34720 ENHANCING JOB OPPORTUNITIES Eastern Europe and the Former Soviet Union This report is part of a series undertaken by the Europe and Central Asia Region of the World Bank. The series covers the following countries: Albania Latvia Armenia Lithuania Azerbaijan Moldova Belarus Poland Bosnia and Herzegovina Romania Bulgaria Russian Federation Croatia Serbia and Montenegro Czech Republic Slovak Republic Estonia Slovenia FYR Macedonia Tajikistan Georgia Turkey Hungary Turkmenistan Kazakhstan Ukraine Kyrgyz Republic Uzbekistan ENHANCING JOB OPPORTUNITIES ENHANCING JOB OPPORTUNITIES Eastern Europe and the Former Soviet Union Jan J. Rutkowski and Stefano Scarpetta with Arup Banerji, Philip O'Keefe, Gaëlle Pierre, and Milan Vodopivec Europe and Central Asia Region ©2005 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 1 2 3 4 08 07 06 05 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not neces- sarily reflect the views of the 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 judgement 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 publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work 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; Internet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. ISBN-10: 0-8213-6195-3 ISBN-13: 978-0-8213-6195-5 e-ISBN: 0-8213-6196-1 DOI: 10.1596/978-0-8213-6195-5 Cover photo by: Y. Hadar. Cover design by: Naylor Design, Inc. Library of Congress Cataloging-in-Publication Data Rutkowski, Jan J., 1954­ Enhancing job opportunities : Eastern Europe and the former Soviet Union / Jan J. Rutkowski, Stefano Scarpetta. p. cm. Includes bibliographical references and index. ISBN-13: 978-0-8213-6195-5 ISBN-10: 0-8213-6195-3 1. Job creation--Europe, Eastern. 2. Job creation--Former Soviet republics. 3. Labor market-- Europe, Eastern. 4. Labor market--Former Soviet republics. I. Scarpetta, Stefano. II. Title. HD5764.7.A6R88 2005 331.12'0420947--dc22 2005044730 Contents Foreword xiii Acknowledgments xvii Acronyms and Abbreviations xix 1. Overview 1 Changing Labor Markets in the Region 5 The Drivers of Labor Demand during the Transition 19 The Role of the Region's Policy and Institutions 24 The Policy Challenge: Promoting Job Creation in the Region 39 2. Main Labor Market Developments during the Transition 61 An Economically Diverse Region with Differing Labor Markets 61 Unemployment and Underemployment: Major Economic and Social Problems 63 Rebounding Real Wages, but Widening Wage Differentials 88 The Changing Nature of Jobs during the Transition 93 Labor Market Outcomes: Disappointing during the Transition? 97 Summary: Key Stylized Facts on Labor Market Transition in the Region 99 vi Contents 3. Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 107 The Employment-Output Link during the Different Phases of the Transition 108 Any Role for Macropolicy to Influence the Employment-Output Link? 114 Summing Up: Employment Prospects in CEE and CIS Countries 118 4. Restructuring, Productivity, and Job Creation 125 The Required Transformation of the Transition Economies and Progress So Far 126 What Is the Role of Firm Restructuring and the Entry and Exit of Firms for Job Creation? 131 What Is the Role of Firm Restructuring and the Entry and Exit of Firms for Productivity and Output Growth? 137 What Drives Restructuring of Existing Firms? 147 How Many Firms Enter and Exit the Market in Transition Countries? 149 Summing Up: Entry Conditions and Incentives to Create Jobs Are Essential for Improving Job Creation in the Region 151 5. The Investment Climate and Job Creation 155 Importance of Investment Climate for Job Creation 157 Employers' Views on the Major Obstacles to Firms' Operation and Growth in the Region 158 The Impact of Investment Climate on Job Creation in the Region 167 Investment Climate: International Comparisons and Variations within the Region 173 Summing Up: Promoting a Better Investment Climate to Foster Job Creation 183 6. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 193 The Role of Labor Market Policies and Institutions 194 The Divergent Paths of Wage Determination during the Transition 194 Employment Protection Legislation Remains Strict despite Reforms, Although Enforcement Is Variable 209 Contents vii Taxes on Labor 217 The Role of Passive and Active Labor Market Programs 221 Summing Up: The Challenge of Labor Policy Reforms in Transition Economies 233 Bibliography 243 Index 259 Boxes 1.1 Geopolitical Country Groups Reflect Economic and Institutional Differences among the Region's Countries 6 1.2 In Most of the Region's Countries, Higher Investment Rates Are Necessary to Accelerate Economic Growth and Job Creation 40 2.1 Do Geopolitical Groupings Help in Assessing the Economic Performance of the Transition Countries? 64 2.2 The Challenge of Job Creation in Turkey 68 2.3 Employment in Moldova 69 2.4 International Migration Patterns in the Region 72 2.5 Growth and Job Creation in Low Income CIS Countries 78 2.6 Relative Position of Women in the Labor Market Has Not Deteriorated during the Transition, and New Employment Opportunities for Women Emerged in the Expanding Services Sector 80 2.7 Internal Migration in the Region in Search of Jobs 86 2.8 An Increase in Educational Wage Premiums Has Been an Important Factor behind the Rise in Wage Inequality 90 2.9 The Surge in Informality during the Transition: Key Features and Policy Challenges 94 3.1 An Empirical Investigation of the Possible Links between Employment, Output, and Macroeconomic Policy 119 4.1 Economic Development and the Employment Structure 128 4.2 A Consistent International Firm-Level Database 132 viii Contents 4.3 Assessing the Impact of Labor Reallocation on Productivity Growth 138 4.4 The Decomposition of Productivity Growth Using Firm-Level Data 143 5.1 Small Entrepreneurs Complain about the Business Environment in Bulgaria 158 5.2 Service Sector Employment Rate as an Indicator of Job Creation Potential 168 5.3 Stringent Employment Protection Regulations May Forestall Job Destruction, but at the Same Time They Discourage Job Creation 172 5.4 What the Official Data on Entry Barriers Do Not Show: Romania 176 6.1 The Role of Labor Market Policies and Institutions: Some International Evidence 195 6.2 Wage Bargaining in Estonia: A Radical Reformer 200 6.3 Innovative Ways of Targeting the Poor 226 6.4 Public Works and Workfare: An Alternative to the Unemployment Benefit? 228 Figures 1.1 The Payoff to Reforms in Transition Economies: Higher Output, but Still Insufficient Jobs, 1992­2003 2 1.2 Unsynchronized Job Creation and Job Destruction Can Give Rise to Unemployment 9 1.3 Unemployment Continues to Be High in Most Transition Economies 10 1.4 Employment Rates Have Declined and Are below the Lisbon Target of 70 Percent 11 1.5 Different Patterns of Labor Reallocation: The Czech Republic (CEE) vs. the Kyrgyz Republic (CIS) 15 1.6 Wage Inequality in the CIS Is Higher than in the CEE 18 1.7 Labor Reallocation Has Played an Increasing Role in Promoting Labor Productivity Growth in Russia 21 1.8 The Rate of Job Creation Is Higher for More Productive Firms in Moldova 22 1.9 Firm Entry and Exit Are Critical for Productivity Growth 25 Contents ix 1.10 Major Obstacles to Firm Activity, 2002 28 1.11 The Tax Wedge on Labor in the Region Is High, Often Higher than in Most OECD Countries 29 1.12 Obstacles to Business Operation and Growth Vary by Subgroup 31 1.13 The Region's Countries Have More Stringent Regulations on Hiring and Firing than OECD Countries Do 37 1.14 Labor Regulations Seem to Be a Binding Constraint Only in the New EU Member Countries and Not in the Other Parts of the Region 38 1.15 Access to Finance Is More Difficult in Transition Economies than in Market Economies at Similar Income Levels 43 1.16 Constraints Reported by Firms Vary across the Region's Countries 46 2.1 Unemployment Is High in Most CEE and SEE Countries, 2003 67 2.2 Employment-to-Population Ratio Is Low in Most of the Region's Countries 76 2.3 More Workers Are Hired in Regions with a Developed Services Sector, Educated Workforce, and Infrastructure (Poland's Regions, 1997) 84 2.4 Real Wages Have Rebounded in the Mid-1990s 89 2.5 Wage Inequality in the CIS Is Higher than in the CEE 91 2.6 Informal Sector Accounts for a Substantial Share of Total Employment, Especially in CIS 94 3.1 Output per Capita Growth Is Largely Driven by Productivity Growth 109 3.2 Inflationary Pressures Have Declined over Time in Most Countries 112 3.3 Employment Adjustment Has Been More Marked in CEE than in CIS Countries 113 3.4 Real-Wage Adjustments Have Been More Marked in CIS than in CEE Countries 114 3.5 Real Interest Rates Have Increased in Recent Years in CEE Countries 115 3.6 Share of Gross Fixed Capital Formation as a Percentage of GDP 116 3.7 Share of GFCF and Productive GFCF as a Percentage of GDP 117 x Contents 3.8 Loosening of the Fiscal Stance in CEE Countries in Recent Years 118 3.9 Real Wages in the Public Sector Have Increased More Rapidly than in the Private Sector in CEE Countries 122 4.1 Different Patterns of Labor Reallocation across Transition Economies 129 4.2 Large Job Flows in Transition Economies 134 4.3 Unsynchronized Job Creation and Destruction Can Give Rise to Unemployment or Underemployment 135 4.4 Job Flow Rates, Selected Transition Countries, 1990­2001 136 4.5 Decomposition of Labor Productivity Growth, CEE Countries 140 4.6 Contribution of Reallocation to Russian Labor-Productivity Growth, 1986­2001 144 4.7 Sources of Productivity Growth in Transition and Emerging Economies 145 4.8 Relationship between Net Entry Contribution and Productivity Growth of Incumbents 146 4.9 Effects of Foreign and Domestic Privatization on Multifactor Productivity Growth (MFP) 148 4.10 Effects of Foreign and Domestic Privatization on Productivity, Employment, and Wages 149 4.11 How Many Firms Enter and Exit the Market? 150 5.1 Most Frequently Reported Major Obstacles to Firm Operation in the Region 159 5.2 Obstacles to Business Operation and Growth Vary by Subgroup 163 5.3 Smaller Firms Are More Constrained by the Investment Climate 166 5.4 Market Service Employment Is Higher in Countries with Easier Access to, and Lower Cost of, Credit 171 5.5 Excessive Market Regulation Hurts Job Creation 171 5.6 In Some of the Region's Subgroups, Time Spent Dealing with Government Regulations Is Still Substantial 174 5.7 Protection of Property against Crime Can Be Costly 174 Contents xi 5.8 Firms in the Region Rely to a Lesser Degree on Capital Coming from Formal Institutions than Do Firms in Other Regions 175 5.9 Starting a Business Is Not Easy in Many of the Region's Countries 178 5.10 Job Creation in the Region Is Likely to Be Hampered by Difficult Access to Credit 180 5.11 Markets in the Region Tend to Be Overregulated 181 5.12 Corruption Is High in the Region 182 6.1 Density and Bargaining Coverage, Early 2000s 198 6.2 Minimum-Wage-to-Average-Wage Ratio, 2002 206 6.3 Minimum Wage in Ukraine Accounts for a High Percentage of the Market Wage of Low-Skilled Workers, but It Is Not Enforced 207 6.4 Employment Protection Legislation in EU-8 and Other Selected Countries during the Transition 212 6.5 Transition Countries Have More-Stringent Regulations on Hiring and Firing than Do OECD Countries 213 6.6 There Are Significant Differences within the Region's Countries on EPL 214 6.7 Informality Tends to Be Higher in Countries with Strict EPL 216 6.8 The Region's Tax Wedge on Labor, 2003 218 6.9 Tax Wedge on Labor, the Region, and OECD, Early 2000s 219 6.10 Surge in the Tax Wedge in EU Transition Countries during the 1990s 220 6.11 Unemployment Benefit Replacement Rates Have Declined in EU Transition Countries during the 1990s 223 Tables 1.1 A Typology of Employment Protection Legislation and Enforcement 37 1.2 A Summary of Key Policy Measures to Improve Labor Market Outcomes in Transition Economies of the Region 45 2.1 Differences between Middle-Income European and Low-Income CIS Labor Markets 66 xii Contents 2.2 Flows into and out of Unemployment 74 2.3 Large Regional Disparities in Unemployment Rates, 2003 82 3.1 Significant Changes in the Correlation between Macroeconomic Variables and Labor Market Indicators 111 4.1 The Employment Structure in CEE and CIS Countries 127 4.2 The Evolution of Productivity, Wages, and Unit Labor Costs 142 5.1 Contribution of Investment Climate Components to Change in Service Sector Employment in the CEE EU Countries over the Past Decade 169 5.2 The Importance of Determinants of Job Creation Varies by Subgroup in the Region 172 6.1 EPL and Enforcement Typology 215 Foreword In the last 15 years, the countries of Eastern Europe and the Former Soviet Union have made impressive progress in their historical transi- tion from centrally planned to market economies. Building the insti- tutional foundations of a market economy, they have developed a vibrant private sector and opened themselves to international trade. After an initial economic recession, this has ignited economic growth, which in turn has brought about higher incomes and reduced poverty. Currently, many of the countries in the Region are well-functioning market economies; eight of them have joined the European Union and an additional three transition countries, together with Turkey, are on the road to EU accession. Such a successful transition, achieved over a short period of time, exceeded many observers' expectations. But the transition and recent growth have been disappointing in one key area--jobs. In countries across the region, job opportunities remain scarce. In many cases, job creation in the expanding new pri- vate sector still falls short of job destruction in the declining old sec- tor. As a result, employment has decreased substantially in virtually all countries in the Region, and high unemployment has emerged in some of them. While some transitional unemployment was expected, the surprise lay in its persistence. Many workers displaced by struc- tural shifts failed to find new jobs, and a significant number have either been unemployed for over a year or are in low-productivity xiii xiv Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union occupations. In some of the new member states of the European Union, as well as in some acceding countries, the unemployment rate tends to be in double digits--as it is in some of the cohesion countries that joined the EU in previous rounds of enlargement. This carries significant social and political costs--long-term unemployment threatens to erode skills, aggravates poverty and inequality, and even- tually leads to social marginalization. In the CIS countries, the jobs problem lies more in the quality of jobs--which are less productive and pay correspondingly little. Going forward, poor job prospects, if not improved, will act as a brake on the substantial poverty reduction that has occurred in the Region since 1998 and will undermine polit- ical support for reforms. Obviously, these problems are more pro- nounced in some countries--where reforms have been slow or delayed--than in others. But the lack of productive job opportunities is a common thread in virtually all transition economies. This study, Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, focuses attention on the causes of these disappointing labor outcomes and points to the solutions. Why has the job creation record of transition economies been disappointing? One reason is that the size of the new private sector that generates jobs is still relatively small in many of the Region's countries. And there is still enough room for firms to increase output through downsizing and retrench- ment--what is sometimes termed "defensive restructuring." For employment to grow, more firms need to engage in "strategic restruc- turing," which requires investing productively, hiring more workers, and finding new production niches. And why do firms not engage in more strategic restructuring? Often, they are discouraged by the poor investment climate--substantial risks, barriers, and costs associated with doing business. Two other factors also play a role. In countries in the Region with strong institutions and enforcement capacity, strict employment pro- tection legislation can actually inhibit hiring. Again, for many firms in the Region, the lack of adequate skills among available workers also proves to be a significant constraint to growth. The pace of job creation and the reduction in unemployment in the Region can and should be accelerated. This study recommends a two-pronged strategy to create more and better jobs. First, in every country across the Region, the investment climate needs to be further improved to encourage new firms to enter the market, and for exist- ing firms to grow. The specific needs differ across countries. In the low-income countries of the Former Soviet Union, the needs are basic: continued development of the institutions of a market econ- omy and reducing risks. In the middle-income countries, such as the Foreword xv Russian Federation and Ukraine, and in Southeastern Europe, the challenge is removing administrative barriers. In the new and prospective European Union members of Central and Eastern Europe, the necessity is improving the quality of regulations. Across the region, better governance, investments in human capital and in infra- structure, and ultimately building competitive economies integrated in a world economy are other key challenges. Second, countries in the Region need to have institutional and reg- ulatory reform to develop an adaptable labor market, where core worker rights are effectively protected but employers are not unduly constrained in adjusting the size and skill composition of their work- force to the changing product demand. Workers need to have incen- tives and skills necessary to take up available jobs, and be capable of moving to where the jobs are. This book, part of a new series of regional studies, is intended as a contribution to the World Bank's goal to work more effectively with clients and partners in the Region to reduce poverty and foster eco- nomic growth through better utilization of labor resources. It com- plements the companion study on growth, poverty, and inequality, which examines in greater detail the critical link between labor mar- ket outcomes and poverty. Forthcoming studies on trade, infrastruc- ture, migration, and demographics look at the key economic and social opportunities and challenges. I hope that these studies stimu- late debate, promote better understanding, and spur action to bring about prosperity for all. Shigeo Katsu Vice President Europe and Central Asia Region Acknowledgments This study was prepared by a team led by Jan Rutkowski and Stefano Scarpetta, who were also the main authors, and comprising Arup Banerji, Philip O'Keefe, Gaëlle Pierre, and Milan Vodopivec. Additional inputs were provided by John Haltiwanger and Paolo Saavedra. It draws on commissioned background papers by a number of researchers, including Tito Boeri, David Brown, Pietro Garibaldi, John Earle, Peter Huber, Paloma Lopez-Garcia, and Paolo Verme. This work was supported by the Chief Economist of the Europe and Central Asia (ECA) Region, Pradeep Mitra, who has provided essential guidance throughout its implementation. The study bene- fited from useful comments and suggestions provided at various stages by Michal Boni, Daniela Gressani, Vladimir Gimpelson, Ros- tislav Kapelyushnikov, János Köllő, William Maloney, Mamta Murthi, Michelle Riboud, Michal Rutkowski, and Jan Svejnar. Earlier versions of this study were presented at a World Bank Con- ference on Labor Market Stock-Taking: "What is the Role of Labor Market Policy in Promoting More and Better Jobs?" held in Washing- ton, DC, on November 18­19, 2004, and at the Centre for Analysis of Social Exclusion (CASE) Conference titled "Europe after the Enlarge- ment," held in Warsaw on April 8­9, 2005. The study benefited from the discussion following its presentation, and in particular from com- ments provided by Mikhail Dmitriev, János Köllő, and Jan Svejnar. xvii xviii Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union The study also benefited substantially from comments received from participants in a series of consultation meetings held in April 2005 at the International Labour Organization (ILO) in Geneva, the European Commission in Brussels, the Organisation for Economic Co-operation and Development (OECD) in Paris, and the U.K. Department for International Development (DFID) and the European Bank for Reconstruction and Development (EBRD) in London. The World Bank Office of the Publisher coordinated the book design, editing, and production. Dina Towbin was the production edi- tor. Francis Speltz and Timothy Whitehead edited the manuscript. Acronyms and Abbreviations AFR Africa ALMP active labor market program AW average wage BEEPS Business Environment and Enterprise Performance Survey BiH Bosnia and Herzegovina CEE Central and Eastern Europe CIS Commonwealth of Independent States CIS-7 low income CIS countries: Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan EAP East Asia and Pacific EBRD European Bank for Reconstruction and Development EC European Commission ECA Europe and Central Asia EIRO European Industrial Relations Observatory EMU Economic and Monetary Union EPL employment protection legislation EU European Union EU-15 European Union before the 2004 enlargement EU-8 New EU-member countries (the Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, the Slovak Republic, and Slovenia) xix xx Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union EUA-3 EU-accession countries (Bulgaria, Croatia, and Romania) FBiH Federation of Bosnia and Herzegovina FDI foreign direct investment FYR former Yugoslav Republic GDP gross domestic product GFCF gross fixed capital formation GNI gross national income ICS investment climate survey ILO International Labour Organization (plenary body) or International Labour Office (the secretariat and publisher) IMF International Monetary Fund IQR interquartile range IZA Institute for the Study of Labor LAC Latin America and the Caribbean LFS labor force survey MENA Middle East and North Africa MFP multifactor productivity growth MNA Middle East and North Africa MW minimum wage NUTS nomenclature of territorial units for statistics OECD Organisation for Economic Co-operation and Development PPP purchasing power parity SAM Serbia and Montenegro SAR South Asia SEE Southeastern Europe SOE state-owned enterprise SSA Sub-Saharan Africa U.S. United States UB unemployment benefit UI unemployment insurance U.K. United Kingdom UN United Nations UNDP United Nations Development Programme Note: All dollar amounts are U.S. dollars ($) unless otherwise indicated. CHAPTER 1 Overview This study examines labor market developments and strategies for enhancing job opportunities in Eastern Europe and the Former Soviet Union during their historic transition from centrally planned to mar- ket economies. (This group of countries is called "the Region" in this study.)1 It addresses three key questions: How has the process of eco- nomic transformation associated with the transition affected the labor market, and who are those who have been hardest hit? How far have the different countries progressed in the transformation process, and what are the key constraints against creating more and better jobs? How can public policy help promote job creation? The transition from centrally planned to market economies has entailed profound institutional reforms and far-reaching structural changes. The countries of Eastern Europe and the Former Soviet Union have achieved impressive results in economic liberalization, privatizing state-owned enterprises (SOEs), and opening up their economies to international trade and capital flows. After a major plunge in the early phases of the transition, gross domestic product (GDP) per capita has recovered in many countries, being at or above the pretransition levels in many. Overall, there is a clear association between the reform effort (proxied by the European Bank for Recon- struction and Development [EBRD] transition index) and the recov- ery of GDP since the early 1990s (figure 1.1). 1 2 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Despite some progress in recent years, labor market conditions remain difficult in most countries. For example, the employment-to- working-age-population ratio--a summary measure of the utilization of labor resources--is currently lower than at the beginning of the transition and often below the average of industrial countries (figure 1.1). Slack labor markets assume different dimensions in the Region's countries: high open unemployment, falling labor force participation, or a low-productivity employment trap. In faster-restructuring countries such as Poland and the Slovak Republic, unemployment was 19 and 18.2 percent of the labor force in 2004, respectively, despite high rates of economic growth. In Hun- gary, the unemployment rate is relatively low (around 6 percent), but this rate is partly the result of falling labor participation: many work- ers have given up their job search and left the labor force. FIGURE 1.1 The Payoff to Reforms in Transition Economies: Higher Output, but Still Insufficient Jobs, 1992­2003 180 4.5 100) = 160 4 (1992 140 3.5 2003 120 3 rate, 2003 100 2.5 index employment 80 2 EBRD capita, 60 1.5 per GDP 40 1 of 20 0.5 Index 0 0 Rep.* Fed. Rep. Rep. Ukraine Belarus Romania Bulgaria Hungary Poland Slovenia Estonia Kyrgyz Azerbaijan Russian Czech Lithuania Slovak GDP per capita (left axis) Employment rate (left axis) EBRD index 2003 (right axis) Sources: World Bank 2004c; ILO LABORSTA database; and EBRD Transition Report 2004. Note: Data for the Kyrgyz Republic are for 2002. Data for Bulgaria, the Czech Republic, Poland, Slovenia, and the Slovak Republic refer to 1993; for Romania, data re- fer to 1994. The EBRD transition index is the simple average of the published indicators on Price Liberalization, Forex and Trade Liberalization, Small-Scale Privatiza- tion, Large-Scale Privatization, Enterprise Reform, Competition Policy, Banking Sector Reform, and Non-Banking Financial Institutions. Overview 3 Open unemployment is less of a problem in the slower-reforming countries of the Former Soviet Union, such as Azerbaijan, Moldova, and Ukraine. However, much of their workforce is stuck in low-pro- ductivity employment in unrestructured and probably nonviable enterprises or has had to move back to subsistence agriculture. The employment levels of the central-plan period, when employment was not only a right but also a duty for most of those of working age, cannot be sustained in a market-based system. However, actual labor market conditions in most countries in the Region clearly indi- cate significant slack. This has led to persistent disparities in the labor market. In the Region's economies, especially in Central European (CEE) and Southeastern European (SEE) countries, unemployment is typically of long duration, and many unemployed have difficulties in finding new jobs. The surge in unemployment was largely expected during the transition, given the need for major reallocation of labor across sectors, firms, and locations. But the persistence of unemployment-- the share of those without a job for more than a year--is 40 to 50 percent in the EU-8 transition countries (annex 1.1), compared to below 40 percent in advanced market economies. A common assumption at the beginning of the transition was that high unem- ployment would be transient, declining once the emerging private sector started creating jobs. In reality, workers displaced in the "old" sectors have had limited chances to find jobs in the "new" sectors. In the Commonwealth of Independent States (CIS),2 underemploy- ment (low-productivity employment and hidden unemployment) is more prevalent. Promoting the creation of more and better jobs remains a priority in the Region. Politically, creating viable and lasting employment will strengthen support for further market-oriented reforms. More broad- based employment growth--job opportunities for many workers and not only for the highly skilled and those in large cities--will also reduce social costs. For example, in Poland, a household whose head is unem- ployed is five times more likely to be poor than a household whose head is employed. In most other countries, such as Georgia, the Kyrgyz Republic, or the Russian Federation, unemployment raises the risk of poverty by at least a factor of two (World Bank 2005b). The new market economy environment has contributed to the widening of wage disparities.3 Although the increase in wage dispar- ities is consistent with growing productivity differentials, market dis- tortions have also played a role (particularly pronounced in CIS countries). The worst affected have typically been those who are the most vulnerable to shocks and least able to adjust to the new market 4 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union paradigms: mostly less skilled and older workers. Thus, low employ- ment and significant earnings inequality have translated into higher poverty and lower social welfare. These outcomes have improved in recent years, but serious problems persist for some of the weakest strata of the population. What, then, is the way to more and better jobs in the Region? The specific mix of policies that each country needs to adopt varies, depending upon its particular economic and labor market situation. However, for all transition economies, addressing the issue of under- utilization of labor requires a two-track strategy: increasing the pace of job creation by encouraging firm entry and expansion and facili- tating the reallocation of jobs and workers away from old, less pro- ductive firms toward new, more productive firms. The first part of this strategy requires further improvements in the investment climate and lowering the costs of doing business. Most countries in the Region have made substantial progress in their investment climate. However, there are remaining constraints. Oper- ating a business in low income CIS countries carries high risks (for example, policy unpredictability, insecure property rights, weak con- tract enforcement, and unreliable infrastructure). In the middle income CIS countries (Belarus, Kazakhstan, Russia, and Ukraine), businesses face considerable administrative barriers (for example, numerous permits, inefficient regulations, and red tape). Doing busi- ness in CEE countries is hampered by high direct costs (for example, high taxation). Although there is substantial subgroup commonality, key constraints to firm entry and growth--and thus priorities for investment climate reform--are country-specific. They have been identified in this report in two ways: by comparing the regulatory and institutional settings in the Region's countries with those of other economies with a vibrant private sector and by exploiting enterprise surveys that directly ask entrepreneurs about the constraints they face in their operation and in their potential for expansion. The second part of the strategy requires promoting adaptable labor markets and effective social safety nets to protect incomes of workers displaced in the wake of enterprise restructuring. In some countries (mainly in CIS and SEE), employment protection legislation (EPL) is stringent, but enforcement capacity is weak. These countries need to simultaneously liberalize EPL and promote compliance so that core worker rights are protected. In other countries (mainly CEE EU mem- bers), enforcement capacity is strong, and EPL is relatively flexible. They need to focus on addressing specific constraints to labor market flexibility. For example, where few regulations on temporary employ- ment exist but regular employment is highly protected, these coun- Overview 5 tries need to reduce protections granted to permanent workers to avoid creating labor market duality. The rest of this overview develops these arguments in four parts: · A summary of the key labor market developments during the tran- sition, highlighting clear country differences in labor market con- ditions in the Region (see box 1.1 for a discussion on country grouping) · A review of the process of job creation and destruction that exam- ines enterprise restructuring, privatization, and entry of new ven- tures · Highlights of the key constraints in the business environment that curb investment and job creation · The set of policies still needed to create more and better jobs in the Region Changing Labor Markets in the Region The relatively poor record of job creation across the Region, despite the resumption of economic growth, remains the key area of concern for most policy makers. In the early phases of the economic transfor- mation of the Region's countries, employment declined because of the major contraction in output. However, in the subsequent recovery periods, employment has not resumed as much as output has. This is the case especially in CEE countries, where GDP growth has been sustained but employment has been fairly stagnant. Therefore, as in some EU-15 countries (annex 1.1), certain transition economies have recently experienced "jobless growth," with the number of jobs in the formal sector hardly responding to the growth in output. Poland is a good example: since the late 1990s, employment has been falling and unemployment persisting, despite relatively high rates of economic growth. In Hungary, the relatively weak labor demand has led to a fall in labor force participation. In other countries, such as those in the CIS, the link between output and employment has been stronger. In the early years of transition, this was associated with negative or very small output growth. Even when economic growth resumed in recent years, the strong link with employment has been associated with a generally poor quality of employment, with a rapid increase in the informal sec- tor (including agriculture) as an employer of last resort. Four main features characterize the labor markets of the Region during the transition: 6 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 1.1 Geopolitical Country Groups Reflect Economic and Institutional Differences among the Region's Countries The Region is large and economically diversified. Its subgroups of countries differ in their levels of economic development, institutions, industrial structure, and progress in market-oriented reforms. These differences, in turn, influence the characteristics and evolution of their labor markets. One way to group countries in the Region is to apply a traditional geopolitical grouping (annex 1.1). This approach clearly shows these countries' differences in their levels of economic devel- opment, institutional settings, and economic structure: · The following table shows a close correspondence between geopolitical location and differ- ences in the level of economic development, as measured by the GDP per capita. All CEE EU-member countries belong to the group of the richest countries in the Region. All low in- come CIS countries belong to the group of the Region's poorest countries. Thus, geopolitical location is a good predictor of the level of economic development. Geopolitical Location and Level of Economic Development Are Closely Related Income groups: GDP per capita,a 2002 Geopolitical groups Top tercile Middle Bottom tercile Mean value US$ CEE EU members 8 12,984 CEE EU accession 1 2 8,032 SEE 3 1 5,685 Middle income CIS 3 1 6,122 Low income CIS 7 2,428 Sources: World Development Indicators; World Bank database; Bank staff calculations. Note: Numbers stand for the number of countries within the cell. a. GDP at PPP (current international dollars). 1. In many countries, job creation still lags behind job destruction, leading to employment decline. Overall, job reallocation (the sum of job creation and destruction) in the Region increased dramati- cally in response to transition--from less than 10 percent of the workforce in the late 1980s to more than 25 percent in the 1990s. The main differences across countries, however, are in job cre- ation. Although countries lagging behind on reforms, such as those in the CIS, were able to contain job destruction in the early phases of transition, job destruction later rose to high levels but was not accompanied by new private initiatives to create new jobs. Overview 7 · Similarly, a strong overlap exists between geopolitical location and quality of institutions and governance. Government effectiveness varies significantly between geopolitical groups, but variation within groups is relatively small. · Geopolitical location also reliably corresponds to the progress of economic transition (as measured by the International Bank for Reconstruction and Development index). Countries that have either become members of the European Union or are candidates for membership have advanced the most in implementing market reforms. Economic transition is less ad- vanced in the middle income and--especially--the low income CIS countries, while SEE countries occupy the middle position. Naturally, use of the geopolitical taxonomy occasionally results in a misclassification of some countries from the economic perspective. Bulgaria and Romania, EU accession countries that thus fall into the CEE group, are closer to the SEE group in economic development. The border- line between middle and low income CIS countries is in some cases blurred. For example, in government effectiveness and regulatory quality, Armenia performs better than other low in- come CIS countries. Georgia, the Kyrgyz Republic, and Moldova, in turn, are more advanced in market-oriented reforms than other CIS countries. Source: Bank staff analysis. 2. The nature of jobs has changed. The incidence of regular formal sector jobs has fallen, while the incidence of temporary and casual informal sector jobs has increased. This has led to increases in the segmentation of the labor market. There has also been a shift from less skilled, blue-collar manufacturing jobs toward more skilled, white-collar service sector jobs. 3. Wage growth has accelerated in recent years, but wage differen- tials have also widened substantially. Enterprise restructuring has led to significant improvements in productivity growth, which in turn has led to marked real-wage increases. At the same time, 8 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union returns to education and to market-related skills have risen, lead- ing to a greater dispersion in the distribution of wages and a grow- ing share of workers in low-paid jobs. 4. There is strong regional concentration of job creation within coun- tries, which has brought about large regional disparities in labor market conditions. These features of labor market adjustment to the transition shock have been common, to a varying degree, across the Region's sub- groups of countries. At the same time, however, there have been con- siderable cross-regional differences in the nature of labor market adjustment, depending on region-specific features of the labor mar- ket. These four labor market factors are now discussed in turn. Persistent Unemployment and Lack of Productive Jobs in the Region The observed labor market outcomes in the Region's countries reflect different stages of the transition process and differences in underlying job dynamics (figure 1.2). All countries have experienced a surge in job destruction by downsizing and closing businesses. They have also seen job creation by new and expanding firms. However, the size of job creation and destruction, as well as their timing, has varied markedly across countries. For example, Estonia and Slovenia experienced a short period of large job destruction in the early phase of the transition. This was fol- lowed by a period in which job creation and destruction went hand in hand (synchronized flows). More recently, incipient signs in these countries are emerging of job creation exceeding job destruction. In contrast, many other countries are still struggling with job cre- ation falling short of job destruction (for example, see Romania in fig- ure 1.2). In Russia, job flows have been remarkably low, with job creation lagging behind job destruction for most of the past decade. In those countries where job destruction has been contained by slowing down the restructuring process, larger flows can be expected once the required transformation process has unfolded. During the initial stage of the transition, high job destruction and low job creation resulted in growing unemployment and falling employ- ment. Today, unemployment rates are in double digits in several CEE economies, in some cases above the Organisation for Economic Co- operation and Development (OECD) or EU-15 averages (figure 1.3). Unemployment rates reach highs of more than 30 percent in the former Yugoslav Republic of (FYR) Macedonia and 19 percent in Poland. In Overview 9 FIGURE 1.2 Unsynchronized Job Creation and Job Destruction Can Give Rise to Unemployment Synchronized job flows Less synchronized job flows Estonia Russian Federation 14 14 12 12 10 10 8 8 employment employment 6 6 total 4 total 4 of of % 2 % 2 0 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Slovenia Romania 14 20 18 12 16 10 14 8 12 10 employment 6 employment 8 total 4 total 6 of 4 2 of % % 2 0 0 1993 1994 1995 1996 1997 1998 1999 1994 1995 1996 1997 1998 1999 Job creation Job destruction Net employment losses Net employment gains Source: Scarpetta and Vodopivec 2005. most CIS countries, open unemployment tends to be lower (with Geor- gia, at more than 12 percent, being an outlier), but many workers are engaged in low-productivity activities or have moved back to subsis- tence agriculture (see the following discussion). In CEE countries, high and persistent unemployment has been accompanied by a major fall in labor force participation rates, as work- ers became discouraged by lack of job opportunities and gave up their job search. In these countries, both open unemployment and low labor force participation have led to a low ratio of employment to working- age population (figure 1.4).4 Employment-to-population ratios are below the EU's average and, for most countries, clearly below the Lis- bon target of 70 percent of the working-age population.5 However, in recent years there are signs that this negative trend in employment has been reversed in some countries, most notably the Baltic States--rais- ing hopes that employment prospects are finally improving. In CEE and SEE countries, the salient feature of unemployment is its long duration and prevalence among the young and the less edu- 10 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.3 Unemployment Continues to Be High in Most Transition Economies Unemployment Rate in 2003, Percentage Kyrgyz Rep.* income CIS Georgia Low Moldova Ukraine income CIS Kazakhstan Russian Fed. Middle Macedonia, FYR SEE Albania Serbia & Montenegro Poland Slovak Rep. Croatia Bulgaria Lithuania CEE Latvia Estonia Czech Rep. Romania Slovenia Hungary EU-15 Turkey OECD average 0 5 10 15 20 25 30 35 40 Unemployment rate in 2003, % Sources: Labor Force Surveys (ILO [LABORSTA database], OECD); * 2002; and Labour Market in the CIS countries, Statisti- cal Abstract, Interstate Statistical Committee of the CIS, Moscow 2004. cated. These are problems observed in many EU countries, but assume particular severity in some of the Region's countries. More than 50 percent of the unemployed in many CEE countries have been without a job for more than one year (long-term unemployed), and Overview 11 FIGURE 1.4 Employment Rates Have Declined and Are below the Lisbon Target of 70 Percent Employment-to-Working-Age-Population Ratio in 2003 Georgia income Kyrgyz Rep.* CIS Low Moldova* Kazakhstan income CIS Russian Fed. Middle Ukraine Serbia & Montenegro * SEE Macedonia, FYR Czech Rep. Estonia Slovenia Latvia Lithuania CEE Romania Slovak Rep. Hungary Bulgaria Poland EU-15 Lisbon strategy average target Turkey 0 10 20 30 40 50 60 70 80 Ratio of employment to working-age population, 2003 Sources: ILO (LABORSTA database); OECD; Eurostat; and Bank staff calculations. Note: * = 2002 the proportion is still higher in SEE countries. The chances to escape unemployment are low by the standards of flexible labor markets and are similar to those seen in some EU countries with a stagnant pool of jobless people. Turkey, a nontransition economy, provides a telling 12 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union contrast of a more dynamic labor market. Long-term unemployment there accounts for some 30 percent of total unemployment, which is similar to the proportion observed among more flexible EU labor markets. Unemployment rates in transition economies are particularly high among less educated workers. For example, in Poland, a worker with primary education (eight years) is more than four times more likely to become unemployed than a worker with tertiary education. The employment rate for workers with primary education is less than 20 percent, compared with close to 80 percent for workers with tertiary education (United Nations Development Programme [UNDP] 2004). Youth is another group that suffers from high unemployment: on average, youth unemployment rates are twice as high as overall unemployment rates. In Bulgaria and the Slovak Republic, more than one-third of the youth labor force is unemployed, which also makes them poorer. In many CIS countries, low open unemployment and high employment rates hide significant problems: delayed enterprise restructuring with persistent overstaffing and, especially in low income CIS countries, the dominance of low-productivity jobs in the informal sector to earn subsistence income. The latter feature is typi- cal of developing countries, where social protection is ineffective and thus unemployment is not an affordable option for most workers. The natural implication is that unemployment in the CIS may worsen as restructuring progresses further. Enterprises will downsize more aggressively to be competitive, as they did in CEE, and unprofitable firms will close. Thus, inflows into unemployment are likely to increase as restructuring progresses. The extent to which this will cause an increase in unemployment will depend on the ease with which the displaced workers will find jobs in the new private sector (or in the informal sector). The Changing Nature of Jobs The greatest change since transition has been the shift from secure, though not highly rewarding, employment, to less stable jobs with greater earning potential. But in the low income CIS, casual and less formal jobs have also increased dramatically to levels observed in developing countries. For example, self-employment accounts for about 20 percent of total employment in EU transition economies and for about 50 percent of employment in low income CIS coun- tries. Similarly, informal sector employment as a share of total employment is estimated at around 40­50 percent in the CIS. Overview 13 Another feature of the changing nature of jobs in transition countries is the rapid increase in fixed-term employment contracts. The characteristics of the jobs offered under temporary or informal arrangements and those of self-employment vary across individuals and countries. Temporary employment has been largely used by firms to circumvent the often rigid firing regulations for those on open- ended contracts. For example, in Croatia, most workers are hired using fixed-term contracts (World Bank 2003b). Informal jobs loom large among young, poorly educated blue-collar workers. They also seem to be a more frequent destination after a period of unemploy- ment than formal sector jobs are. Informal workers also tend (on average) to have lower wages than formal sector workers do. The observed wage differentials between formal and informal sector jobs are largely the result of selection of workers in the two sectors: low- skilled and less experienced workers tend to be sorted into informal jobs, often in small firms with low-productivity potentials. The nature of self-employment also varies greatly. For some low- skilled workers, especially in the poorer CIS countries, own-account jobs in retail and agriculture are subsistence activities. But for other, more skilled workers, self-employment is sometimes a preferred alternative to formal sector employment because self-employment offers better earning opportunities and more scope for entrepreneur- ship, especially in the more dynamic CEE countries. The nature of jobs has also changed because of sectoral shifts and deindustrialization. Most CEE countries have witnessed a fall in the number of blue-collar manufacturing jobs and an increase in white- collar service sector jobs. In contrast, in most CIS countries, deindustri- alization was associated with an increase in agricultural employment. Figure 1.5 illustrates these divergent patterns of job reallocation. In the Czech Republic, market services as a share of total employment increased by about 5 percentage points during the transition, while manufacturing's share fell by about 3 percentage points. In contrast, in the Kyrgyz Republic, market services gained little as a share of total employment. However, there was a dramatic fall in the share of manu- facturing (about 17 percentage points) and a huge shift toward agricul- ture, which increased its share by some 20 percentage points. It is not clear whether the significant increase in the share of agricultural employment in the CIS countries is only temporary or represents a more profound and long-lasting reversion toward employment pat- terns more typical of countries with relatively low income per capita. The skill content of jobs has shifted upward, especially in the new sector. The profiles of job creation and destruction during the transi- tion differed in skills, occupations, and experience. Many jobs have 14 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union been destroyed in low-skilled activities or in highly specialized man- ufacturing activities. Many jobs have been created not only in rela- tively more skilled activities but also in certain service activities that require low- and medium-level skills that are nonetheless different from those of the lost manufacturing jobs. In Estonia, for example, the share of nonmanual workers in total employment increased by 8 percentage points during 1990­2000; at present, nonmanual workers account for 55 percent of total employment. In Russia, the share of workers with tertiary education increased by 6 percentage points dur- ing 1992­2000 and currently exceeds 20 percent. A similar increase occurred in Poland and other CEE economies (Peter 2003). This skills mismatch between jobs destroyed in the old sector and those created in the new sector is one of the factors behind unemployment in tran- sition economies. It requires reforms in the education system and efforts of requalification (Commander and Köllő 2004).6 Changes in the nature of jobs have affected men more than women. Many jobs have been lost in sectors dominated by male employment--heavy industry and the extraction industry--while new activities have been created in services where women tend to have easier access. As a result, men have suffered relatively more job losses than women have during the transition. For example, the female-employment-to-population ratio in CEE, at 54 percent, is close to the EU-15 average (56 percent), while the male ratio, at 65 percent, is significantly below the EU-15 average (73 percent).7,8 Wage Growth, Inequality, and Low-Paid Jobs Wage increases in recent years have been associated with firm restructuring and productivity improvements. Enterprise restructur- ing benefited the insiders (that is, those workers who kept their jobs and those who managed to find new ones).9 Wage growth has been the fastest in the low income CIS countries, where it averaged 2.8 percent annually. It was slightly slower in the CEE and SEE and in middle income CIS economies, where it averaged 2.5­2.6 percent annually during 1996­2002. The change in the wage-setting mechanism, from administrative to market-based, has led to an increase in wage dispersion. The wage distribution was compressed under central planning because of the egalitarian ideology and the centralized wage-setting mechanism. For example, the wage premium for university workers over those with vocational education was just 20 percent.10 Changes in the structure of the economy, and thus changes in relative demand for different types of labor, have given rise to various wage premiums to worker Overview 15 FIGURE 1.5 Different Patterns of Labor Reallocation: The Czech Republic (CEE) vs. the Kyrgyz Republic (CIS) The structure of employment in the Czech Republic Evolution over the transition and with respect to benchmark Agriculture Industry 20 45 . 1989 Benchmark 1990 1992 . . 1996 40 . 15 .. 2000 . 2002 1990 1989 35 employment . employment 10 1992 30 total Benchmark . total . of . 1996 2000 2002 of 5 . .. % . % 25 9000 9500 10000 10500 11000 11500 9000 9500 10000 10500 11000 11500 GDP per capita/PPP terms GDP per capita/PPP terms Market Services Nonmarket Services 26 32 Benchmark . .. 2002 2000 . Benchmark 1996 24 . 30 . 2002 . 2000 . 22 employment 28 . . employment 1992 . . 1996 . . total 26 1992 total 20 . 1990 of of 1989 % 24 . 1989 18 1990 % 9000 9500 10000 10500 11000 11500 9000 9500 10000 10500 11000 11500 GDP per capita/PPP terms GDP per capita/PPP terms The structure of employment in the Kyrgyz Republic Evolution over the transition and with respect to benchmark Agriculture Industry 55 30 . 2000 1990 50 25 . 1989 . .. 1996 45 Benchmark 1992 . 20 Benchmark employment . employment . 40 . . total 1992 total 15 1996 35 . of 1990 of 1989 .. % % . 30 10 2000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 GDP per capita/PPP terms GDP per capita/PPP terms Market Services Nonmarket Services 25 26 . 1992 1990 24 . 1989 Benchmark . 20 22 . 1996 employment . . employment 1996 20 . .. 2000 15 . 2000 total . . total 18 of . 1992 . 1990 of Benchmark % % 16 10 1989 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 GDP per capita/PPP terms GDP per capita/PPP terms Source: Bank staff calculations based on the "Industrial Database for Eastern Europe" of the Vienna Institute for International Economic Studies. Note: The market benchmarks are obtained through regressions of sectoral employment shares on GDP per capita and its square on a sample of 50 countries. 16 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union and firm characteristics (such as education and skills, occupation, or industry) that were not widespread in socialist times. A university- educated worker in Poland currently earns (on average) 70 percent more than a worker with basic vocational training. In the private sec- tor, this premium to university education is even higher (160 per- cent).11 These natural changes not only have contributed to the increase in wage inequality over most of the transition period but also have led to better market efficiency as wages become better aligned with productivity differentials. One should note that, although they remain high by the standards of industrial countries, wage inequali- ties have shown some decline in recent years (World Bank 2005b). The gender pay differential has declined in many countries in the Region and is relatively small. On average, women earn some 20 per- cent less than men in comparable jobs, which is modest by interna- tional standards (Paci 2002). Thus, the gender wage differential has not been a factor behind the rise in wage inequality. After all, women in the Region benefited from the increase in returns to schooling more than men because (on average) they are better educated than men. Though wage inequality is presently high in virtually all transition economies, there are substantial differences among the Region's sub- groups of countries (figure 1.6). In European transition economies, wage inequality is high by EU standards, but still within the OECD range (although at its upper end). In contrast, in most of the CIS coun- tries, wage disparities have reached very high levels characteristic of many developing countries. As an illustration, in CEE countries, the top-decile worker usually earns 4 to 5 times as much as the bottom- decile worker. In the Kyrgyz Republic, this ratio rises to 9, in Russia to 11, and in Azerbaijan to 13. In contrast, in the EU, the decile ratio varies between 3 and 4. In countries where there is high earnings inequality, the labor mar- ket is divided into an increasingly large high-wage segment and a cor- respondingly large low-wage segment. For example, in the Czech Republic and Slovenia (where earnings inequality is relatively low), less than 5 percent of all wage and salary workers earn less than 50 percent of median earnings. In Serbia and Montenegro, this fraction exceeds 20 percent; in Russia, it goes up to 25 percent; and it reaches 30 percent in Azerbaijan.12 The high incidence of low pay is of partic- ular importance because it may translate into poverty. However, in the Region, consumption inequality tends to be lower than wage inequality because of multiple-earner households and extensive transfer systems in many countries. In the low income CIS countries and some SEE countries, remittances and other private transfers may also boost consumption levels (World Bank 2005b). Overview 17 The minimum wage serves as a floor in European transition economies, reducing wage dispersion. In CEE and SEE, minimum wages are set at around 40 percent of the average wage. In contrast, in the CIS, minimum wages are much lower in relative terms. As a rule, they account for less than 20 percent of the average wage-- often around 10 percent--which means that they are unlikely to be binding. This allows firms to maintain low-paid jobs, and as such it contributes to wage dispersion. Geographical Labor Market Disparities Labor market conditions vary substantially across geographical areas of most of the Region's countries, and in many cases such geographi- cal disparities have persisted or even increased over time. For exam- ple, in 2003, differences between the regions with the highest unemployment rate and those with the lowest unemployment rate (usually the capital) exceeded a factor of 3 in all countries of the Region except Romania and Slovenia. These disparities in unemploy- ment are also associated with large disparities in wages, with wages being lower in higher-unemployment regions. In Russia, the dispar- ity in regional labor market conditions is much more pronounced, reflecting the immense economic heterogeneity of Russian regions. Regional disparities tend to be persistent--regions showing better performance at the outset of the transition also tended to perform bet- ter in later phases--suggesting that they may be of a long-term nature, rather than a transitory phenomenon. Job creation and employment opportunities are clustered mainly around large urban agglomerations with diversified industrial structures--particularly with large and expanding service sectors, developed infrastructure, and a skilled workforce. In contrast, employment prospects are dire and unemploy- ment high in monoculture (highly specialized) industrial regions, which suffer from idiosyncratic demand shocks. For example, in Bul- garia and Moldova in the early 2000s, employment was expanding almost exclusively in the capital region. During the same period, employment in Croatia grew in only 4 of 21 regions and in Poland in only 3 of 16 regions. Unemployment in rural agricultural regions is usually lower than that in regions with an industrial monoculture because workers in such regions have recourse to subsistence agricul- ture and can take temporary and seasonal jobs. In these regions, the problem is underemployment, rather than open unemployment. The labor market inequalities are sustained by relatively low inter- regional migration. The main factors accounting for low interregional labor migration rates in transition economies include the following: 18 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.6 Wage Inequality in the CIS Is Higher than in the CEE Ratios of 9th decile to 1st decile, 2002 Azerbaijan Russian Fed. Kyrgyz Rep. Serbia & Montenegro Armenia Estonia* Romania Ukraine Bulgaria Belarus Lithuania Hungary* Latvia Poland Slovenia OECD level Czech Rep. 0 2 4 6 8 10 12 14 Low income CIS CEE Middle income CIS SEE Sources: UNICEF TransMONEE database; and Bank staff calculations. Note: * Signifies 2001. · High nationwide unemployment rates may discourage internal migra- tion because they indicate falling probability of finding employ- ment. For example, in Bulgaria in the early 2000s, the unemployment rates were high in absolute terms, even in rela- Overview 19 tively low unemployment regions (for example, 14 percent in the capital city). A similar situation occurs in Poland. · Skills mismatch. New jobs are created mainly in the service sector in urban regions, and these jobs may require skills not available to unemployed blue-collar workers in other regions. For example, it is estimated that in Bulgaria and Lithuania, more than 20 percent of the unemployed do not find a job because their skills fall short of employer requirements (World Bank 2002a and 2002c). Related to this is the inefficiency of the spatial matching (that is, the process by which unemployed workers find employment in other regions). · Social benefits might have provided disincentives for migration, especially at an early stage of the transition (they were gradually reduced or abolished). In European transition economies, these were mainly cash benefits, while in CIS economies, they were usu- ally in-kind benefits, such as firm-provided housing. · Informal safety nets from informal sector employment or income from subsistence farming may induce job searchers to stay at home, rather than move elsewhere in the country. · Inefficiencies in the housing markets greatly increase migration costs (for example, through high rents). The cost of renting a studio apartment in Warsaw accounts for around 70 percent of the aver- age monthly net wage of a less skilled worker. · Liquidity constraints may make migration an unaffordable option. The Drivers of Labor Demand during the Transition The observed changes in the Region's labor markets during the tran- sition have been strongly influenced by the process of firm restruc- turing and massive reallocation of resources across firms, sectors, and locations. Faced with radical changes in the institution and market environment, firms in all countries have been forced to adapt their behavior. Some have been able to seize new opportunities, but many others have simply tried to survive in a market environment to which they were unaccustomed. The ensuing process of restructuring has involved major upgrading of technologies and production processes and hiring or (more often) laying off workers. As a result, some firms have managed to boost productivity, but some have been forced to exit the market. The process of economic transformation has been characterized by two interlinked processes: 20 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union 1. Within-enterprise reforms, spurred by higher competitive pressure. These reforms comprise improvements in governance and man- agement; better organization of producing and marketing; and introduction of new, more efficient technologies. 2. A more efficient allocation of resources, which also involved realloca- tion of jobs and workers away from declining, less productive firms, sectors, and regions toward expanding, more productive ones. Resource reallocation is important to promote output and productivity growth in any market economy, but it has assumed an even greater role in transition economies, given their highly distorted industrial structure inherited from the central-planning period. An important role in the process of reallocation of resources has been played by the entry of more productive firms and the exit of obsolete ones. Restructuring of existing firms, reallocation of resources across them, and the entry and exit of firms have all contributed to shape the evolution of productivity and employment. The effects on pro- ductivity are massive. For example, figure 1.7 shows that the contri- bution to productivity growth of the entry and exit of firms and the reallocation of labor across incumbent firms in Russia has increased significantly during the transition period. In addition, there is evi- dence that more productive firms tend to create new jobs, while inef- ficient firms are destroying them in a number of countries, including Croatia, Lithuania, and Moldova (figure 1.8 illustrates this pattern for Moldova). Firm Restructuring Firm restructuring was a result of imposing market discipline on inherited enterprises and encouraging the creation and expansion of new enterprises. Discipline forced old enterprises to release assets and labor, which were then potentially available to restructured and new enterprises. It did this by hardening budget constraints, introducing competition in product markets, providing exit mechanisms, and monitoring managerial behavior to generate incentives for produc- tion and innovation. Encouragement entailed policies to create an attractive and competitive investment climate in which restructured and new enterprises had incentives to hire labor and to invest in expansion (World Bank 2002e, xvii). Within-enterprise reforms engendered by the imposition of disci- pline resulted largely in labor shedding and what is called "defensive Overview 21 FIGURE 1.7 Labor Reallocation Has Played an Increasing Role in Promoting Labor Productivity Growth in Russia 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 ­4 ­2 0 2 4 6 8 % growth Exit Entry Cross Between Source: Brown and Earle 2004. Note: Entry: the contribution of new firms to overall labor-productivity growth; Exit: the contribution of exiting firms--as they tend to be less productive than the average, their exit contributes to raising productivity growth (on average); Cross: the product of changes in productivity and changes in market shares--it is negative if firms with rapidly growing produc- tivity lose market shares; Between: captures the gains in aggregate productivity coming from a reallocation of resources from lower to higher productivity firms. restructuring." Often, the impetus for defensive restructuring was privatization. In the "old" enterprises, productivity gains have been achieved largely through shedding redundant labor, which allowed the old firms to reduce costs and thus stay in the market. In addition, discipline led to the closure of other old, but inefficient firms, thus 22 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.8 The Rate of Job Creation Is Higher for More Productive Firms in Moldova Employment Growth (percentage) by Productivity Level in 2001 10 5 % 0 growth, ­5 Employment ­10 ­15 ­20 Low Medium High Productivity levels, 2001 Source: Rutkowski 2004a. raising average productivity. In addition, the entry of new firms fur- ther stimulated efficiency enhancements by incumbent firms. In background work to this report, which analyzed the effects of restruc- turing and reallocation of labor on productivity for a group of CEE countries, Russia, and Ukraine,13 productivity growth was found to be clearly led by defensive restructuring and the associated labor shedding.14 Indeed, those firms that achieved higher-than-average productivity growth during the transition period also experienced sig- nificant downsizing, with a fall in their share of total employment. At the same time, many low-productivity industries, sheltered from competitive pressure, have managed to contain job destruction, but may still have to go through a period of downsizing and restructur- ing, which will likely cause further increases in unemployment. Strategic restructuring--with new investment and job creation-- seems largely confined up until now to foreign-owned firms. For example, evidence for four countries (Hungary, Romania, Russia, and Ukraine) suggests that all privatized firms experienced significant improvements in their productivity performance. The acceleration in productivity was much stronger in foreign privatized firms, possibly because they had better characteristics even before the ownership change but also because of foreigners' advantages in accessing finance, new technologies, the latest managerial techniques, and Overview 23 world markets, which outweighed any disadvantages resulting from unfamiliarity with local conditions. But where the differences between foreign and domestic privatiza- tion become even more evident is in employment. Foreign privatiza- tion is associated with increased employment and wages in all countries. Domestic privatization has been less kind to workers: wages suffered in all four countries, and employment rose in only a few instances. There are also significant differences across types of work- ers: employment composition and relative wage changes have been significantly biased toward white-collar employees. This is consistent with the idea that foreign privatization, by promoting the upgrading of production processes with new technologies, tends to be skill-biased. It also suggests that although foreign privatization has improved the welfare of workers overall, it has also increased inequality. Reallocation of Labor and Firm Demographics In all market economies, many firms enter and exit most markets every year. But in transition countries, the pace of firm creation and destruction during the transition has been phenomenal. In CEE coun- tries, about 20­25 percent of firms were created or destroyed during the transition (compared with an average of 10 percent in OECD countries). Firm entry largely outpaced firm exit, which is obviously related to the process of transition and is not sustainable over the longer run. Yet it points to the fact that new firms not only displaced obsolete incumbents in the transition phase but also filled in new markets that were either nonexistent or poorly populated in the past. The entry of new firms contributed markedly to job creation. As an example, in Russia, before the transition, firm turnover (entry and exit of firms) accounted for less than 20 percent of overall job turnover. During the transition, the contribution of firm turnover to job flows increased strongly. In the countries for which firm-level data are available, in the initial phases of the transition, the entry of firms contributed to about 40 percent of total job creation in Estonia, Hungary, and Latvia and to more than 70 percent in Romania and Slovenia.15 Moreover, a stronger (positive) correlation exists in tran- sition countries than in OECD countries between the number of entrant firms in a given country and industry and the overall job cre- ation in that country and industry. By contrast, there is almost no cor- relation between exit of firms and job destruction, while in OECD countries this correlation is stronger. In other words, the entry of new firms plays a significant role in job creation. However, the exit of obsolete firms, while promoting productivity, does not strongly con- 24 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union tribute to job destruction, which largely comes from downsizing of surviving firms. Not surprisingly, the contribution of firm entry to overall job cre- ation declined over the transition period. After having filled the pre- transition void in certain activities, the characteristics of job flows converged toward those observed in market economies: job creation and destruction come increasingly from within firm adjustment, rather than from the entry and exit of firms. In this context, Russia and Ukraine--two laggard reformers--saw the reverse patterns, with an increasing role of firm entry in total job creation in the second half of the 1990s. Firm turnover is also important in promoting productivity. Con- trary to the evidence in most OECD countries, where the lack of expe- rience and small size often make new firms less productive than the average incumbents, in transition economies new firms have been more productive than existing firms. They have been able to fill in new market niches and adopt new and more efficient technologies. Thus, firm entry is critical not only for job creation but also for pro- ductivity growth. Figure 1.9 shows the contribution of the entry of more productive firms and the exit of less-productive ones to aggre- gate productivity in manufacturing. During the transition, in Estonia, Latvia, and Romania, 30 to 40 percent of overall productivity growth resulted from the creation and destruction of firms. The Role of the Region's Policy and Institutions All countries in the Region have made major progress in reforming their policy and institutional settings to accommodate the require- ments of a market economy. But large differences remain in the investment climate across countries in the Region and, in different ways, all countries still have an ambitious agenda ahead to remove constraints to firms' entry, investment, and job creation. The initial deterioration of labor market conditions in most economies of the Region was the result of the transition shock, which caused a dra- matic fall in output and a correspondingly large fall in labor demand. The resulting shedding of redundant labor and closing of loss-making enterprises caused large inflows into unemployment. Yet, the ability of the Region's economies to recover from the transition shock and to reduce unemployment has been largely determined by the rate of creation and expansion of new firms and consequently by the size of the new sector, as well as by the ability of existing firms to restructure to ensure profitability and improve competitiveness. Overview 25 FIGURE 1.9 Firm Entry and Exit Are Critical for Productivity Growth The Contribution of Entry and Exit of Firms to Total Labor Productivity Growth Estonia Latvia Korea, Rep. of Indonesia Taiwan, China Romania Chile Portugal Netherlands Slovenia Colombia United States Hungary France W. Germany Argentina 0 5 10 15 20 25 30 35 40 45 50 % of total productivity growth Countries in the Region Other countries Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Data show the sum of the contributions from new entrant firms and exiting firms to total labor-productivity growth in manufacturing. Data cover different pe- riods for the countries. For all but three countries, the contributions are calculated on the basis of five-year rolling windows. For Hungary, Indonesia, and Romania, the data refer to three-year rolling windows and tend to underestimate the contribution of new firms to total labor-productivity growth. The decision of firms to create more, and more productive, jobs and that of workers to stay in the labor market and seek more- rewarding jobs depend on a complex set of factors: 1. Macroeconomic policy setting. Macroeconomic instability and financial crises have both contributed to discouraging enterprises from undertaking new investment and creating new jobs. At the same time, the combination of relatively high interest rates and loose fis- cal policy may have contributed in some CEE countries to reduc- ing the employment content of growth. 2. The cost of doing business. Private investment is often discouraged in the Region's countries by a set of factors that increase the costs of new investment, the risk associated with it, and barriers to compe- tition faced by firms. 3. Wage flexibility has increased during the transition in the Region; however, there are still instances in which wage floors or govern- ment intervention prevents wages from adjusting. 26 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union 4. Employment protection legislation. Labor reallocation is also influ- enced by regulations on hiring and firing. Although several coun- tries have reformed these regulations to better conform with market requirements, reforms have often focused on creating flex- ibility at the margin (for example, temporary contracts in CEE). And in the CIS countries, labor flexibility is often ensured by non- compliance with rigid regulations. 5. Social benefits played an important role in smoothing the costs of the transition, but overly generous benefits (as in the early phases of the transition in several CEE countries) weakened job search incentives.16 Each of these five factors--four on the demand side and one on the supply side--has played a different role across the countries of the Region. But, in general, the dominant role has been played by the demand-side factors, while the supply-side effects were relatively limited. Indeed, in transition economies, the rate of job destruction has been--and, in many cases, still is--higher than the rate of job cre- ation, which has led to a fall in the number of jobs and an increase in unemployment. As a result, the number of unemployed per job vacancy is high in most transition economies. At the same time, the fraction of the jobless who receive unemployment benefits (20­30 percent) is relatively small. This suggests that the outflows from unemployment to jobs are mainly impeded by the lack of job open- ings, rather than by labor supply disincentives.17 Macroeconomic Policies The transition to a market economy has been characterized in some countries by macroeconomic instabilities and financial crises that have had repercussions in the neighboring countries. Since the mid-1990s, there has been a tendency, especially in CEE countries, toward relatively tight monetary policy and loose fiscal policy. This combination is estimated to have contributed to jobless growth.18 The tightening in monetary policy necessary to reduce inflationary pressures caused high real interest rates in the CEE since 1995. Empirical evidence suggests that high real interest rates tend to reduce capital investment and thus slow down the required adop- tion of new technologies and production processes, as well as the development of new activities. In turn, a slower pace of capital for- mation can easily translate into the postponement of defensive restructuring (see below), with less hiring of workers and thus a lower rate of employment growth. Overview 27 There is also evidence of a loosening in fiscal policy in the after- math of the 1998 recession, when employment was sharply declin- ing.19 Because fiscal loosening implied a likely future increase in interest rates, and possibly in the tax burden as well, forward-looking entrepreneurs reduced up-front hiring (Boeri and Garibaldi 2004). In addition, there was a crowding-out effect induced by an increase in labor costs. In the past five years, real wages in the public sector increased at a higher pace than those in the private sector in several CEE countries. In addition to worsening the fiscal balance, the associ- ated high wage growth in the CEE public sector may have crowded out skilled workers from the private sector. GDP growth also had a strong effect on employment in the second phase of transition, but again macroeconomic policies may have diluted its impact. There is evidence that the rise in real interest rates and the loosening of the fiscal stance contributed to weakening the potential effect of GDP growth on employment and unemployment (Boeri and Garibaldi 2004). In other words, employment and unemployment would have been more responsive to changes in GDP growth had the macroeconomic environment not changed at the same time. Cost of Doing Business Improving the investment climate requires further action not only in reducing the costs of investment and the risk associated with it but also in promoting competition in the market (World Bank 2004d). Priorities along these three dimensions differ from one country to another.20 High costs of doing business discouraged the creation of new firms and firm growth, leading to a lower pace of job creation. Objective indicators of the costs of doing business, as well as entrepreneurs' per- ceptions, indicate that there is substantial room for improving the investment climate in the Region and thus fostering job creation. In most of the subgroups of countries in the Region, economic and reg- ulatory policy uncertainty and macroeconomic instability are seen as major obstacles (figure 1.10). High taxes, burdensome tax adminis- tration and associated corruption, and problems with financing access and cost also loom large. Not surprisingly, most firms in the Region claim that taxes have a major impact on their costs. This is common to most firms around the world, and it is not by itself a source of major concern for policy mak- ers. If properly used, tax revenues contribute to the provision of pub- lic goods and mitigate market failures. Two main aspects of the tax system have, however, a specific bearing on firms' costs in the Region. 28 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.10 Major Obstacles to Firm Activity, 2002 Proportion of Firms That Report the Following as Major or Severe Obstacles in the Region Economic & regulatory policy uncertainty Macroeconomic instability Tax rates Tax administration Cost of financing Corruption Access to financing Anticompetitive informal practices Customs & trade regulations Legal system/conflict resolution Crime, theft & disorder Business licensing & operating permits Skills & education of available workers Electricity Access to land Labor regulations Telecommunications Transportation 0 5 10 15 20 25 30 35 40 45 % of firms in the Region that report the above as major or severe obstacles Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new Business Environment and Enterprise Performance Survey (BEEPS) of 2005 are broadly consistent with those reported in this figure. However, some changes may have occurred for individual countries or subgroups of countries in the Region. Tax rates--especially payroll taxes--are particularly high in European transition economies, reflecting, to a large extent, inefficiencies in the pension and health care systems. Taxes on labor use are among the highest in the world--in the Czech Republic, Hungary, and Romania, the tax wedge on labor use is around or even above 50 percent, while it is around 40 percent in the OECD countries (figure 1.11). Labor taxes lower labor demand by raising labor costs, but they also lower labor supply by reducing the real consumption wage.21 In addition, they contribute to the growth of the informal sector. The result is lower formal sector employment. The employment effect of these taxes is especially strong in CEE economies, which combine high tax rates with effective enforcement.22 Also, tax administration (for example, tax inspections) is particularly burdensome in the CIS coun- tries, being associated with bureaucratic harassment and corruption. Many firms in the Region also point to the difficulty of accessing finance and the costs of it as major constraints on their ability to invest Overview 29 and create new jobs. There are large differences in access to finance in the Region. For example, while domestic credit to the private sector accounts for more than 100 percent of GDP in the European Union, it is less than 30 percent in CEE, and it drops to 16 percent (on aver- age) in the middle income CIS countries and to 10 percent in the low income CIS countries. The comparison between Croatia, with one of the most developed credit markets in the Region, and the Kyrgyz Republic, with one of the least developed credit markets in the Region, is stunning: in Croatia, domestic credit to the private sector represents more than 50 percent of GDP, while it represents only 4 percent in the Kyrgyz Republic. But even in Croatia, the credit mar- ket is less developed than in some developing countries, such as the Republic of Korea and Malaysia, where credit totals considerably more than 100 percent of GDP.23 Difficult access to credit is most common for small and newly estab- lished firms, which account for a large part of total job creation in most of the Region's countries. Improving financial markets and easing access to credit may therefore lead to significant improvements in the FIGURE 1.11 The Tax Wedge on Labor in the Region Is High, Often Higher than in Most OECD Countries The Difference between Labor Cost and Take-Home Pay as a Percentage of Labor Cost 70 pay 60 take-home 50 and cost 40 costs labor of labor % 30 a as 20 between 10 Difference 0 OECD OECD OECD OECD CEE SEE Low Turkey high-income mid-income low-income all income countries countries countries countries CIS Sources: OECD 2004b for OECD and CEE OECD, except Slovenia (based on 2001); Bank staff estimates for others (using 2003). Tax wedge calculated with reference to the average worker without dependents. For countries in the Region, aver- age workers do not include those involved in agriculture. OECD groupings exclude CEE OECD: (i) low (less than 40% wedge) includes Australia, Iceland, Ireland, Japan, Korea, Mexico, New Zealand, Portugal, Switzerland, UK, and USA; (ii) mid (40-50%) includes Canada, Denmark, Greece, Luxembourg, Norway, Spain and Sweden; (iii) high (51+ %) includes Aus- tria, Belgium, Finland, France, Germany, Italy, and Netherlands. 30 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union labor market. In Latvia, where access to credit is considered one of the best in the Region and comparable to the EU-15 average, successful entrant firms almost double their employment in the first four years of life. By contrast, in Romania, where access to credit is considered one of the worst in the Region, successful new firms increase their employ- ment by a mere 10­20 percent over the same time horizon. Labor regulations do not rank among top business concerns in the Region, especially in the CIS. In the less-advanced economies of the CIS, they are dwarfed by more important and binding constraints. Although strict on the books, labor regulations are often only weakly or selectively enforced, and firms do not comply with them. Finally, in the slower-reforming transition economies (again, mainly in the CIS), where the restructuring effort still lies ahead, few firms have gone through major downsizing, constrained by high dismissal costs. However, in European transition economies, where enforcement is stronger and enterprise restructuring more advanced, firms fre- quently view labor regulations as a significant obstacle to their oper- ation and growth. In CEE, the proportion of firms complaining about labor regulations reaches as much as 30 percent, and it is only slightly lower in SEE (25 percent). For comparison, in low income CIS coun- tries, this proportion is only 10 percent; it is somewhat more in mid- dle income CIS countries, which reflects in part lax enforcement and in part lower pressures on adjustment.24 Although many investment climate concerns are common for most transition economies, there is also substantial variation in the relative importance of the obstacles. In the transition economies that recently joined the EU, labor-related issues, such as skills of the available workers, labor regulations, and taxation, figure prominently (figure 1.12, panel A). In other European transition economies, a combina- tion of poorly functioning institutions (for example, ineffective legal systems and industrial conflict resolution) and bad governance (for example, anticompetitive and informal practices, as well as corrup- tion) limit firm growth and job creation (figure 1.12, panels B and C). In the middle income CIS countries, the predominant part is played by administrative barriers, such as numerous licenses and operating permits, burdensome tax administration (with associated bureau- cratic harassment and extortion), and inefficient regulations (figure 1.12, panel D). In the low income CIS countries, the constraints are more basic, ranging from unreliable infrastructure to underdeveloped institutions of a market economy (figure 1.12, panel E).25 One reason why labor-related issues are more pronounced in CEE, compared with those in other subgroups in the Region, is a faster pace of enterprise restructuring. First, a widespread job reallocation gives rise to a skills mismatch because the newly created jobs differ in Overview 31 FIGURE 1.12 Obstacles to Business Operation and Growth Vary by Subgroup Panel A--CEE, EU-Member Countries Skills of available workers Labor regulations Access to land Tax rates Crime, theft & disorder Anticompetitive/informal practices Telecommunications Transport Less important More important Legal system/conflict resolution Customs & trade regulations Licensing & operating permits Cost of finance Access to finance Tax administration Economic & regulatory policy uncertainty Corruption Electricity Macroeconomic instability Average Panel B--CEE, EU-Accession Countries (Bulgaria, Croatia, and Romania) Legal system/conflict resolution Access to finance (e.g. collateral) Corruption Anticompetitive/informal practices Licensing & operating permits Economic & regulatory policy uncertainty Cost of finance Tax rates Macroeconomic instability Less important More important Crime, theft & disorder Skills of available workers Labor regulations Transport Access to land Telecommunications Customs & trade regulations Electricity Tax administration Average skill content from jobs that were destroyed. Thus, the relatively large job reallocation in CEE explains why employers there see the skills of available workers as a major constraint to firm growth. Second, labor regulations become an effective obstacle only if firms face competitive pressures to adjust the size and the composition of their workforce to stay profitable. This in turn explains why employers in faster-restruc- turing CEE countries complain more often about labor regulations, although on the books they are less strict than in other subgroups of countries in the Region. 32 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.12 (continued) Panel C--Southeastern Europe Legal system/conflict resolution Anticompetitive/informal practices Customs & trade regulations Electricity Corruption Crime, theft & disorder Transport Cost of finance Less important More important Labor regulations Economic & regulatory policy uncertainty Macroeconomic instability Telecommunications Access to land Access to finance tax rates Licensing & operating permits Tax administration Skills of available workers Average Panel D--Middle Income CIS Countries: Belarus, Kazakhstan, Russia, and Ukraine Licensing & operating permits Tax administration Access to land Customs & trade regulations Skills of available workers Access to finance Crime, theft & disorder Macroeconomic instability Economic & regulatory policy uncertainty Less important More important Tax rates Cost of finance Labor regulations Corruption Telecommunications Transport Legal system/conflict resolution Anticompetitive/informal practices Electricity Average Administrative costs of setting up a new business affect the cre- ation of new firms most in the CIS and SEE. Costs for setting up a new business tend to be higher in the Region than in many industrial and emerging Asian economies, being particularly constraining in many of the CIS countries and in the countries of Southeast Europe. By contrast, most Central and Eastern European countries of the Region--most notably the Baltic States--have made significant improvements in reducing start-up costs. For example, the number of Overview 33 FIGURE 1.12 (continued) Panel E--Low Income CIS Countries: Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan Tax administration Electricity Customs & trade regulations Access to finance Licensing & operating permits Access to land Telecommunications Cost of finance Transport Tax rates Corruption Less important More important Macroeconomic instability Crime, theft & disorder Skills of available workers Economic & regulatory policy uncertainty Labor regulations Legal system/conflict resolution Anticompetitive/informal practices Average Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Bars show the deviation from the regional and factor averages. Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individual countries or subgroups of countries in the Region. start-up procedures is just 3 in countries such as Ireland and Sweden and 7 or 8 in the Baltic States, but their number is 14 in Azerbaijan and Ukraine and 19 in Belarus. Thus, reducing business start-up costs in the Region may foster firm entry, facilitate formal sector growth, and as such bring about faster job creation. Reducing the risks for firms and for workers is also a key challenge because they operate in a milieu of greater uncertainty. Firms in the transition economies have emerged during a period of profound eco- nomic and social transformation characterized by major macroeco- nomic shocks and profound structural changes. All these factors have certainly affected firms' perceptions about risk. Not surprisingly, most countries in the Region point to macroeconomic and political insta- bility as one of the main sources of concern in their operations. But other factors also contribute to the perceived business insecurity: security of property rights and contract enforcement, frequent regu- latory changes, arbitrary and selective application of the law, bureau- cratic harassment and extortion, and crime are important issues in virtually all countries of the Region, although less so in the new EU members. Thus, strengthening the rule of law, developing effective market institutions, and improving governance are key elements of 34 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union reducing risk and uncertainty and thereby fostering investment and job creation. Workers also have to confront uncertainties in their career prospects and job stability. During rapid transition, in which many jobs are destroyed and others are created in different firms, sectors, and often locations, these uncertainties are high. Government can help workers cope with these rapid changes by insuring against the risk of job and income loss and improving the matching function of the labor market. Better labor market information, job search assis- tance, training, and facilitation of labor mobility can contribute to achieving this goal. Virtually all transition countries introduced income support schemes and a range of active labor market programs (ALMPs), although they are more generous (in coverage, benefit level, and duration) in CEE than in other subgroups, especially in the low income CIS countries. However, balancing income support to promote reallocation with possible job-search disincentive effects has been, and still is, a challenge in many CEE countries. Introducing greater competition into markets has helped longer- term job creation in the Region. One of the most visible and pervasive changes during the transition was the introduction of competitive forces into markets for goods and services--by liberalizing prices and trade flows, by imposing harder budget constraints on state firms, and by shifting ownership to the private sector (World Bank 2002e). More- competitive product markets bring easier firm entry, less administra- tive barriers to firm growth, and greater firm efficiency, including wage adaptability. All these changes are conducive to faster job creation, although productivity gains can lead to job losses in the short term. There are, however, large differences in the pervasiveness of com- petitive forces in the markets of the Region's countries, with product markets being most competitive in more-advanced CEE countries and least competitive in the CIS. For example, 90 percent of firms in Poland face significant competitive pressures and thus incentives to improve efficiency, while only 40 percent of firms face similar pres- sures in Georgia.26 Some of these differences have been the result of different modes of privatization, which have often been the main trigger of firm restructuring and the promotion of efficiency. The positive effect of privatization on productivity has been stronger in CEE countries than in the CIS countries because the for- mer have placed fewer restrictions on--and sometimes actively encouraged--foreign participation in privatization and concentrated ownership. The defensive restructuring strategy adopted by many domestically privatized firms could go on for a while, given the large overmanning in previously state-owned firms, but further improve- Overview 35 ments in their performance will soon depend on their ability to invest, adopt new technologies, and hire new staff.27 Promoting competition via easier entry of new firms and leveling the playing field can speed up the transition from defensive restructuring to strategic restructur- ing and job creation. Wage Flexibility The wage determination process in the Region has been decentralized and liberalized during transition and generally resembles that in mature market economies. But liberalization has gone much further in CEE than in SEE and the CIS.28 In CEE countries, where the new private sector plays a bigger role than in other parts of the Region, wages are mostly determined at the enterprise level and linked to worker productivity. Union presence in the new private sector is weak and competition is high, so there is little space for workers to appro- priate part of the firms' market rents29 or for wages to be set above the competitive level. Evidence exists that regional wages are sensi- tive to regional unemployment rates (the so-called wage curve), as they are in mature market economies (Huber 2004). In contrast, wage setting in SEE and the CIS is less competitive, largely because of a still dominant role played by the "old" sector, including public and many privatized enterprises. In some cases, wage rigidities in the form of binding minimum wages could have contributed to unemployment among disadvan- taged groups. In many CEE countries, minimum wages are centrally set at a relatively high level (around 40 percent of the average wage). At this level, minimum wages may produce some disemployment effect among the less-skilled and younger workers, especially in eco- nomically depressed regions of the country where underlying market wages are well below the national average. For example, in econom- ically depressed regions of Poland, the minimum wage is roughly 10 percent less than the average wage of low-skilled workers and 25 per- cent less than that of young workers (Rutkowski and Przybyla 2002). Similarly, the recent substantial minimum wage increase in Hungary was found to adversely affect job retention and job-finding probabili- ties of low-wage workers (Kertesi and Köllő 2003). In SEE and especially the CIS, wage determination is more cen- tralized and still influenced by the government. This is related to a higher share of the public sector in employment and the behavior of privatized enterprises, which often mimics that of state-owned firms. In addition, union bargaining power is significantly higher in the public sector than in the new private sector. As a result, the 36 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union wage structure in the formal sector in SEE and the CIS is more rigid. Nonetheless, these countries have achieved a considerable degree of de facto wage flexibility through nonpayment or delayed payment of wages. Employment Protection Legislation Employment protection legislation (EPL) is (on average) stricter in the Region than in OECD economies (figure 1.13). Historically, employment protection has been particularly strict in CIS and SEE countries, and somewhat less strict in most CEE countries.30 This means that employers in the CIS and SEE may face high--mainly procedural--costs of firing redundant labor. This is likely to discour- age them from hiring in the period of economic upturn, to avoid future firing costs in some subsequent downturn. Apparently strict EPL in the Region might have contributed to a slow pace of job cre- ation. However, the enforcement capacity in countries with strict EPL--largely in the CIS--is often weak, which means that the actual "bite" of the EPL tends to be less than implied by the law (table 1.1).31 In contrast, EPL is more binding in CEE because of stronger enforcement, despite more liberal regulations.32 This assessment is consistent with the perceptions of employers, who deem labor regu- lations a significant obstacle in the new EU member countries (and also in Turkey), but not in the other parts of the Region (figure 1.14). Available evidence suggests that the combination of stringent EPL with effective enforcement did indeed slow job creation in the Region. For example, in Croatia and the Slovak Republic, where EPL before the reforms was extremely strict and compliance was high, the rates of job destruction and job creation were much below those in CEE economies with more liberal EPL, such as the Baltic States, Hun- gary, or Poland (World Bank 2002a and 2003a). Thus, excessive pro- tection of existing jobs comes at a price of slower job creation and also slower enterprise restructuring and productivity improvements. Moldova, where the rate of job turnover is relatively high, shows that strict EPL has little effect on enterprise restructuring if enforcement is weak (World Bank 2004a). All in all, EPL has likely slowed the pace of job creation in those countries of the Region that combined strict regulations with effective enforcement, such as the countries of the former Yugoslavia. In the rest of the Region, high firing costs might have adversely affected job creation in selected firms and had some effects on the composition of employment, but the overall effect on employment has likely been limited. Overview 37 FIGURE 1.13 The Region's Countries Have More Stringent Regulations on Hiring and Firing than OECD Countries Do 0.7 0.6 regulated 0.5 More index 0.4 firing 0.3 and 0.2 Hiring regulated 0.1 Less 0.0 Sub- East Asia Eastern Latin Middle OECD South CEE SEE Middle Low Turkey Saharan & & America East & Asia income income Africa Pacific Central North CIS CIS Europe Africa The Region Temporary employment regulations Overall employment regulations Source: World Bank, Doing Business Database (2005). TABLE 1.1 A Typology of Employment Protection Legislation and Enforcement Flexible EPL More restrictive EPL Very rigid EPL Weaker enforcement Albania, Kazakhstan Armenia, Georgia, Azerbaijan; Belarus; Russian Federation, Bosnia and Herzegovina; Serbia and Montenegro, Kyrgyz Republic; Moldova; Turkey Ukraine; Uzbekistan Intermediate enforcement Bulgaria Croatia; Macedonia, FYR; Romania Stronger enforcement Czech Republic, Latvia, Lithuania, Estonia, Hungary, Poland, Slovenia Slovak Republic Source: Bank staff calculations. Note: (a) Strictness is defined as whether country EPL index below world average EPL index in Doing Business 2005 (flex- ible), up to 20% higher (more restrictive) and more than 20% higher (very rigid). (b) The degree of enforcement is measured by the size of the informal economy--the higher the informal economy, the weaker the estimated degree of enforcement (Schneider and Klinglmair 2004 and country studies, where available). The typology is approximate because the indices are subject to margins of error. 38 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.14 Labor Regulations Seem to Be a Binding Constraint Only in the New EU Member Countries and Not in the Other Parts of the Region Relative Importance of Labor Regulations, 2002 CEE, EU-8 Turkey SEE Less important More important Middle income CIS CEE, EUA-3 Low income CIS Average Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Bars show the deviation from the regional and factor averages. Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individual countries or subgroups of countries in the Region. Social Benefits One way to accommodate labor market slack in most of the Region's countries has been through the withdrawal of older workers through the promotion of early retirement schemes and disability pensions. This contributed to raising social expenditures and led to the surge in taxes on labor use. Moreover, social benefits were set at a fairly gen- erous level at the beginning of the transition in many of the Region's countries, creating some disincentive effects on the supply side. The same happened with unemployment insurance benefits that pro- vided income support for a long period of time (18­24 months in some countries) and a high replacement rate. But unemployment insurance benefits have been streamlined in all countries, and their generosity--and, in particular, their coverage--is fairly low nowa- days: only 20­30 percent of the unemployed receive unemployment benefits in most of the Region's countries.33 In CEE, unemployment benefits are more generous than in the CIS, but are still at the lower end of the OECD range. In the CIS, the generosity is so low that few unemployed have an incentive to register, which partly accounts for low registered unemployment in those countries. Unemployment and other social benefits may have the largest effect in economically depressed regions. Despite their low overall generosity, unemployment benefits and other social benefits in CEE and SEE may create labor-supply disincentives in depressed regions, where they account for a higher proportion of the market wage. Fam- Overview 39 ily benefits and social assistance benefits are set at a national level and do not consider the large differences in the cost of living across areas of the Region's countries. Although very limited in the capitals and other dynamic regions, they represent a higher proportion of under- lying market wages in more depressed areas. In CEE, social benefits were a factor behind longer job-search dura- tion and labor force withdrawal (especially initially, when their gen- erosity was greater), but they facilitated restructuring and labor reallocation. For example, a study for the Slovak Republic found that workers who received unemployment benefits or social assistance spent more time (about two months longer) unemployed than work- ers who were not entitled to these benefits. At the same time, however, benefit recipients found new jobs more often, and these jobs turned out to be better matches than the ones obtained by nonrecipients (World Bank 2002d). In contrast, in the CIS, the lack of an effective social safety net made workers extremely reluctant to part with their jobs, contributing to the slower pace of enterprise restructuring. In addition, poor social safety nets in CIS countries caused widespread informal sec- tor employment, including work in subsistence agriculture. The Policy Challenge: Promoting Job Creation in the Region The Region's countries face the daunting challenge of fostering the creation of more and better jobs. This challenge is not unique to these countries. Most Western European neighbors have faced high and persistent unemployment for a long time, and only some of them have succeeded in significantly improving labor market conditions. Creating more and better jobs requires improvements in the invest- ment climate, a climate in which firms of all types find the incentives to invest productively, adopt new technologies, and ultimately hire more workers in more-productive, and thus more-rewarding, jobs. In the countries of the Region, this already complex task is part of an even broader process of economic transformation associated with profound institutional and structural changes. Ultimately, structural reforms and improvements in the investment climate are necessary to foster high investment rates and create foundations for sustainable economic growth (box 1.2). In all of the Region's countries, there is substantial scope for invest- ment climate improvements, although the conditions vary substan- tially across country groups. Obstacles to firm growth are generally more severe in SEE or low income CIS countries than in the new EU transition countries, where structural reforms have progressed the 40 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 1.2 In Most of the Region's Countries, Higher Investment Rates Are Necessary to Accelerate Economic Growth and Job Creation Traditional growth theories emphasize the role of saving and investment in physical capital as important drivers of long-term economic growth. Investment is needed to expand and upgrade physical capital and thus promote output growth and productivity. And if technological progress is also embedded in physical capital, investment is needed to adopt new technologies and again to promote growth. Recent growth theories have also stressed the importance of complemen- tary investment in human capital and the fact that there could be endogenous forces that coun- teract the traditional decreasing returns in investment in both physical and human capital. The transition process from a central-plan to a market economy entails a massive investment in physical capital to replace obsolete equipment with new and to expand in new sectors and ac- tivities. The experience of some of the rapidly growing countries in East Asia clearly suggests that their extraordinary performance was also promoted by high investment rates, sustained for a prolonged period of time (World Bank 1993). For example, during the 1990s, the investment- to-GDP ratio in East Asian countries, such as Korea, Malaysia, or Thailand, averaged 36 percent. In most transition economies of the Region, the gross-investment-to-GDP ratio was around 20 percent over the past decade (figure A), largely resulting from low domestic savings. Higher in- vestment rates may be necessary to accelerate or sustain economic growth and create new jobs. This in turn requires incentives for domestic savings and an environment in which firms find it profitable to invest and engage in strategic restructuring. The impact of investment on economic growth and employment is also determined by the structure of investment. A part of gross investment is simply the replacement for depreciat- ed capital and, as such, may not contribute to job creation.a Moreover, investment can either expand "productive" capital (such as machinery and equipment) and thus contribute to job creation or add to "nonproductive" capital (for example, housing), in which case its effect on employment is negligible. There is a substantial variation across countries of the Region in the proportion of total investment allocated to productive uses. Productive (nonhousing) in- vestment as a share of gross fixed capital formation (GFCF) ranges from less than 60 percent in the Slovak Republic to about 80 percent in Poland and to some 90 percent in the Czech Republic. As a comparison, in Korea, productive investment accounts for 85 percent of GFCF.b a. Gross investment minus capital depreciation is net investment, which would be a more relevant indicator of the poten- tial for job creation; however, data on net investment are not available on a comparative basis. b. National accounts of OECD countries (OECD 2003), average share for 1999­2003. Overview 41 FIGURE A In Many Transition Economies of the Region, Investment Rates Are below 20 Percent, Weakening Growth Prospects Gross Fixed Capital Formation as Percentage of GDP (1999­2003 average) Azerbaijan Czech Rep.* Estonia Slovak Rep. Latvia Slovenia Hungary* Belarus Croatia* Poland Kazakhstan Lithuania Bosnia & Herzegovina Albania* Romania Ukraine Georgia* Uzbekistan Armenia Bulgaria Kyrgyz Rep. Russian Fed. Moldova Macedonia, FYR Serbia & Montenegro Tajikistan 0 5 10 15 20 25 30 35 Gross fixed capital formation as % of GDP, 1999­2003 average Low income CIS CEE Middle income CIS SEE Source: World Development Indicators. Note: * = 1998­2002. 42 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union most. But even in the most-advanced transition economies of CEE, the costs of doing business can be substantially lowered. Entrepre- neurs in CEE still complain about high regulatory and economic pol- icy uncertainty, difficult access to finance, poor governance, onerous regulations, and administrative barriers. There is also much to be done to improve infrastructure, from improving road conditions to developing the information and communication technology.34 After all, the experience of developed market economies shows that improving the investment climate is a continuous process that should respond to changing economic conditions, globalization, and techno- logical progress.35 The Region's countries also need to develop an adaptable labor market. Currently, labor codes in the Region, especially in the CIS and in SEE, go well beyond what can be observed in most OECD countries. But the overly generous provisions remain largely on paper because firms find it too costly to comply and thus evade or circum- vent rigid labor regulations. A possible solution is two-pronged. First, employment protection regulations need to be revised to remove unduly restrictive and overly detailed provisions. Second, social part- ners--trade unions and employers--should be given a greater role in determining employment relations. This requires that both sides are adequately represented and that their power is balanced. Employers' organizations, adequately representing the interests of new private firms, should develop and have due voice, especially in the CIS and in SEE. At the same time, workers need to have adequate representa- tion in the private sector--where currently unions are often nonex- istent--so that their rights are protected. Deregulation of employment relations, coupled with strengthening the role of collective bargain- ing, can benefit both sides. Workers would benefit from de facto rather than de jure employment protection because firms would have less incentive for noncompliance and enforcement would be easier. At the same time, employers would be freed from regulations that are particularly costly to firms but of less value to workers. Challenges facing countries of the Region have their source in a combination of the Region's history and its income level. Numerous administrative barriers (for example, permits) or strict employment protection are largely the legacy of central planning, although they are quite typical of developing countries across the world (Turkey is one example). Difficult access to finance or inefficient legal systems reflect institutions of a market economy that are underdeveloped because of both the history of central planning and the stage of eco- nomic development. Economic policy uncertainty, corruption, and crime have been engendered by the regime changes, although again Overview 43 they are typical of much of the developing world. Finally, poor infra- structure is closely associated with relatively low income levels. How severe are the policy challenges? How far are the Region's subgroups from their natural comparators (that is, countries at a sim- ilar level of economic development that have been market economies)? This study focuses on one central dimension of the investment climate--access to finance--which is seen as a major obstacle to firm growth in virtually all countries of the Region. Figure 1.15 indicates that credit markets are underdeveloped in all of the Region's subgroups relative to their income level. For example, Arme- nia, Georgia, and Ukraine are at an income level similar to Bolivia's. But domestic credit to the private sector (as a share of GDP) is much higher in Bolivia: twice as high as in Ukraine and more than five times as high as in Armenia or Georgia. In the same vein, access to credit in Lithuania or the Slovak Republic is more limited than in Chile or Malaysia, in spite of a similar GDP level. This suggests that historical factors, such as the legacy of central planning, still have a bearing on the Region's economic performance. Evidence provided in this report indicates that access to finance has a significant impact on job creation. This, in turn, suggests that the Region's labor market problems at least partly stem from the uncompleted transition. Thus, FIGURE 1.15 Access to Finance Is More Difficult in Transition Economies than in Market Economies at Similar Income Levels 120 2003 GDP, 100 of % as 80 sector 60 private to 40 credit 20 Domestic 0 High-income Upper-middle-income Lower-middle-income Low-income countries countries countries countries Countries outside the Region Countries in the Region Source: World Development Indicators. Note: * = Slovenia. 44 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union further structural reforms are necessary to develop the institutions of a market economy and foster labor demand. There is no question that the Region's countries have been through major reforms already. The EBRD transition index attests to the many reforms that have been made in the past 15 years in privatization, enhancing competition, developing the financial market, and so forth. Reforms have to be seen, however, as a process through which con- straints are identified and continuous improvement and fine-tuning of policies are made. International experience has shown that essential elements of reform processes include setting priorities, managing indi- vidual reforms, maintaining momentum, and strengthening govern- ment capabilities (World Bank 2004d). Setting priorities and sequencing reforms are of particular importance because too broad a reform agenda is difficult to implement and the impact is likely to be limited because of resource constraints, including "political capital," and opposition from the groups whose interests are likely to be hurt. The policy priorities need to be set according to country circum- stances. Although the transition countries of the Region have gone through similar economic changes over the past 15 years, they differ in their initial conditions, paths, speeds of reform, and consequently in their levels of economic development (figure 1.16). The next part of the overview highlights some of the main policy challenges the report has identified for each of the Region's subgroups: CEE, SEE, the middle income CIS countries (Belarus, Kazakhstan, Russia, and Ukraine), and the low income CIS countries. They focus on the labor market area of policy intervention, although in some subgroups, the improvements in the investment climate and the restoration of sus- tained growth are the key priorities for job creation, calling for a much broader policy agenda. Main policy priorities for each of the Region's subgroups are summarized in table 1.2. Central and Eastern European Countries: Reducing the Cost of Labor and Mobilizing Labor Supply These countries have generally started the macro and structural reform process earlier than other transition countries and have often adopted more aggressive reforms. They have achieved impressive results, with output exceeding its pretransition level. They have also successfully created institutions of a market economy, which has led to their EU accession. But these impressive results should not hide the turbulent phases of this massive transformation and its associated social costs. Moreover, impressive macro and structural reforms have not yet delivered major improvements in labor market conditions. Overview 45 TABLE 1.2 A Summary of Key Policy Measures to Improve Labor Market Outcomes in Transition Economies of the Region The Region's subgroups Policy priorities Central and Eastern Europe · Reduce taxes on labor through reforming pension and health care systems · Enhance wage adaptability (for example, through reducing/differentiating minimum wages) · Liberalize employment protection regulations for permanent contracts · Improve labor supply incentives in backward, low-wage regions by adjusting benefit levels Southeastern Europe · Enhance product market competition and reduce administrative barriers · Reduce taxes on labor · Decentralize wage determination · Liberalize employment protection legislation Middle income CIS · Enhance product market competition and reduce administrative barriers · Decentralize wage determination · Deregulate labor relations and focus on enforcement of key labor standards · Improve the effectiveness of the social safety net to encourage labor reallocation Low income CIS · Develop key institutions of a market economy and invest in infrastructure · Reduce administrative barriers · Deregulate labor relations and focus on enforcement of key labor standards · Develop social safety net, particularly for those without access to formal insurance mechanisms Source: Bank staff calculations. Job opportunities in the formal sector are scarce, and unemployment is often high and persistent. There is a substantial potential in CEE to foster job creation through addressing country-specific investment climate constraints. Although the investment climate in CEE is generally hospitable and notably bet- ter than in other subgroups of the Region, each CEE country can improve its investment climate along specific dimensions. For exam- ple, there is room for lowering start-up costs in Hungary and the Slo- vak Republic. Registering property is difficult in Croatia, and contract enforcement is extremely time-consuming in Poland. In Bulgaria and Romania, there is substantial room for improving access to finance and the efficiency of the legal system. In many CEE countries, corrup- tion is still a serious problem. In general, the challenges are bigger for the EU accession countries (Bulgaria, Croatia, and Romania), where entrepreneurs identify more obstacles to business operation and more often perceive them as severe as in the new EU member countries. Reducing taxes on labor is necessary to encourage hiring. To decrease such high rates in a fiscally viable way, reforms of health and pension systems must make the systems more efficient and sustain- able. In particular, reviewing early retirement, disability, and sickness benefit policies should be part of the reform agenda, with the aim of tightening the eligibility criteria and eliminating the widespread abuse of the systems. 46 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 1.16 Constraints Reported by Firms Vary across the Region's Countries Security & stability Labor Regulation Infrastructure Taxation Finance Ukraine Georgia Bulgaria Source: From WDR 2005, with data from EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individual countries or subgroups of countries in the Region. Another way to facilitate job creation in CEE is to provide for greater adaptability of real wages to productivity and local labor mar- ket conditions. Decentralized (firm-level) wage setting has developed in most CEE countries, but some countries, such as Slovenia, have kept highly centralized bargaining. Some wage rigidities are also pres- ent in the state-owned and privatized firms, as well as in noncompet- itive sectors of the economy. Most countries have kept a rigid coefficient system as a basis for wage setting in the public sector. Sig- nificantly, wage flexibility is limited by the mandated minimum wages, which tend to be relatively high in CEE. Creating subminimum wages for young workers or in economically depressed regions, as in many EU countries, would promote job creation for less-productive workers who are hardest hit by unemployment. Poland set the example in CEE by introducing a youth subminimum wage (at 80 percent of the regular minimum wage) in an attempt to alleviate youth unemployment. Reform in this area is also important because social benefit programs are often tied to the minimum wage Overview 47 (that is, they are relatively high when the minimum wage is relatively high), creating additional disincentive effects on labor supply. One common strategy to promote flexibility in the labor market in CEE countries has been the liberalization of fixed-term and tempo- rary contracts. In this respect, CEE countries have followed the expe- rience of many Western European countries that, faced with mounting pressure to promote flexibility, liberalized these atypical contracts while maintaining strict regulations for regular contracts. In Western Europe, this strategy has reinforced duality in the labor mar- ket, with some workers enjoying high and even increased protection in regular jobs, but others bearing the cost of greater flexibility by moving from one short-term job to another. In CEE countries, creating jobs, even of short duration, can be ben- eficial, especially in a context of significant long-term unemployment. Still, it is important in the long term to foster regular employment cre- ation. In this context, employment protection regulations for regular contracts should be liberalized in accordance with the recent relax- ation of restrictions on the use of temporary contracts. More generally, employment relations should be reformed to pro- mote direct agreements between workers and employers.36 For his- torical reasons, in many countries of the Region, the statutory labor standards are still very high, and detailed regulations cover virtually all aspects of employment conditions (for example, working time, employment protection, wage formation, and in-work benefits). This leaves little room for negotiations and thus impedes the development of genuine collective bargaining systems and stifles social dialogue. Instead, overregulation contributes to noncompliance and a steady drift toward informal arrangements. A greater role given to collective bargaining agreements in determining employment relations is thus a way to instill more flexibility into the labor market while balancing the interests of workers and employers. Unemployment and related benefits have been reformed several times in most CEE countries, mainly to reduce the initial excessive generosity of the system. Therefore, the system does not currently create major labor supply disincentives. However, it may give rise to the "unemployment trap" in backward regions, where benefits account for a high proportion of the prevailing market wage of low- skilled labor. This is reinforced by the fact that some countries have extended benefit duration in backward areas or apply lax eligibility criteria. Although social assistance benefits are needed to cushion households from falling into persistent poverty, targeting and means testing should be reinforced to improve the effectiveness of these schemes. 48 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Passive labor market programs are not sufficient to deal with the problems of long-term unemployment and discouragement. Stronger job creation by private firms is obviously the main avenue for address- ing this serious social and economic issue. But active labor market programs, if properly designed and implemented, can also contribute by providing greater incentives for the unemployed to go back to work and for employers to employ disadvantaged individuals. Evi- dence suggests that ALMPs can improve the labor market prospects of some groups, although they do not increase aggregate employment and such programs may be costly. It is therefore crucial to pay special attention to their design and efficient targeting. They also need to be accompanied by monitoring and impact evaluation. The former enables the management of programs and fine-tuning, if necessary, while the latter provides feedback for future policies. Investments in human capital are necessary to keep pace with the growing demand for high skills. Although the workforce in CEE is (on average) well educated, employers have difficulty finding work- ers with the right skills. The problem of the skills gap and mismatch hits the most advanced reformers in the subgroup. In countries like the Czech Republic, Hungary, Latvia, and Poland, 35­40 percent of employers see the skills of the available workforce as a significant obstacle. (The proportion is even higher in Estonia.) Evidence sug- gests that a substantial portion of the long-term unemployed in CEE lack the skills required by the employers. In some part, this reflects the failings of the educational system: workers need to be equipped with broad and transferable skills and provided with life-long learn- ing opportunities to be able to meet the challenge of a dynamic labor market and to support firms' strategic restructuring. Southeastern European Countries: Greater Stability, Law Enforcement, and More Competitive Markets This subgroup of countries tends to be at an intermediary position between the most advanced CEE countries and the CIS countries. Unemployment is particularly high and persistent in SEE, and infor- mality is pervasive. The reforms in the subgroup have been delayed and slowed by the regional conflict, which has also led to the dis- ruption of regional infrastructure and economic ties. The subgroup of countries fares worse than CEE in product market competitive- ness, administrative barriers, effectiveness of the legal system, cor- ruption, crime, and disorder. Substantial investment climate improvements are thus necessary to foster firm growth and formal sector job creation. Overview 49 Investment climate improvements in SEE require reforming insti- tutions and changing government policies and behavior. One priority is to reduce corruption, which is widespread. Another is to enhance competition, because anticompetitive or informal practices are regarded as a major obstacle by almost 30 percent of firms in the sub- group. The same proportion of firms sees as a major obstacle to their operation the inefficient system of conflict resolution, which points to the need to reform the legal system. Finally, barriers to firm entry need to be removed, because starting a business in SEE can be diffi- cult, time-consuming, and costly.37 The tax wedge on labor is slightly lower in SEE than in CEE coun- tries, but it remains higher than the bottom end of OECD countries. Moreover, tax rates are identified as a major obstacle to operating a business by around 30 percent of firms. SEE countries should aim for a gradual reduction in payroll taxes, with a particular focus on health insurance taxes. This should be accompanied by efforts to improve compliance on tax payments by firms, including SOEs. Moreover, to make the reduction of payroll taxes fiscally viable, public-funded pro- grams should be made more efficient. This entails tightening the eli- gibility criteria (for example, for early retirement and disability pensions) and reducing the abuse of the benefit system (for example, sickness benefits and disability pensions) by able-bodied individuals. Wage setting in SEE may hamper achieving labor market equilib- rium. Wage pressure arises because unions tend to be influential in these countries (partly because of the history of self-management before 1990). But the role of the government in wage setting also remains strong. Thus decentralized bargaining at the level of the firm should be encouraged to enhance wage flexibility. Sectoral wage negotiations appear to be a particular problem. For example, through the setting of specific minimum wages by sector, such bargaining leads to extremely high levels of wages. For example, in Bosnia and Herzegovina, the minimum wage in the 2004 telecommunications sector collective agreement was double the economywide minimum wage (itself a high 55 percent of the average wage), equivalent to 110 percent of the economywide average wage. The compression of the wage distribution that results leads to high unemployment for vulnerable groups such as youth and for some less developed regions. Employment protection legislation remains fairly strict in SEE and clearly is not helping workers overall because a great number of "out- siders" are either employed in the informal sector or unemployed (figure 1.13). In fact, the share of formal employment decreases and an informal economy develops where enforcement of these strict reg- 50 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ulations is difficult, especially because of the weak capacity of the administration. The way forward is to deregulate labor relations while protecting core worker rights and to leave the determination of employment conditions to bargaining between employers and work- ers. This should be coupled with improving the effectiveness of labor inspection in enforcing core labor standards. Unemployment benefits tend to be at appropriate levels of gen- erosity in SEE countries. However, work test requirements (that is, ensuring that the unemployed is available and willing to take a job, if offered) could be strengthened, especially in regions where labor demand is more robust. Also, employment offices should make a clear distinction between the unemployed who register to find work and those who register to gain eligibility for health care and other bene- fits. This will help to focus resources on those who need job search assistance and at the same time will provide a clearer picture of unem- ployment.38 ALMPs are currently insignificant in SEE countries. More funda- mental investment climate improvements are needed to encourage job creation and improve labor market outcomes. However, the cost- effectiveness of ALMPs can be improved. ALMPs in SEE should be targeted at vulnerable groups (that is, those running a high risk of long-term unemployment). Given the limited capacity for designing costly programs, the most cost-effective way to improve the matching of unemployed to jobs in these countries is to improve the capacity of employment agencies to identify job vacancies and to provide job search assistance. Finally, to address the mounting issues of long-term unemployment, back-to-work benefits could be useful. Social assistance performance is mixed across SEE countries. The evidence at the level of the subgroup suggests that the effectiveness of programs has been greater in countries where financing is more cen- tralized, enabling targeting of the poorest areas. The current range of programs could be streamlined into core benefits to provide a decent but sustainable level of social assistance. Middle Income CIS Countries (Belarus, Kazakhstan, Russia, and Ukraine): Removing Administrative Barriers, Promoting Restructuring, and Improving the Safety Net These four countries have a high economic potential; however, the challenges are profound. Administrative barriers are still pervasive, and governance is often poor. Corruption is widespread. Competition is limited. Inefficient and burdensome regulations and numerous permits hamper business growth, as do underdeveloped financial markets. Overview 51 Onerous tax administration and customs and trade regulations are seen as the key constraints to firm operation and growth in this sub- group. However, these are likely to capture a broader spectrum of investment climate issues, which include complex and often unintel- ligible regulations, their discretionary interpretation, bureaucratic harassment (for example, frequent, lengthy, and disruptive tax inspections), selective enforcement and associated uncertainty, and extortion. Therefore, improving the quality of regulations, reducing the red tape, and limiting the scope for bureaucratic discretion (along with other general improvements in governance) would create a more favorable investment climate and would support the develop- ment of the private sector and job growth. Another important reform concerns wage determination. Even more than in SEE countries, the government has a strong role in wage setting in middle income and other CIS countries. In fact, the formal private sector tends to follow wage increases in the public sector. The role of the government in wage setting should be reduced and replaced with direct bargaining between employer and worker representatives. In most countries, this implies encouraging independent trade unions and employers' associations that are representative of the whole spec- trum of firms (including small and newly established private firms). Although labor codes were revised, labor relations tend to be over- regulated, and employment protection remains relatively strict (except in Kazakhstan), but often not enforced (figure 1.12 and table 1.2). This imposes rigidities, leaves little room for collective bargain- ing, and creates incentives for informal employment. As in other countries of the Region, deregulating labor relations and focusing on a set of core standards of worker protection, which can be effectively enforced, would be more efficient. This would more effectively pro- tect basic worker rights while supporting labor market adaptability and reducing incentives for informality. Given the low generosity of unemployment benefits in middle income CIS countries, mandatory severance pay for dismissed work- ers is one way to provide income support and facilitate job search, as well as make firms internalize the costs of dismissal. The mandatory severance pay should be related to job tenure with the terminating employer, with a ceiling so as not to impose excessive and unen- forceable costs on firms.39 At the same time, the procedural costs of dismissal (such as preventive measures, advance notice, required notifications, and negotiations) should be considerably reduced to facilitate enterprise restructuring and productivity growth. The social safety net is underdeveloped in middle income CIS coun- tries, inhibiting enterprise restructuring and locking workers into low- 52 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union productivity jobs. To facilitate labor reallocation, middle income CIS countries need to move from the principle of protecting jobs to that of protecting workers. Income-support schemes need to be strengthened to reduce the social cost of worker displacement. Obviously, such schemes are costly and can create adverse labor-supply incentives. One option that addresses these concerns is workfare or public works. In essence, such schemes provide the unemployed with income, conditional on per- forming some socially useful work. The wage rate under such programs needs to be low to ensure self-selection of the neediest and to avoid com- petition with regular jobs. The participants should be encouraged to look for regular jobs, which may mean that work is less than full-time. Countries that have not already done so should also consider using alternative approaches to targeting social transfers, such as proxy means testing or community identification of beneficiaries within national guidelines. In social insurance, middle income CIS countries should continue reforms of the pension system: this will adjust the system's dependency ratio to changing demographics and encourage other reforms that will stimulate formal sector labor supply to improve the contributions base for a given size of the working-age population. Given the large share of informal employment, a significant pro- portion of people are not covered by social protection; therefore, a move to universalize social protection could be considered. Decou- pling minimum pension schemes and health insurance from the labor market status of people (for example, through creating a universal minimum pension or expanding health insurance coverage to infor- mal workers) are options currently considered for implementation in other countries outside the Region, such as Mexico. As with any of the measures proposed above, fiscal considerations would limit the menu of feasible options. Low Income CIS Countries: Developing Institutions to Support the Formal Sector This group of countries has drifted closer to the situation of low- income developing countries. They have great difficulties in creating jobs in the formal sector and have among the largest shares of infor- mal economy (for example, in Azerbaijan and Georgia, the informal economy is estimated to account for more than 60 percent of GDP).40 This situation means that large parts of the population are essentially unprotected against economic risk and receive no help from the state. One key factor behind the low pace of job creation in the formal sec- tor in the low income CIS countries is underdeveloped institutions of a market economy. This includes lack of enforcement of rules and regu- Overview 53 lations, economic and regulatory policy uncertainty, underdeveloped financial markets, inefficient regulations, bureaucratic harassment and corruption, and (last, but not least) poor and unreliable infrastructure. In particular, the unreliability of public infrastructure is a significant hurdle for firms in the low income CIS countries. It places them at a competitive disadvantage in relation to both their near neighbors in the middle income CIS group and their more-distant competitors in CEE (Steves, Fankhauser, and Rousso 2004). Thus, to meet the challenge of job creation, low income CIS countries need to strengthen the institu- tional underpinnings of a market economy to lessen the risks associated with doing business. They also need to reform their regulatory frame- work to increase its efficiency and remove administrative barriers (for example, customs and trade regulations rank high as a significant impediment to business operation in entrepreneurs' perceptions). An important component of the progrowth strategy is also the develop- ment and proper maintenance of the infrastructure. Employment protection legislation tends to be strict in low income CIS countries, compared with the average for the entire Region (figure 1.12). At the same time, enforcement is lax or selective, meaning that the seem- ingly rigid regulations seldom are a binding constraint on firms. However, flexibility through nonenforcement is not an efficient combination because it undermines the rule of law, increases uncertainty, deprives workers of protection, and is not conducive to long-term productivity growth. Therefore, low income CIS countries, even more than other countries of the Region, need to focus employment legislation on core and enforceable labor standards. As in the middle income CIS countries, mandatory severance pay can be a way of providing workers with income protection (see above), but this option should be coupled with a substan- tial reduction in administrative procedures associated with layoffs. In addition, low income CIS countries need to review the wage determination system and encourage direct bargaining between inde- pendent employers' associations and labor unions. This entails less- ening the role of government and the development of genuine representation of labor and business, representing all segments of the economy (including the new private sector). The restricted administrative capacity, large underreporting of earn- ings, and limited financial resources reduce the efficiency of current social protection programs, which end up protecting a few. Although these constraints limit the range of options available to these countries, a minimum level of protection for some basic risks could be provided to all, financed through general revenues. For example, a move to a flat- rate unemployment benefit system could be considered, financed through general revenues rather than payroll taxes. Moreover, given 54 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union widespread informality, the need to reinforce social protection for those without access to formal insurance mechanisms should be considered. As in the middle income CIS countries, this could include greater use of public works (or workfare) schemes that are open to all those willing to work. However, the program wage rate should be set below the market level to ensure self-targeting of the most needy. In addition, social funds can serve to improve opportunities in poor areas. Finally, given the large share of agriculture, policies should aim at improving employment opportunities in rural areas and raising the productivity of agricultural employment (World Bank 2005b). Invest- ment in infrastructure and in human capital in agricultural regions can encourage the development of agribusiness and the creation of off-farm employment. * * * Productive job opportunities in the countries of Eastern Europe and the Former Soviet Union have to increase substantially to improve the growth prospect in the Region and reduce poverty. Workers who lost their jobs in the declining regions, industries, and firms often find it diffi- cult to find a new, gainful job in the formal sector of the economy because, for most of the transition so far, fewer regions, industries, and firms expanded than declined. As a result, an accumulated pool of unem- ployment has developed especially in CEE and SEE. In the low-income CIS, by contrast, many laid off workers resorted to low-productivity jobs in the informal sector and in agriculture, because unemployment was not an affordable option. To meet the challenge of job creation, the Region needs to have more productive, competitive, and expanding firms. Lowering the costs of doing business and removing key constraints to firm operation and growth are the main routes to more and better jobs. Many of the constraints differ across subgroups of countries in the Region, but there is also a common thread. The investment climate can be improved and job creation spurred by reducing economic and policy uncertainty, easing access to credit, lowering the burden of taxation, and removing administrative barriers to firm entry and growth. But the process of moving from the old jobs to the new ones also needs to be supported. This requires an adaptable labor market where employers have incentives to hire workers--incentives that are not weakened by unduly stringent employment regulations. At the same time, workers should have incentives to look for new jobs, even in other regions of the country or in other occupations. This requires lowering the cost of mobility (for example, by developing the housing and mortgage mar- kets) and improving access to lifelong learning opportunities. Overview 55 ANNEX 1.1 Geopolitical Taxonomy of the Region's Countries Czech Rep. Estonia EU Hungary member Latvia countries Lithuania (EU-8) Poland Slovak Rep. Slovenia CEE Bulgaria EU Croatia accession Romania countries (EU-3) Albania Bosnia & Herzegovina SEE Macedonia, FYR Eastern Serbia & Montenegro Europe and the Former Soviet Union Belarus Middle Kazakhstan income Russian Federation CIS Ukraine Armenia Azerbaijan Low Georgia income Kyrgyz Rep. CIS Moldova Tajikistan Uzbekistan Note: The study focuses on the transition economies of Eastern Europe and the Former Soviet Union. Turkey is also in- cluded in some of the comparative analyses. The transition economies of Eastern Europe and the Former Soviet Union, plus Turkey, are all part of the World Bank's Europe and Central Asia (ECA) Region. Turkmenistan is not included in this Overview due to a lack of reliable country data. 56 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ANNEX 1.2 Summary of Key Differences between Labor Markets in the Middle-Income European and the Low-Income CIS Economies Middle-income European Low-income CIS Labor market characteristics transition economies transition economies Main labor market issue Unemployment Underemployment Adjustment to transition shock Quantity (employment);labor moving Price (wages); labor shifting to expanding services sector back to subsistence agriculture Investment climate Generally favorable Less favorable Market-oriented reforms and enterprise restructuring Advanced Less advanced Employment protection legislation Relatively liberal Strict Enforcement of labor market regulations Relatively strong Weak Unemployment insurance system Developed Largely ineffective Economic structure Dominance of services and industry Dominance of agriculture Informal sector Marginal, driven by tax evasion and Large, driven by lack of job opportunities avoidance of regulations in the formal sector Labor force location Largely urban Largely rural Labor force skills High Mostly low Productivity (wages) High Low Earnings inequality Modest High Source: Bank staff analysis. Endnotes 1. See annex 1.1 for the taxonomy of the countries of this Region. The study focuses on the transition economies of Eastern Europe and the Former Soviet Union. Turkey is also included in some of the comparative analyses. The transition economies of Eastern Europe and the Former Soviet Union, plus Turkey, are all part of the World Bank's Europe and Central Asia (ECA) Region. Turkmenistan in not included in this Overview because of a lack of reliable country data. 2. The Former Soviet Union, except the Baltics. 3. These inequalities largely reflect differences in workers' productivity. However, in some cases wage differentials may be the result of the lack of more productive jobs for workers rather than their low-productivity potential. 4. Unless indicated otherwise, data on the Region's employment and unemployment presented in this report come from the ILO LABORSTA database, which contains national labor force survey (LFS) results. The data may differ from national sources and may not be strictly compara- ble across countries. 5. In March 2000, the EU heads of state and governments agreed on a strat- egy to promote competitiveness and a dynamic, knowledge-driven econ- omy in Europe. The main issues for the realization of the Lisbon agenda for 2010 were the necessary investment in research and development, reduction of red tape to promote entrepreneurship, and achieving an employment rate of 70 per cent (60 percent for women). The actual employment-to-working-age-population ratio in the EU-15 was about 65 percent in 2004. Overview 57 6. In principle, the skills mismatch problem can be addressed by training programs. However, in practice, the potential of training programs to successfully address large skills mismatches is limited for at least two rea- sons. First, if the skill profile of displaced workers is much different from the skill profile of new jobs, successful retraining can be extremely, if not prohibitively, costly. Second, for retraining to be successful, unfilled vacancies must exist for which the training participants can apply. If, however, job openings are few, as is often the case in transition economies, then training is of little help in finding a new job. Existing evaluation of training programs indeed confirms that training is often not very effective in addressing the mass unemployment resulting from large-scale industrial restructuring (Betcherman, Olivas, and Dar 2004). 7. 2002 data. 8. In virtually all modern societies, the female labor force participation rate is significantly less than the male one, which accounts for the higher male-employment-to-population ratio. 9. One reason why productivity gains have been translated into higher wages rather than employment is that they have been largely achieved through defensive restructuring by firms (that is, by eliminating over- staffing and shedding redundant labor). High hiring and firing costs may have contributed to discouraging firms from engaging in strategic restructuring and adopting new technologies and new workers (see the firm restructuring section in this chapter). The substantial real-wage growth, despite high unemployment, reflects a combination of market and institutional forces: employers maximizing profits by using efficiency wages (that is, wages set above the market-clearing level to motivate workers), union pressures, and government wage policies. 10. Rutkowski 1996. 11. Background calculations for World Bank (2001), using data from the firm-based 1998 Survey of Earnings. 12. Earnings dispersion--and thus the incidence of low pay--may be over- estimated because of the practice of underreporting wages to avoid taxes on labor. 13. See Scarpetta and Vodopivec (2005) for the detailed analysis. See Gros- feld and Roland (1996) for a discussion of defensive versus strategic restructuring. 14. It should be stressed that, from a sectoral or macro perspective, defensive and strategic restructuring can be complementary--rather than oppos- ing--strategies. Without a strategic reallocation of resources toward more productive use and the entry and expansion of a job-creating new private sector, it would not be politically feasible to engage in downsiz- ing and defensive restructuring. Correspondingly, without hard budget constraints on state and unrestructured privatized firms, there is too much leakage of resources into the least efficient parts of the economy, crowding out the new private sector. 15. Scarpetta and Vodopivec 2005. 16. Generous social benefit schemes, including early retirement, were used (especially at the early phase of the transition) in CEE countries to facil- itate enterprise restructuring and to move difficult-to-reemploy dis- placed workers (especially the older ones) out of the labor force. 58 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Obviously, these schemes created adverse labor supply effects and con- tributed to low employment rates among older workers in CEE. How- ever, they were deemed necessary to maintain political support for reforms. Moreover, the alternative to outflows to economic inactivity was either higher open unemployment or higher hidden unemploy- ment, as in most of the CIS countries. 17. For example, in the early 2000s, there was only 1 vacancy in Poland for every 200 filled positions, which is four to five times less than in the United States, indicating the weakness of labor demand. At the same time, there were about 50 job seekers per vacancy. For comparison, in the United States, the unemployment-to-vacancies ratio is in the 2 to 3 range. The data on vacancies for Poland come from a special establish- ment-based survey of labor demand and in principle cover all available vacancies, not just those registered with employment offices (GUS 2003). 18. A caveat is necessary: conventional measures of the deficit may not fully capture changes in the fiscal policy stance. Many governments in the Region have reined in quasi-fiscal deficits, reduced arrears, and begun to get a grip on contingent liabilities since the financial crisis in Russia, which does not necessarily show in the fiscal deficit data. 19. It should be stressed that from an analytical standpoint, the association between fiscal loosening, output growth, and job creation is ambiguous. 20. This report uses primarily two sources of information to identify obsta- cles to business operation and growth: the EBRD­World Bank Business Environment and Enterprise Performance Surveys (also known as the investment climate surveys) and the World Bank Doing Business Data- base. For countries, the main route to identifying bottlenecks to firm growth and job creation is regular dialogue with the representatives of employers, including those representing newly established small firms. Additional insights can be gained through focus groups of employers and specialized surveys. 21. A growing body of evidence suggests that high levels of taxes on labor use can have detrimental effects on growth and on employment. This is particularly the case when taxes--in particular, social security contribu- tions to pay for social benefits--are not transferred onto lower wages and thus raise labor costs for firms. If workers fully valued these benefits and were willing to trade them off for lower wages or greater effort, high taxes on labor use might not constrain firms' performance and job cre- ation. The pass-through of taxes on wages, however, depends also on the existence of wage floors (for example, the minimum wage) and wage- bargaining systems. Evidence suggests that in countries with sectoral wage bargaining--as opposed to those with decentralized wage bargain- ing or fully centralized wage bargaining--the pass-through is more lim- ited because workers have less incentive to internalize the effects of their bargaining on overall employment and unemployment. For evidence on the OECD countries, see Daveri and Tabellini (1997); also Elmeskov, Martin, and Scarpetta (1998). For evidence of the impact of the tax wedge on employment and unemployment in the Region during the transition, see Haltiwanger, Scarpetta, and Vodopivec (2004). 22. Employers in the Region often describe these taxes as substantially rais- ing labor costs and having a strong negative impact on their hiring deci- Overview 59 sions. For example, high taxes on labor were identified as the most important barrier to firm growth in a 2003 survey of small and medium- size enterprises in Poland (PKPP 2003). 23. The data come from the World Bank (2004c). 24. EBRD­World Bank Business Environment and Enterprise Performance Survey (BEEPS) data. 25. The conventional view is that low income CIS countries inherited well- developed infrastructure for their level of income. However, the public services infrastructure is unreliable, and the level of maintenance is low, which creates sizable obstacles to business. For example, power outages occur almost two-and-a-half times more frequently in low income CIS countries than in middle income CIS countries and almost 13 times more frequently than in CEE countries. The supply of water and the availabil- ity of mainline phone systems follow a similar pattern (Steves, Fankhauser, and Rousso 2004). 26. See World Bank (2004d), World Development Report 2005, based on data from BEEPS II. 27. Domestic privatization has usually been associated with management and employee buyouts and dispersed ownership. Foreign privatization, in contrast, has been associated with outsider control and more concen- trated ownership. 28. This is not to deny that wages may be above the market-clearing level because of efficiency wages and other reasons, reflecting profit-maxi- mizing behavior by firms. The point is that (except for minimum wages) there is no clear indication of institutionally induced wage rigidity in CEE. 29. "Rent" is extra profit that firms earn if there is little competition. Work- ers can share in part of this rent and raise their wages (appropriate it) if their bargaining power is high. 30. Most countries of the Region have reformed their EPL to reduce employ- ment adjustment costs. Recently, substantial reforms were carried out in Bosnia and Herzegovina, Croatia, Poland, Serbia and Montenegro, and the Slovak Republic. 31. Enforcement is often selective and targeted at larger firms, where there is greater potential for extortion. So, even if compliance with EPL is low in general, some, mainly larger, firms may still be adversely affected by strict EPL. 32. Strong trade unions are an important part of the enforcement mecha- nism, so in countries where they are stronger (mainly in CEE and SEE), employment protection regulations tend to be more binding than in countries where unions are weaker (for example, in the CIS). Also, larger firms, where unions tend to be stronger, more often see EPL as a constraint than do small firms, where unions are weak. 33. For example, in Poland, unemployment benefit coverage was reduced from about 80 percent in the early 1990s to less than 20 percent in the early 2000s, largely by tightening the eligibility criteria (by introducing a requirement of 180 days of covered employment within the last 12 months). Benefit duration, initially indefinite, was reduced to 12 months in 1991. In addition, a flat-rate benefit (36 percent of the average wage) replaced an earnings-related benefit (up to 70 percent of one's previous 60 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union earnings). In Hungary, revisions to unemployment benefit regulations focused on reducing benefit duration: from two years in 1990 to three months to one year (depending on length of employment over the last four years) in 1993 (Dorenbos 1999). 34. For example, a recent investment climate assessment for Poland identi- fied improving the quality of physical infrastructure as one of top priori- ties (World Bank 2004b). 35. OECD (2005) data suggest that foreign direct investment moves away from economies with more-regulated markets (for example, France and Germany) toward economies with less-regulated markets (for example, the United Kingdom and the United States). 36. Naturally, employment protection regulations need to be consistent with the ratified ILO conventions and, for the EU-member countries, relevant EU directives. 37. For example, in Serbia and Montenegro--the largest country in SEE-- starting a business requires going through 11 procedures and takes 51 days. In Albania, the cost of starting a business represents one-third of the gross national income (GNI) per capita. 38. Croatia (which in this report is considered a CEE country, but has shared the problem of inflated unemployment registers with other former Yugoslav republics) provides an example of such an approach. It recently revised its unemployment statistics to distinguish between active job seekers and the passive unemployed. 39. In most OECD countries, the maximum mandatory severance pay does not exceed three months of salary (OECD 1999). 40. Schneider and Klinglmair 2004. CHAPTER 2 Main Labor Market Developments during the Transition Economic transition is associated with profound changes in the labor market. Labor resources are allocated by forces of demand and supply rather than by the decisions of the central planners. This fundamen- tal institutional change, along with other transition shocks, spurs widespread employment and wage adjustments. Unemployment emerges and the wage structure changes. The nature of jobs changes, too. This chapter documents and discusses these changes. It shows that Eastern Europe and the Former Soviet Union (the Region) is an economically diverse region, and therefore its subgroup labor mar- kets are also diverse.1 The chapter looks at the labor market adjust- ment during the transition--the dynamics of employment and unemployment and the changes in the wage level and structure. Then it shows the changing nature of jobs in the wake of the transi- tion. Next the chapter compares initial expectations with actual labor market outcomes. It concludes by summarizing key stylized facts on labor market transition in the Region. An Economically Diverse Region with Differing Labor Markets By its very nature, the transition--from a centrally planned to a mar- ket economy--that started in the Region in the early 1990s implied 61 62 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union changes in the basic institutions of the economy. It decentralized eco- nomic decision-making processes, liberalized price and wage setting, and exposed enterprises to competition. Profitability and competi- tiveness increasingly became the major criteria of firm survival and growth. These major changes in the rules of the game, along with dis- integration of traditional economic links, resulted in a substantial fall in output. This in turn led to a fall in labor demand, which forced firms to reduce employment or wages or, in practice, some combina- tion of the two. Unprofitable firms went out of business or were trans- formed through restructuring and changes in governance and ownership. At the same time, new private firms began to enter the market. This process occurred in virtually all transition economies, although at varying speeds in different countries. Some countries, mainly in Central and Eastern Europe (CEE), implemented key mar- ket-oriented reforms early in the transition and were determined to continue economic restructuring, while others, mainly in Southeast- ern Europe (SEE) and the Commonwealth of Independent States (CIS), adopted a more gradual and less decisive approach. Economies of the Region vary vastly in their levels of economic development, which influences the functioning of the labor market (see box 2.1). At one end of the spectrum are countries that are now members of the European Union; at the other end are the low- income countries of Central Asia. The first group is characterized by developed market institutions, more-advanced economic structures, and, correspondingly, relatively high levels of labor productivity. By contrast, the transition economies of Central Asia are much closer to those in low-income developing countries in their economic infra- structures. Their market-based institutions are relatively less devel- oped or not effective on the ground, economic structures are dominated by agriculture, and their labor productivity tends to be low. Other economies of the Region, such as those of Southeast Europe, Russia, and Ukraine, fall somewhere between these two polar cases. The differences in the levels of economic development across the Region are associated with significant differences in the key features of labor markets. The largely urban, services-oriented, and formal labor markets of EU transition economies perform differently from the mainly rural, agricultural, and informal labor markets of Central Asia. For example, the predominantly rural and informal labor mar- kets in Tajikistan or Uzbekistan are very much different from the mostly urban and formal labor markets of the Czech Republic or Hun- gary. Similarly, the labor markets in Belarus or Ukraine, where mar- ket-oriented reforms are less advanced, differ from the labor markets Main Labor Market Developments during the Transition 63 in the Baltic States, which have a well-functioning market economy and are members of the European Union. Labor market adjustments to the transition shock have been dif- ferent in the transition economies of Europe and those of Central Asia. In Central and Eastern Europe, the high pace of enterprise restructuring and other structural reforms has led to an initial strong fall in employment and far-reaching reallocation of labor away from declining industries toward expanding ones, mainly in the services sector. In contrast, in Central Asia, the pace of enterprise restructur- ing has been slower. The drop in output associated with the transition was accommodated mainly by a fall in real wages. Many workers who were released from the oversized manufacturing sector moved back to subsistence agriculture. And the lack of job opportunities in the formal sector has led to the fast growth of the informal labor mar- ket and self-employment as a coping strategy.2 In European transition economies, the main problem is unem- ployment, which to a large extent results from the fast pace of enter- prise restructuring. In CIS economies, the main problem is underemployment (hidden unemployment) and low productivity (for example, in subsistence agriculture). The low open unemploy- ment in Central Asia reflects both the predominantly agricultural and informal nature of their economies--agriculture and informal sectors act as employers of last resort--and the slow pace of enterprise restructuring (for example, overstaffing is still prevalent in many enterprises). The key differences between labor markets in the middle-income European transition economies and those in the low-income CIS economies are summarized in table 2.1. These differences must be borne in mind while reviewing labor market development in the rest of this report. Unemployment and Underemployment: Major Economic and Social Problems The transition shock caused a substantial fall in the number of for- mal sector jobs in all transition economies of the Region, although it was more marked in CEE and SEE than in the CIS economies. This in turn has led to considerable underutilization of labor. Underuti- lization of labor manifests itself in different forms, depending on policies toward the "old" enterprise sector and the prevailing mode of enterprise adjustment. Aggressive enterprise restructuring, encouraged by imposition of the market discipline that prevailed in 64 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.1 Do Geopolitical Groupings Help in Assessing the Economic Performance of the Transition Countries? Eastern Europe and the Former Soviet Union is a large and economically diversified region. It en- compasses subgroups of countries that differ in their level of economic development, institu- tions, industrial structure, and progress in market-oriented reforms. These differences, in turn, influence the characteristics and evolution of their labor markets. The profound heterogeneity in the economic and structural characteristics of the countries of the Region makes it very difficult to analyze them as a group. One way to group the transition countries is to apply a traditional geopolitical grouping: Central and East European countries, which comprise two groups: CEE EU (EU members) and CEE EUA (EU-accession countries); Southeastern European countries (SEE); and CIS, which comprises middle income CIS countries (Belarus, Kazakhstan, Russia, and Ukraine) and low income CIS countries (Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan).3 This geopolitical grouping also has a close bearing in the levels of economic development, insti- tutional arrangements, and economic structure. For example, a close correspondence exists be- tween the geopolitical location and differences in the level of economic development, as meas- ured by GDP per capita (table A). All CEE EU-member countries belong to the group of the richest countries of the Region. And all low income CIS countries belong to the group of the poorest countries of the Region. Thus, the geopolitical location is a good indicator of the level of economic development. Table A. Geopolitical Location and Level of Economic Development Closely Related Income groups: GDP per capita a), 2002 Geopolitical groups top tercile middle bottom tercile Mean value $ CEE EU 8 12,984 CEE EUA 1 2 8,032 SEE 3 1 5,685 Middle income CIS 3 1 6,122 Low income CIS 7 2,428 Source: World Development Indicators World Bank database; Bank staff calculations. Note: Numbers stand for the number of countries within the cell. a) GDP at PPP (current international $) Similarly, there is a strong overlap between the geopolitical location and the quality of institu- tions and governance. For example, government effectiveness varies significantly between geopolitical groups, but variation within groups is relatively small (table B). Main Labor Market Developments during the Transition 65 Table B. Geopolitical Groups and Government Effectiveness Government Effectiveness Index, 2002 Geopolitical groups top tercile middle bottom tercile mean value CEE EU 8 0.67 CEE EUA 1 2 ­0.07 SEE 3 1 ­0.62 Middle income CIS 1 3 ­0.74 Low income CIS 2 5 ­0.85 Source: Kaufmann, Kraay, and Zoido-Lobaton 2004; Bank staff calculations. Note: Numbers stand for the number of countries within the cell. Finally, geopolitical location allows one to predict fairly well the progress of economic transition. Roughly speaking, the closer a country group is to Brussels, the more advanced it is in imple- menting market reforms (table C). Economic transition is most advanced in CEE countries and least advanced in middle and low income CIS countries, while SEE countries occupy the middle position. Table C. Geopolitical Groups and Economic Transition EBRD index of the progress of the transition, 2001 Geopolitical groups top tercile middle bottom tercile mean value CEE EU 8 3.45 CEE EUA 1 2 3.03 SEE 2 2 2.61 Middle income CIS 2 2 2.36 Low income CIS 3 4 2.53 Source: EBRD Transition Report (2004); Bank staff calculations. Note: Numbers denote the number of countries within the cell. Naturally, there are cases when the use of the geopolitical taxonomy results in a misclassifica- tion of some countries from the economic perspective. For example, Bulgaria and Romania, which are EU-accession countries and thus fall into the CEE group, are closer to the SEE group in economic development. In turn, Croatia, which until 2004 was not officially an EU-accession country, in economic indicators is closer to EU transition economies than to EU-accession coun- tries, let alone SEE countries. Finally, the borderline between middle and low income CIS coun- tries is blurred in some cases. In sum, the Region's economic diversity is well captured by the traditional geopolitical country grouping. This grouping fairly well reflects differences in the levels of economic development, institutional arrangements, and governance and in the progress of market-oriented reforms. 66 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union TABLE 2.1 Differences between Middle-Income European and Low-Income CIS Labor Markets Middle-income European Low-income CIS Labor market characteristics transition economies transition economies Main labor market issue Unemployment Underemployment Adjustment to transition shock Quantity (employment);labor moving to Price (wages); labor shifting back to expanding services sector subsistence agriculture Investment climate Generally favorable Less favorable Market oriented reforms and enterprise restructuring Advanced Less advanced Employment protection legislation Relatively liberal Strict Enforcement of labor market regulations Relatively strong Weak Unemployment insurance system Developed Largely ineffective Economic structure Dominance of services and industry Dominance of agriculture Informal sector Marginal, driven by tax evasion and Large, driven by lack of avoidance of regulations job opportunities in the formal sector Labor force location Largely urban Largely rural Labor force skills High Mostly low Productivity (wages) High Low Earnings inequality Modest High Source: Bank staff analysis. most CEE countries, led to high open unemployment. In contrast, the policy of protection through soft budget constraint, common in most CIS countries, led largely to underemployment (hidden unem- ployment) and associated low productivity (World Bank 2000). Unemployment emerged at an early stage of the transition and rose sharply along with the restructuring process.4 Currently it remains high in most transition economies of CEE and SEE, signifi- cantly above the level characteristic of mature market economies (fig- ure 2.1). But unemployment is also high in Turkey, which is a nontransition economy at a level of economic development similar to that in many countries in the Region.5 Albeit affected by major macro and structural changes over the past two decades and still afflicted by low participation, the country has not gone through the systemic changes of the transition economies (see box 2.2). Unemployment rates in CEE countries are currently well above 10 percent, except in the Czech Republic, Hungary, Romania, and Slove- nia. In SEE countries, unemployment rates are between 10 and 22 percent, except for FYR Macedonia (30 percent). Among the CIS group, Armenia, Georgia, Moldova, Russia, and Ukraine have unem- ployment rates above 6 percent. The other CIS countries are charac- terized by low registered unemployment rates, generally below 4 percent, and with little fluctuation over time. These latter rates, how- ever, should be interpreted with caution because they refer to admin- Main Labor Market Developments during the Transition 67 FIGURE 2.1 Unemployment Is High in Most CEE and SEE Countries, 2003 Kyrgyz Rep.* income CIS Georgia Low Moldova Ukraine income CIS Kazakhstan Russian Fed. Middle Macedonia, FYR SEE Albania Serbia & Montenegro Poland Slovak Rep. Croatia Bulgaria Lithuania CEE Latvia Estonia Czech Rep. Romania Slovenia Hungary EU-15 Turkey OECD average 0 5 10 15 20 25 30 35 40 Unemployment rate in 2003, % Sources: Labor Force Surveys (ILO, and OECD); * 2002 from Labour Market in the CIS countries, Statistical Abstract. Inter- state Statistical Committee of the CIS, Moscow 2004. istrative data that are highly influenced by the availability and gen- erosity of social-benefit provisions.6 Low open unemployment in most CIS countries as a rule masks significant labor market problems. It is associated with artificially sus- tained, low-productivity employment in unprofitable firms, as well as 68 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.2 The Challenge of Job Creation in Turkey Turkey has suffered from poor labor market conditions since structural reforms were introduced in the 1980s. During 1980­2003, employment growth averaged just 0.8 percent, compared with 2.6 percent growth of the working-age population, and the labor force participation rate fell to 48 percent, the lowest in the OECD. Female labor force participation has been exceptionally weak, standing at less than 20 percent in urban areas. Currently, the working-age population is grow- ing at more than 2 percent per year, and by 2020 more than 21 million new workers will have reached working age. Accommodating the needs of these new and existing workers represents a critical challenge for Turkey over the coming decades. Turkey's labor market outcomes reflect the interaction of demographic and economic factors. Like most other developing countries, Turkey is experiencing a rapid demographic transition that has generated a surge of new entrants to the labor market. Although these demographic forces have the potential to promote economic growth, absorbing them in the labor market requires sustained firm dynamism and job creation. Inadequate human and physical capital investment have so far prevented rapid expansion and job creation. Meanwhile, the demographic trends have been accompanied by a transformation of a fundamentally rural-agrarian society into an ur- ban-industrial one. Cultural and institutional factors and an evolving occupational structure have led to changing roles, particularly for women, and have affected household preferences and la- bor force participation decisions. In the economic sphere in the early 1980s, Turkey abruptly abandoned a long-standing and in- ward-looking statist policy framework and began a transition toward a more outward-looking, widespread informal sector and agricultural employment to earn sub- sistence income. Productive job opportunities are few, and underem- ployment is common. See box 2.3 on Moldova and World Bank (2003a) on Russia. In recent years, unemployment has fallen in a few countries, most notably in the Baltic States. But in many other countries, such as Poland and the Slovak Republic, high unemployment rates persist despite resumed economic growth. And in other countries, mainly in those where reforms have been slow and delayed, such as Romania (until recently) and Serbia and Montenegro, unemployment continues to rise. Long-Term Unemployment The most worrisome and socially costly feature of unemployment in transition economies, particularly visible in CEE and SEE, is its long Main Labor Market Developments during the Transition 69 free-market development model. The change quickly turned the economy around from a pro- longed crisis in the late 1970s and produced spectacular results in trade performance. But sub- sequently long-standing policy failures were perpetuated in the new environment, contributing to inflation, macroeconomic instability, and weak investment. At the same time, integration into the global economy, though otherwise beneficial, increased vulnerability to external events, in- cluding two regional wars, economic disruptions in trading partners, and financial contagion. The solution to the problem of generating employment is growth. In the medium term, the out- look is reasonably favorable. The external environment is promising, with growth and import de- mand picking up in European and the Region's trading partners. Turkey rebounded quickly from the 2001 crisis, and the confidence of international investors has been restored, at least for now. The government is implementing an IMF program, and early negotiations on EU admission seem likely, which should bolster reforms and raise Turkey's attractiveness as a destination for foreign direct investment (FDI). Nevertheless, medium-term growth will do little more than keep the labor market in a holding pattern. Making serious inroads into unemployment and low labor force participation will require a substantial improvement in macroeconomic performance. In the longer term, Turkey is well situated in location and factor endowments. Some of the stress of absorbing large numbers of new labor market entrants will be alleviated as the work- ing-age population gets older and its rate of growth slows. A recent overhaul of the educational system aimed at increasing human capital investment promises to increase competitiveness vis-ŕ-vis international rivals. However, increasing the quantity and quality of investment is crucial to exploiting future opportunities and realizing Turkey's potential. Policies to improve the busi- ness climate and achieve macroeconomic stability must be top priorities. Source: Keyfitz 2004. BOX 2.3 Employment in Moldova Moldova is an example of a country with relatively high employment and low unemployment rates, but where a large proportion of jobs is of low productivity and thus probably unsustain- able. As much as 20 percent of privatized firms reported overstaffing in 2001. Moreover, 1 of every 10 workers was on a forced unpaid leave, and the average duration of such leave was three months. Many Moldovan firms close for a part of the year. Because of the low level of ac- tivity in poorly performing firms, the actual hours of work are very low. An average worker in Moldova works only 28 hours per week, while his or her Polish counterpart works 42 hours. Similar problems, although to a lesser degree, occur in Russia (World Bank, 2003f). Source: World Bank 2004a. 70 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union duration. Once unemployed, workers find it difficult to find a new job. Consequently, a large fraction of the unemployed in the Region's transition economies are long-term unemployed (that is, jobless for more than a year). In CEE, 40 to 50 percent of the unemployed have been without a job for more than one year, and this proportion is higher than 60 percent in Bulgaria and the Slovak Republic. These figures are high by OECD standards and similar to those observed in Southern European countries with a stagnant unemployment pool. Available data also point to a high persistence of unemployment in SEE. For example, more than 80 percent of the unemployed in FYR Macedonia have been without a job for more than one year. The crux of the unemployment problem in European transition economies is thus the low chances of finding new employment, rather than high inflow into unemployment. Russia and Turkey pro- vide an interesting contrast of more dynamic unemployment pools. Russian and Turkish workers tend to leave unemployment sooner, and therefore the incidence of long-term unemployment is much lower (about 30 percent). Labor Force Withdrawal The increase in unemployment has been one symptom of the fall in labor demand during the transition. Another closely related symptom is the often large fall in labor force participation. Many workers become discouraged by the lack of job opportunities and give up the job search, ceasing to be part of the labor force. In 2002, the econom- ically active population accounted for around 60 percent of the work- ing-age population in Hungary and FYR Macedonia and for more than 70 percent in Belarus, the Czech Republic, Estonia, Lithuania, and the Slovak Republic. In this context, Bosnia and Herzegovina, Moldova, and Turkey are clear outliers, with official participation rates below 55 percent (in the first two cases, down from more than 60 percent in the mid-1990s). From an economic viewpoint, the effects of labor force withdrawal are similar to those of unemployment: both reduce the amount of available labor and, as such, translate into lower output and diminish growth prospects. For instance, in the early 2000s, the unemploy- ment rate in Poland was much higher than in Hungary. However, so was the labor force participation rate. As a result, both countries had similar employment rates, which is the most relevant measure of the utilization of labor resources.7 Some decline in labor force participation was probably to be expected during transition. Participation rates were particularly high during the Main Labor Market Developments during the Transition 71 central-planning period, when work was not only a right but also a duty for all people of working age. The withdrawal of older workers from the labor market--often taking advantage of early retirement schemes-- imposed significant costs on fragile budgets, but was probably inevitable, given the difficulty of reemploying the workers. In contrast, the with- drawal of women and youth is reducing the growth potential of transi- tion economies and contributing to widening income disparities.8 Greater International Migration Flows Cross-border migration has increased in the Region during the tran- sition. Although in some countries migration was linked to conflicts and ethnic reasons in the early years of the transition, migration has increasingly been driven by economic reasons in most recent years. Pressures from a growing working-age population and the lack of employment opportunities have led nationals from the low income CIS countries toward richer countries of the Region, especially Rus- sia, which have themselves become both significant receiving and sending countries. Many nationals of these countries migrate to West- ern Europe because of increased economic payoffs to migration. In particular, temporary migration (that is, either seasonal or moving back and forth) is sizable (box 2.4). Falling Employment Rates and Persistent Unemployment The level of unemployment depends on the probability of becoming unemployed and on the expected duration of the unemployment period. The latter in turn depends on the probability of escaping unem- ployment by finding a job. Therefore, the level and duration structure of unemployment vary depending on the relationship between the inflow rate and the expected duration of unemployment. In a dynamic labor market, the risk of losing a job may be high, but so are the chances of finding a new one. Therefore, the expected duration of unemployment will be short, which will result in a "mod- erate" unemployment rate (or a low unemployment rate if inflows into unemployment are low). An opposite example is that of a stag- nant labor market. In such a market, the risk of losing a job may be low, but so are the prospects of finding a new one. The expected duration of unemployment periods will be long, and the level of unemploy- ment will be elevated, despite low inflows into joblessness. If inflows into joblessness increase (for example, because of economic down- turn, restructuring, or demographical pressures), unemployment will reach a high level. 72 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.4 International Migration Patterns in the Region The Changing Nature of Migration In the first half of the 1990s, international migration in the Region was a direct consequence of the breakup of the communist system in Eastern Europe and the Soviet Union. Because of the creation of new countries in the former Soviet Union, the former Yugoslavia, and the former Czechoslovakia, millions of people became de facto migrants (that is, foreign born): for example, within the Soviet Union, 43 million individuals were living outside their homeland. Moreover, the end of restrictions to movement led to the resumption of pre-Iron Curtain era migration (for ex- ample Jewish emigration to Israel or ethnic German migration to Germany). In addition, civil wars and conflicts in the area led to further ethnic migration. From the mid-1990s, international migration became increasingly economically driven as indi- viduals weigh the expected costs and benefits of moving to another country to improve their quality of life. Overall, two main directions of flows have emerged: migration from Eastern Eu- rope to Western Europe (especially to Germany and Austria) and migration from low income CIS countries to resource-rich CIS countries (especially to Russia, but also transiting through Azer- baijan and Kazakhstan). The movements from new EU members to old EU members are expected to be limited and temporary. The experience of Southern European countries joining the EU with much lower in- come levels than the then members is illustrative. After an initial increase in migration when join- ing, and as economic and social development occurs, through (for example) foreign direct in- vestments spurred by membership, emigration starts declining to be progressively replaced by temporary or circular (seasonal) migration. All new EU members are currently both significant sending and receiving countries, and a few (the Czech Republic, Hungary, the Slovak Republic, and Slovenia) have already become net immigration countries. The movements from low income CIS countries and among CIS countries have to be viewed in the context of a history of movements within the Former Soviet Union (FSU), making northern FSU countries attractive destinations for those in southern FSU countries. Young and growing populations in the poorly developed Central Asian countries are likely to move north to FSU countries, which are seeing aging and decreasing working-age populations. Belarus and Russia have already become net immigration countries. Lack of progress of reforms, poor governance, and low quality of public services in some CIS countries are strong incentives for out migration. A typical example of a flexible, dynamic labor market is that of the United States. There, 66 percent of the unemployed find a job within a year (Boeri and Terrell 2002). The median duration of unemploy- ment fluctuates around 10 weeks, and correspondingly the incidence Main Labor Market Developments during the Transition 73 Again, as the differentials in governance decrease across countries, migration flows are expect- ed to become multidirectional rather than directed at a few receiving countries, and return mi- gration will become more attractive. Recent Evolution Accurate data on migration are notoriously difficult to obtain, but some tentative observations can be made from available data. Consistent with the changing nature of migration just de- scribed, countries that had above-average migration flows in the early phase of transition saw these subsequently decrease. Net emigration in absolute terms was largest in Lithuania and Poland. Relative to population size, net emigration was largest in Estonia. Hungary reported, both in absolute and in relative terms, the largest net gain from migration, followed by the Czech Republic. Although the number of Central Asian nationals living permanently in Russia has in- creased moderately, there is evidence that temporary migration has increased significantly. International Migration and Labor Markets The receiving country may be affected in two main ways: flows of new migrants may have an impact on wages and employment of prior residents, and immigration may have fiscal implica- tions. The current literature on immigration in the United States, Europe, and Australia suggests only small effects on labor market outcomes of native populations. The fiscal contribution of mi- grants is found to depend on the composition of immigration, highly skilled migrants contribut- ing the most to the fiscal balance. For the sending country, on the one hand, migration may relieve tensions on the labor market (for example, in resource-poor countries that have growing working-age populations, such as Central Asian countries). On the other hand, the emigration of disproportionately young and bet- ter-educated individuals may be an issue because growth may be further depressed by poor la- bor supply. There is evidence that circular or return migration may be welcomed by both sending and re- ceiving countries. Receiving countries may prefer temporary migration that responds to specif- ic labor demand and may be reduced in economic downturns. Sending countries may benefit from greater remittances, as migrants keep stronger ties to their home country with a view to coming back, and from better-skilled returning migrants (although this is not always the case). For example, there is evidence that Albanians who have been abroad earn more than those who never went, while the stayers would have earned even more abroad than those who left. Sources: World Bank forthcoming; Lucas 2005. of long-term unemployment is very low: in 2002, the number of unemployed with a period longer than one year accounted for 8.5 percent of total unemployment in the United States, and usually the percentage is even lower (OECD 2003). Labor markets in CEE are in 74 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union sharp contrast to the U.S. example. As can be expected, flows from employment to unemployment tended to be higher, as transition economies underwent rapid restructuring, which by its very nature is associated with a high rate of job destruction. However, the main dif- ference lies not in the inflow rate, but in the outflow rate, which was much lower in CEE than in the United States. The chance that a worker finds a job within a year after entering unemployment is sub- stantially higher in the United States than in virtually all transition economies. Roughly speaking, the probability that an unemployed person finds a job in transition economies is half that in the United States. Thus, all else being equal, if the outflow rate from unemploy- ment to jobs in Poland, say, were the same as in the United States, then the Polish unemployment rate would be about one-fourth lower (15 percent instead of 20 percent).9 Some examples of the differences in flows into and out of unem- ployment are shown in table 2.2. Data in table 2.2 suggest that high unemployment in transition economies was initially the product of relatively high inflows into unemployment and limited outflows to jobs. Most of the economies were thus in the high unemployment/depressed (stagnant) labor mar- ket state. As the restructuring process slowed with the progress of the transition, the initially high inflow rates gradually declined, although TABLE 2.2 Flows into and out of Unemployment As a proportion of the origin stock Employment to Unemployment Out of the Out of the Employment to Out of the Unemployment to Out of the Labor Force to Labor Force to Country Year Unemployment Labor Force to Employment Labor Force Employment Unemployment Central and Eastern European Countries Bulgaria 1994­1995 0.059 0.092 0.323 0.244 0.092 0.044 Czech Republic 1994­1995 0.013 0.028 0.496 0.129 0.042 0.012 Czech Republic 1996­1997 0.08 0.025 0.457 0.101 0.04 0.008 Czech Republic 1998­1999 0.018 0.025 0.335 0.09 0.036 0.017 Poland 1992­1993 0.04 0.076 0.361 0.158 0.095 0.045 Poland 1993­1994 0.04 0.063 0.354 0.159 0.074 0.043 Slovak Republic 1994­1995 0.023 0.045 0.237 0.078 0.018 0.017 Former Soviet Union Estonia 1992 0.048 0.097 0.465 0.093 0.143 0.036 Estonia 1997 0.047 0.04 0.372 0.064 0.074 0.038 Russian Federation 1992­1993 0.032 0.058 0.52 0.157 0.087 0.014 Russian Federation 1995­1996 0.056 0.062 0.395 0.145 0.076 0.034 Memorandum United States 1992­1993 0.028 0.053 0.659 0.288 0.043 0.161 Source: Boeri and Terrell 2002. Main Labor Market Developments during the Transition 75 they still remain higher than in the United States. This entailed a movement toward the moderate unemployment/stagnant labor mar- ket state because the job-finding chances hardly improved and thus the duration of unemployment remained high. Currently, unemploy- ment is a rather stagnant pool: few workers enter unemployment, but also few are able to exit it. However, recent examples of a substantial fall in unemployment in some CEE countries (for example, Lithuania) suggest that these successful countries might have begun a movement toward a moderate unemployment/dynamic labor market. It is obvi- ously an open question whether the movement will continue toward the low unemployment/dynamic labor market state. Employment Rates below EU Average Because of rising unemployment and falling participation, employ- ment rates (employment as a percentage of working-age population) declined sharply in most transition economies. Roughly 15 percent of the population of working age who would have been employed under the old regime has become nonemployed (that is, unemployed or out of the labor force). In virtually all of the Region's countries, the employment rate is presently below the EU average of 65 percent and way below the Lisbon target of 70 percent for 2010 (figure 2.2). For CEE countries, meeting this target would require a major effort of job creation in the years to come. Employment rates remain relatively high in some low income CIS countries. But rather than a sign of good labor market outcomes, these high employment rates are often a symptom of a slow pace of restruc- turing and the permanence of many low-productivity jobs. These include informal sector jobs, subsistence farming, and nonviable jobs in the formal sector characterized also by a high prevalence of wage arrears, forced unpaid leave, and short working hours. These jobs can exist only as long as enterprises are not subject to competition and receive direct or indirect subsidies (for example, tax arrears, nonpayment of utility bills) (see box 2.5).10 The high employment rates in these countries therefore partly reflect low value added and poorly paid work. In some transition economies of both CEE and the CIS, there are signs of incipient employment growth. Since the early 2000s, the employment rate has increased in the Baltic States, Hungary, and Slovenia in CEE, and in Armenia, Georgia, Russia, and Ukraine in the CIS.11 However, whether this growth is sustainable and represents a turnaround point remains to be seen. In some of these countries, especially in the CIS, major restructuring effort still lies ahead, which may imply an acceleration of job destruction. 76 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 2.2 Employment-to-Population Ratio Is Low in Most of the Region's Countries Georgia income Kyrgyz Rep.* CIS Low Moldova* Kazakhstan income CIS Russian Fed. Middle Ukraine Serbia & Montenegro * SEE Macedonia, FYR Czech Rep. Estonia Slovenia Latvia Lithuania CEE Romania Slovak Rep. Hungary Bulgaria Poland EU-15 Lisbon strategy average target Turkey 0 10 20 30 40 50 60 70 80 Ratio of employment to working-age population, 2003 Source: Labor force surveys (ILO). Note: * signifies 2002. Male Employment Rates Affected More As is common in most industrial economies, male employment rates are higher than female employment rates in all transition economies of the Region. This results from the difference in the labor force participation rates, which are lower for women, rather than from the difference in the Main Labor Market Developments during the Transition 77 unemployment rates, which tend to be similar for men and women in the Region. This is not surprising because the female labor force partici- pation rates are lower than those of men in virtually all societies, and the transition countries are no exception to this regularity. If anything, the gap between male and female labor force partici- pation rates is smaller in the Region than in the OECD countries. In nearly two-thirds of the Region's transition economies, women's activity rates increased relative to men's. In effect, there has been a general increase in the share of women in the labor force over the past decade in the Region (Paci 2002). Thus, by international stan- dards, male rather than female labor resources are underutilized in the Region's transition economies. To illustrate this point for the CEE countries, there is a 6-percentage-point gap in the male labor force participation rates between the EU-15 and CEE. This gap is even big- ger in the male employment rate: 64 percent in the CEE compared with 73 percent in EU-15. In contrast, no such gap between CEE and EU-15 exists for the female employment rate. How can these particularly low male employment rates be explained? One possible explanation is that the restructuring process has been biased against manual, less-skilled labor, which has put men at a disadvantage. The industries that have been declining, especially heavy industry, were employing predominantly male labor. In con- trast, the expanding service sector relies to a much greater extent on female labor. Moreover, the restructuring is associated with the increase in the relative demand for higher, mainly nonmanual, skills, which benefits women more, because in many transition economies women tend to be better educated than men. In other words, there is indication that during the transition the demand for male labor, which tends to be manual and less skilled, has been hurt more than that for female labor, which tends to be nonmanual and more skilled. In addition, the female labor force participation rates were very high by international standards in most socialist countries, and there- fore the gender gap in labor force participation was relatively small. The transition shock lowered both male and female employment rates, but although it has brought female employment rates to the EU level, it brought the male rates well below the EU level (box 2.6). Less-Skilled and Young Workers Unemployment rates in transition economies are particularly high among less-educated workers and are disproportionately low among well-educated workers, especially those with university education. For example, a worker with primary education (eight years) in Poland 78 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.5 Growth and Job Creation in Low Income CIS Countries The low income CIS countries historically suffered from a certain degree of underdevelopment vis-ŕ-vis other Soviet republics, and this development gap was increasing during the 1980s. The output fall unleashed by the early transitional postindependence reforms exacerbated this gap as the low income CIS countries experienced a deeper recession between 1990 and 1994 rela- tive to other transitional economies. In the subsequent stabilization period, the macroeconomic trends became progressively positive, and growth was reestablished in all the countries consid- ered after the Russian financial crisis in 1997. A period of substantial and sustainable GDP growth followed. Yet net job creation continued to be negative or slow (especially in the formal sector) during the expansion period, a trend that constrained the transmission of growth benefits to people. Ac- cording to available evidence, the elasticity of poverty reduction to growth was modest and de- layed by two to four years vis-ŕ-vis the initial phase of GDP growth. Thus questions arise as to what factors may be constraining employment growth in the low income CIS countries and what policy mix might be needed to reverse this trend. A number of coexisting factors have constrained job creation in these countries. They are related to both the structural heritage of the pretransition period and the nature of economic transition in the Region. The countries entered the transition period with significant excess labor in most en- terprises. The recession of the early 1990s accentuated this phenomenon as employment de- clined markedly slower than did GDP. The resulting losses in productivity were fully reflected in real-wage losses. In other words, most people kept their job, but lost part of their income. Growth recovery also took place only in a few sectors, mostly capital-intensive, and in a few ex- port-oriented firms. Moreover, enterprises remained largely unrestructured and could still be characterized by significant labor hoarding. Economic growth was in fact supported by either capital investments or by labor-productivity growth (with a steep increase in real wages). There- fore, the later period of mostly jobless growth is largely explained by the productivity-margin gains that were available in existing enterprises. How real-wage growth could have coexisted with excess labor in the market is partly explained by the dual labor market behavior of the low income CIS economies. Many workers who lost their jobs in manufacturing moved back to agriculture, which for many became the only possi- is more than four times more likely to become unemployed than a worker with tertiary education. At the same time, most of the unem- ployed are workers with secondary education, especially those with narrow vocational skills, because in the centrally planned economies this was the most prevalent form of education. This unemployment Main Labor Market Developments during the Transition 79 ble form of subsistence, as well as an escape from urban unemployment. Consequently, they did not compete with formal urban sector workers and did not exert downward pressure on wages. Industry and agriculture followed two rather different paths during the transition. The agricultur- al sector initially experienced a fall in output--but a fall that was much more limited than what was visible in other sectors. This is explained by land privatization that allowed a large number of small farmers to increase production and productivity and progressively replace state farms in the market. But agriculture remained a rather isolated sector, with few backward and forward linkages to the industrial sector. In the industrial sector, the collapse of the state system was not offset by private initiatives. Moreover, the post-1997 growth in the industrial sector tended to be narrowly based and concentrated in a few--typically capital-intensive--industries, while manu- facturing stagnated. As a result, the demand for agricultural production has hardly increased, constraining agricultural production to the subsistence level. Thus, agriculture and industry not only followed different paths, but paths that produced very little employment or benefits for the average household. Summing up, the following features distinguish economic transition in low income CIS economies from that in middle income CIS and other more-advanced transition economies: · Lower economic performance than middle income CIS countries before the transition · A different economic structure, with a smaller urban sector and a larger agricultural sector · Deeper and faster recession than elsewhere in transition during 1991­1996 · Larger output-employment gap during the recession and faster labor-hoarding growth during the 1990s · Slower and narrower structural reforms during the recession · Faster reforms and faster growth rates during the growth period (1997­2003) · Slow or nil job creation during the growth period (explained by productivity gains in existing enterprises) · An agricultural sector currently confined largely to subsistence · A very small industrial manufacturing sector delinked from the agricultural sector Source: Verme 2004. pattern indicates a labor market premium to education and high skills, also visible in the structure of wages (see Widening Wage Dif- ferentials section later in this chapter). Youth is another group that suffers from high unemployment. On average, youth unemployment rates are twice as high as overall 80 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.6 Relative Position of Women in the Labor Market Has Not Deteriorated during the Transition, and New Employment Opportunities for Women Emerged in the Expanding Services Sector Despite an ideology of gender equality and high labor force participation, different forms of gen- der discrimination were evident under the central-plan period in many of the transition countries. In particular, there was evidence of a gender pay gap, mainly because women were prevalently working in low-paid, white-collar occupations. This wage gap was relatively small by interna- tional standards but, in practice, societies continued to reflect the model of a "male breadwin- ner," which considered women as secondary workers. In the context of major economic changes associated with the transition to a market economy, there was a concern that women would be disproportionately affected by worsening labor mar- ket outcomes so that preexisting gender gaps in employment and wages would increase. The reduction in state-financed social services, such as kindergartens, forced many women to with- draw from the labor market to take on family responsibilities. At the same time, jobs with a clear female connotation remained or even expanded during the transition, while many male-oriented jobs were destroyed, contributing to the erosion of the male breadwinner model. After 15 years of transition, female unemployment rates are generally similar to men's rates and are even lower in some cases. In a number of countries, however, female labor force participa- tion declined more sharply than did men's. Two main concerns persist: unemployment rates. This proportion is not different from that in mature market economies; however, given high overall unemploy- ment rates, joblessness among youth often reaches dramatic levels. For instance, in Bulgaria and the Slovak Republic, more than one- third of the youth labor force is unemployed. Youth unemployment matters because it prevents new entrants in the labor market from gaining experience and developing labor mar- ket skills. Youth unemployment is also closely associated with poverty in some of the transition countries. For example, in Bulgaria and FYR Macedonia, the incidence of poverty among the unemployed youth is significantly higher than the average. In this case, the likely reason is that young people in poor families cannot afford to be out of the labor force. Low family income forces people to enter the labor market at a relatively young age. As long as young people are searching for jobs, they are counted as unemployed, and in this sense poverty gives rise to youth unemployment. Obviously, prolonged and unsuccessful job search by young family members can also contribute to poverty. Main Labor Market Developments during the Transition 81 · Although the gender wage gap remains small, there is evidence of some wage discrimina- tion in the Region. This wage gap can largely be explained in gender differences in the re- turns to some key individual characteristics, primarily education. In fact, the relatively small size of the gender pay gap largely results from the fact that women in the Region tend to be better educated than men, which helps to contain the wage gap in many of the transition countries. · New gender differentials may have been created because of the increase in the size of the informal economy. On the one hand, the informal economy may offer employment opportu- nities to women in a context of a small, formal private sector and shrinking public sector. On the other hand, the informal economy does not provide adequate social protection, and there is some evidence that women are more vulnerable as they concentrate in lower-end jobs. All in all, there is little evidence that the transition has been harder on women than on men, al- though women's labor market status varies greatly across countries. There is, however, some concern that the gender dimension of some developing phenomena--such as the rise of the in- formal sector--may hurt women disproportionately, but this hypothesis has not yet been test- ed. More generally, there is a need to mainstream gender issues into the debate about reforms of the labor market and social policy. This is particularly so given that many employment-protec- tion provisions granted to women during the socialist era (for example, generous maternity- leave provisions with extended rehiring guarantees) may actually hurt women's employment prospects in a competitive labor market by raising the cost of employing women relative to that of employing men. Sources: Paci 2002; Paci and Reilly 2004. Either way, youth unemployment is significantly correlated with poverty (see also Kolev and Saget 2003). Those in Backward Areas Labor market conditions vary substantially across geographical areas of most of the transition countries, and in many cases such geograph- ical disparities have increased over time. Large differences in the unemployment rate within countries illustrate the severity of regional labor market disparities. Differences between the regions with the highest unemployment rate and those with the lowest unemployment rate (usually the capitals) exceeded a factor of 3 in all countries but Romania and Slovenia in 2003.12 These disparities in unemployment are also associated with large disparities in living standards. Regional GDP per capita levels ranged from 60 percent to up to more than 200 percent of the national average in most transition economies (table 2.3). The disparity in regional labor market conditions is much more 82 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union TABLE 2.3 Large Regional Disparities in Unemployment Rates, 2003 Average Minimum Maximum Coefficient of variation Capital city region Czech Republic 7.5 4.2 14.8 40.1 4.2 Hungary 6.3 3.3 11.3 31.1 3.6 Poland 20.1 8.5 33.4 25.8 18.3 Romania 7.0 5.9 8.6 14.8 -- Estonia 10.6 0.4 17.4 32.2 9.0 Latvia 10.4 8.2 15.4 22.7 10.8 Lithuania 12.3 7.5 16.9 20.4 11.7 Slovenia 7.3 4.7 10.2 29.8 4.7 Slovak Republic 17.2 7.1 23.9 36.3 7.1 Russian Fed. (2002) 8.0 1.4 44.0 56.3 1.4 Source: Huber 2004. Note: Registered unemployment rate, NUTS III level. pronounced in Russia, which is not surprising, given the immense economic heterogeneity of the country's regions. In Moscow and St. Petersburg, open unemployment is negligible (less than 4 percent), while in Dagestan and (even more so) in Ingushetia, it is alarmingly high (24 and 44 percent, respectively). Even leaving these extreme cases aside, regional labor market conditions within Russia vary much more than those in European transition economies.13 High regional unemployment is usually coupled with lower labor force participation (resulting from discouragement), giving rise to the low employment/population ratio in high-unemployment regions. Dif- ferences between maximum and minimum employment rates range from more than 10 percentage points to more than 25 percentage points in transition economies; however, there are notable exceptions to this pattern. For example, the high-unemployment (nearly 20 per- cent) southwestern region in Romania has at the same time the high- est employment rate (70 percent). In contrast, the low-unemployment (11 percent) capital region has by far the lowest employment rate (56 percent). This confirms that unemployment and nonparticipation are often different signals of poor labor market conditions and should be considered together. Nonetheless, the prevailing pattern is that of high unemployment being associated with low labor force participation, and thus low employment rates (Bornhorst and Commander 2004). Job Creation Concentrated around Large Urban Agglomerations with Modern Economic Structure Regional unemployment inequality reflects strong regional concen- tration of job creation and employment growth. Job creation and Main Labor Market Developments during the Transition 83 employment opportunities are clustered mainly around large urban agglomerations with diversified industrial structures, particularly with large and expanding service sectors, developed infrastructures, and skilled workforces (figure 2.3). In contrast, employment prospects are dire and unemployment high in monoculture (highly specialized) industrial regions, which suffer from idiosyncratic demand shocks. For example, in Bulgaria and Moldova, employment expanded almost exclusively in the capital region (Rutkowski 2003c and 2004). Employment grew in only 4 out of 21 regions in Croatia and in 3 out of 16 regions in Poland (Rutkowski 2003c; World Bank 2001). Unemployment in rural agricultural regions is usually lower than in regions with industrial monoculture because workers in such regions have recourse to subsistence agriculture and can take tempo- rary and seasonal jobs. In these regions, the problem is underem- ployment rather than open unemployment, a pattern more characteristic of developing countries. Regional labor market disparities are not specific to countries of the Region--they also occur in developed market economies. How- ever, transition economies stand out in two respects. First, most tran- sition countries display a degree of variation in unemployment rates that is generally higher than in most industrial (OECD) countries. For example, the coefficient of variation of unemployment for France and the United States is roughly one-half that for the Czech Republic and one-third that for Hungary (Boeri and Scarpetta 1996; Bornhorst and Commander 2004). These large differences in regional unem- ployment rates are associated with strong concentration of net job creation in a few relatively economically vibrant regions and the dominance of job destruction in most other, economically depressed, regions. Second, labor market imbalances tend to be highly persistent in transition economies.14 Regions showing better performance at the outset have also tended to perform better in later phases. Correlation coefficients of the rank position of regions over time suggest strong persistence in the relative position of regions along different labor market indicators (Huber 2004). In other words, regions at the bot- tom (or at the top) of the regional distribution at the beginning of the transition were still in the same position late in the transition. Again, there are some important exceptions. Particularly in Bulgaria, Roma- nia, and Russia, which may be considered as countries that were slightly slower in their reform process, there were some important changes in the regional distribution of unemployment rates, particu- larly in the early transition, and a similar observation applies to Esto- nia for wage levels. 84 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 2.3 More Workers Are Hired in Regions with a Developed Services Sector, Educated Workforce, and Infrastructure (Poland's Regions, 1997) Panel A 18 16 14 12 e,ployment total 10 of % 8 rate, 6 Hiring 4 2 0 0 10 20 30 40 50 60 70 Employment in services as % of total employment Panel B 18 16 14 12 employment 10 total of 8 % rate, 6 Hiring 4 2 0 0 10 20 30 40 50 60 Population with secondary or higher education, % of total population Main Labor Market Developments during the Transition 85 Panel C 18 16 14 12 employment 10 total of % 8 rate, 6 Hiring 4 2 0 0 10 20 30 40 50 Telephone lines per 100 inhabitants Source: Rutkowski and Przybyla 2002. The strong persistence of regional labor market disparities suggests that they may be of a long-term rather than a transitory nature.15 Even though some of the divergence may be transitory, regions in the long run may become clustered into two distinct groups of prosperity: one covering a relatively small number of well-off regions, the other comprising a large number of relatively poor regions. This points to the weakness of equili- brating mechanisms, such as wage adjustment or interregional labor mobility, to accommodate region-specific shocks (Boeri and Scarpetta 1996; Fidrmuc 2004; Huber 2003; Bornhorst and Commander, 2004). Limited Labor Mobility Interregional migration in most transition economies is low relative to that of the European Union.16 Furthermore, results show that migration rates in transition countries have fallen during the transi- tion, although regional disparities have widened.17 Commuting may be a substitute for migration, but evidence is scarce that this is indeed the case. A study on Hungary points to the high cost of transport as a potential barrier to commuting (Boeri, Burda, and Köllő 1998). Still, the pattern of migration is clear: people move away from declining regions, which offer few job opportunities, to expanding regions, often the capital city and other large urban agglomerations, where employment prospects are better (box 2.7). 86 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 2.7 Internal Migration in the Region in Search of Jobs Internal migration within transition economies generally appears quite low by international stan- dards, and it is now declining from a peak in the early 1990s. Still, migration in the Region re- sponds to differences in labor market conditions: people move away from regions where job op- portunities are scarce toward regions offering better employment prospects--found largely in capital regions and other large urban agglomerations. Internal flow levels increased sharply in a number of countries at the beginning of the transition. The largest of these streams were movements to national commercial and investment capitals from isolated, peripheral communities. For example, more than 1 million people relocated to more centralized areas of Russia from Siberia and the Russian North and Far East. This repre- sents about 12 percent of the population of these areas. Russian census takers discovered near- ly 13,000 "ghost towns" in peripheral regions where cities that had once existed have become fully depopulated and another 35,000 where the population had dwindled down to 10 or fewer people. This mass depopulation has been explained, apart from population aging, by large out- migration to urban areas--mainly to Moscow, where the population increased by 1.5 million from 1990 to 2002. Similar patterns of considerable internal migration from peripheral regions to urban centers are evident in other parts of the Region. For example, although Albania and the Kyrgyz Republic have been net emigration countries since the early 1990s, the populations of Tirana and Bishkek have increased 41 and 23 percent, respectively, over the past decade as internal migration in- creased.18 In Poland, a clear pattern exists of people moving to where jobs are: large net migra- tion flows are associated with hiring rates (figure A). Low and falling migration in the face of large regional disparities in income and unemployment rates in the transition countries presents a puzzle. A number of explanations have been put forward: · High nationwide unemployment rates might have discouraged internal migration because they suggest a low probability of find- ing employment (Decressin 1994). For example, in Bulgaria in the early 2000s, the unemployment rates were high in absolute terms even in relatively low-unemployment regions (for example, 14 percent in the capital city). A similar situation occurred in Poland. · Spatial skills mismatches occurred. Unemployed workers in declin- ing regions (largely less-skilled blue-collar workers) might have been reluctant to migrate to expanding regions because they lacked the skills in demand in these regions (white-collar skills in Main Labor Market Developments during the Transition 87 Figure A Job Opportunities Are a Driving Force behind Migration Flows Hiring Rate and Migration Rate in Poland's Regions, 1998 0.4 0.3 0.2 population 0.1 total of 0.0 % 0 2 4 6 8 10 12 14 16 18 ­0.1 rate, ­0.2 migration ­0.3 Net ­0.4 ­0.5 Hiring rate, % of total employment Source: Rutkowski and Przybyla 2002. Note: Each dot represents one region in Poland in 1997. Yet, the overall trend in the Region points to declines in internal migration since the early 1990s. Much of the spike in migration, especially in CIS countries, appeared to be driven by "diaspora" migration--the return of people to their ethnic homelands--and the movement of workers away from peripheral areas that had been assigned to them or their families by the Soviet central plan- ners. These two effects appear to have run their course, and consequently migration has slowed despite persistent income and quality-of-life differentials. Source: World Bank staff. the services sector). In addition, spatial matching (that is, the process by which unemployed workers find employment in other regions) might have been inefficient (Faini and others 1997). · Social protection mechanisms, being set at national levels, provide relatively generous income support to the unemployed or the inac- tive in regions with low costs of living (see chapter 6 for more details), therefore discouraging migration. · Informal sector income may induce job searchers to stay at home rather than move elsewhere in the country. · Inefficiencies in the housing markets have likely constrained migration. This may be especially relevant in countries where rent controls are important and taxation of housing transactions is high. For example, the cost of renting a studio apartment in Warsaw rep- 88 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union resents around 70 percent of the average monthly net wage of a less-skilled worker. · Liquidity constraints could have prevented potential migrants from moving by making the migration costs unaffordable. · Housing and other in-kind benefits associated with employment have effectively raised the opportunity costs of migration, espe- cially in monoenterprise regions in the CIS (Friebel and Guriev 2000). The evidence presented in existing studies--while delivering a far- from-complete picture--suggests that a combination of liquidity con- straints, housing market imperfections, and in-kind transfers may go some way in explaining the low and falling migration rates.19 Rebounding Real Wages, but Widening Wage Differentials The early stage of the transition was associated with a deep drop in real wages, another manifestation of the fall in output and labor demand. During this initial stage, wages fell more in the CIS economies than in European transition economies, which was an unavoidable conse- quence of the limited employment adjustment to the output drop in the CIS. (In contrast, in European transition economies, the adjust- ment was more through employment reduction.) Significant Wage Rebounds in the Mid-1990s Unlike employment, wages have rebounded relatively quickly. Since the mid-1990s, real wages have grown in all transition economies at a relatively high pace.20 However, despite the growth, in most coun- tries the measured real wages are still below the pretransition level (figure 2.4), but in purchasing power, real wages have probably increased more than wage statistics suggest.21 Notwithstanding prob- lems with real-wage measurement, in some early reformers (that is, the Czech Republic, Hungary, and Poland), wages are already higher than before the transition. The real-wage growth that has occurred during the second stage of the transition is a reflection of concomitant productivity improve- ments (see chapter 4). Productivity gains resulting from enterprise restructuring were translated into higher wages, but at the same time contributed to the low elasticity of employment in relation to output growth.22 Therefore, the beneficiaries of productivity growth have Main Labor Market Developments during the Transition 89 FIGURE 2.4 Real Wages Have Rebounded in the Mid-1990s 250 100) = 200 year 150 (base 100 index 50 growth 0 wage ­50 Real ­100 CEE, Russian Fed. Armenia, excluding & Azerbaijan, Croatia Ukraine Georgia, & Tajikistan 1989­95 1995­2003 1989­2003 Sources: UNICEF, TransMONEE database, Bank staff calculations. been the insiders (that is, workers with jobs) rather than the outsiders (that is, those without salaried employment). Widening Wage Differentials The wage distribution was compressed under central planning, result- ing from the egalitarian ideology and the centralized wage-setting mechanism. Wages were not aligned with productivity, and the returns to human capital were relatively low, which dampened the incentives to invest in skills and education. The change in the wage- setting mechanism, from administrative to market-based (though this change has been less pronounced in the public sector, especially in CIS and SEE), has led to a sharp increase in wage dispersion. More- productive and expanding firms tend to pay higher wages in general, and especially to workers whose skills are in high demand.23 Moreover, technological progress and structural shifts, such as that from the manufacturing sector to the services sector, raised relative demand for, and thus wages of, white-collar, skilled workers. At the same time, the process implied deterioration in the relative wages of less-skilled, blue-collar workers employed in declining manufacturing industries. For example, a university-educated worker in Poland currently earns (on average) 70 percent more than a worker with basic vocational training does. In the private sector, this premium to university education is even 90 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union higher (160 percent).24 Under central planning, the differential was merely 20 percent.25 The changes in the structure of the economy, and thus changes in relative demand for different types of labor, have given rise to various wage premiums to worker and firm characteristics (such as education, skills, occupation, industry, and location) and contributed to the increase in wage inequality. In particular, the increase in returns to education is the single most important observable factor accounting for the increase in wage inequality during the transition (box 2.8). BOX 2.8 An Increase in Educational Wage Premiums Has Been an Important Factor behind the Rise in Wage Inequality The rising wage inequality reflects an increase in wage premiums to various worker and firm characteristics. The increase in wage premiums to education is the most important observable factor that has caused the increase in wage dispersion in virtually all transition economies.a Un- der central planning, wage differentials between highly educated workers and less-educated workers tended to be relatively small. Evidence for CEE and Russia shows that the situation changed dramatically during the first few years of the transition (Fleisher, Peter, and Wang 2004). Wages and salaries of well-educated and highly skilled workers have gone up, while wages of less-educated workers have gone down, not only in relative terms but often also in ab- solute terms. These developments have led to a substantial increase in returns to education, es- pecially to university education. At present, the annualized rate of return to education is similar or higher than that in Western European economies (Orazem and Vodopivec 1995; Vecernik 1995; Rutkowski 1996 and 2001; Kertesi and Köllő 2001; Newell and Reilly 1999; Munich, Svej- nar, and Terrell 2005; Peter 2003). The premiums to high skills and education (especially university education) are particularly high in the private sector. For example, in Poland, while less-educated workers earn the same or less in the private sector, university-educated workers (on average) earn twice as much in the private sector as they do in the public sector (Rutkowski 1998). The pace of the increase in returns to schooling has been positively related to the speed of mar- ket reforms. Returns to education increased faster in countries where regulatory and institu- tional constraints on wage setting were removed earlier. For example, in Ukraine, where the pace of market reforms has been slow, the increase in returns to schooling has been modest, and they are still lower than in the faster-reforming economies of CEE and in Russia (Fleisher, Peter, and Wang 2004; Gorodnichenko and Peter 2004). a. Evidence comes from CEE, Russia, and Ukraine. An earlier study (Newell and Reilly 1996) found that returns to universi- ty education in Uzbekistan in 1995 were significantly lower than in CEE, Russia, and Kazakhstan. Main Labor Market Developments during the Transition 91 Wage inequality is presently high in virtually all transition economies; however, there are substantial differences among the Region's subgroups (figure 2.5). In European transition economies, wage inequality is high by EU standards, but still within the OECD range (although in its upper end). That is, wage disparities are large, FIGURE 2.5 Wage Inequality in the CIS Is Higher than in the CEE Ratios of 9th decile to 1st decile, 2002 Azerbaijan Russian Fed. Kyrgyz Rep. Serbia & Montenegro Armenia Estonia* Romania Ukraine Bulgaria Belarus Lithuania Hungary* Latvia Poland Slovenia OECD level Czech Rep. 0 2 4 6 8 10 12 14 Low income CIS CEE Middle income CIS SEE Sources: UNICEF TransMONEE database; Bank staff calculations. Note: * Signifies 2001. 92 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union but not exorbitant. In contrast, in most of the CIS, wage disparities have reached the extremely high levels characteristic of developing countries. The gap between workers at the bottom of the wage distri- bution and those at the top has become dramatic. As an illustration, in EU transition economies, the top decile worker usually earns four to five times as much as the bottom decile worker. In the Kyrgyz Republic, this ratio amounts to 9, in Russia 11, and in Azerbaijan 13. In contrast, in the EU, the decile ratio varies between 3 and 4. In CEE countries, wage distribution has widened mainly at the upper tail, which means that high wages have become more preva- lent. In the CIS, the wage distribution has widened at both the lower and the upper tails, which implies the emergence of both very low and very high wages. As a result, the wage position of low-paid work- ers has deteriorated in the CIS much more than in CEE. The widening earnings differentials contribute to the growing labor market segmentation in the Region.26 In high earnings­inequality countries, the labor market is divided into the increasingly large high- wage segment and the swelling low-wage segment. The high incidence of low pay is of particular importance, as it often translates into poverty (World Bank 2000). For example, in the Czech Republic or Slovenia (where earnings inequality is relatively low), less than 5 percent of all wage and salary workers earn less than 50 percent of median earnings. In Serbia and Montenegro, this fraction exceeds 20 percent; in Russia, it goes up to 25 percent; and it reaches 30 percent in Azerbaijan.27 In European transition economies, minimum wages are set at a relatively high level, accounting for around 40­50 percent of the average wage. This acts as a floor on the wage distribution and is thus likely to lessen wage variation. In contrast, minimum wages in the CIS are much lower in relative terms. As a rule, they account for less than 20 percent of the average wage, often around 10 percent, which, together with weak enforcement, means that they are unlikely to be binding. This allows firms to maintain low-paid jobs and as such con- tributes to wage dispersion. Wage dispersion also reflects the large disparities in regional labor market conditions discussed above. In theory, wage flexibility can act as an equilibrating mechanism and can be expected to lessen regional unemployment disparities. In practice, however, this mechanism is moderately effective, although in most CEE countries, wages were found to be responsive to regional unemployment (Huber 2004).28 Wage flexibility is limited in many cases because of institutional factors, such as collective bargaining agreements or national minimum wages, which often do not allow for regional differences in labor market con- ditions. In addition, even lower wages in economically underdeveloped Main Labor Market Developments during the Transition 93 regions often do not provide a sufficient incentive for firms to enter and create jobs if the infrastructure is poor and skilled labor scarce. The Changing Nature of Jobs during the Transition Job security was one of the salient features of the socialist economic system that has been undermined during the transition. Under cen- tral planning, the bulk of workers had permanent, full-time jobs in state-owned companies. The transition has been associated with the emergence and expansion of self-employment and informal sector employment and--a related development--a growing incidence of irregular, casual, or temporary jobs. Expanded Informal Sector The size of the informal sector (measured by its share in total output or employment) varies significantly across subgroups within the Region, as well as across countries. Still, the informal sector is an important source of jobs in all subgroups and in most countries (fig- ure 2.6). Its role in the Region is much greater than in OECD coun- tries. The informal sector contributes to GDP nearly twice as much in CEE countries as it does in OECD countries (29 and 17 percent, respectively, in 2000). Informality tends to be smaller--although not negligible--in the more-advanced CEE countries, while it is particularly large in economies of the CIS. In CEE, between one in five and one in four workers is engaged in the informal sector. In other European transition economies, the proportion goes up to one in three. In the CIS, every third to every second worker has an informal sector job. The cross- country variation is substantial. On one end of the spectrum are the Czech and Slovak Republics, where the informal sector accounts for around 15 percent of employment, which is close to the OECD average. The percentage is about twice as high in Bulgaria, Estonia, and Latvia and reaches 40 percent in Belarus, Russia, and Ukraine. In Azerbaijan and Kazakhstan, the informal sector accounts for half of the economy. The characteristics of jobs in the informal economy vary across coun- tries, but also within each country across individuals (box 2.9). Self-Employment Also Expanding Another dimension of the change in the structure of employment during the transition is the increase in self-employment. Self-employ- 94 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 2.6 Informal Sector Accounts for a Substantial Share of Total Employment, Especially in CIS Czech Rep. Slovak Rep. Lithuania Poland Hungary Slovenia Romania Croatia Latvia Bulgaria Uzbekistan Georgia Estonia Kyrgyz Rep. Moldova Macedonia, FYR Armenia Russian Fed. Belarus Ukraine Azerbaijan Kazahstan 0 10 20 30 40 50 60 Employment in the informal sector as % of total employment, 1998/99 Low income CIS CEE Middle income CIS SEE Sources: Schneider 2002; Schneider and Klingmair 2004. BOX 2.9 The Surge in Informality during the Transition: Key Features and Policy Challenges The characteristics of informal activities vary within each transition country and across them; nonetheless, there are some common patterns. In a number of countries in the Region, mainly in CEE, the rise in informal activities is associated with high taxes and strict regulations in prod- uct and labor markets. In other countries, including the low income CIS group, the informal economy has a clear rural connotation and has played the role of an employer of last resort to provide subsistence income. Different motives drive firms and workers into informality. Many small firms choose to stay in- formal because fixed costs--such as licenses or permit fees (and even bribes)--tend to impose a disproportionate burden on them, and they prefer to remain invisible to public officials. Pro- fessional workers may prefer to develop their own activities as self-employment rather than working as employees when taxes on labor use are high and social benefits from affiliation to (continues on the following page) Main Labor Market Developments during the Transition 95 BOX 2.9 (continued) the social security uncertain. At the same time, informal employment may be the only available alternative for less-skilled workers displaced in the wake of enterprise restructuring. Microeconomic evidence from Household Panel Data in four transition economies (Albania, Georgia, Hungary, and Ukraine) sheds some light on the incidence of informality29 and self-em- ployment across different groups of workers and on the mobility in and out of informality and self-employment. The incidence of informality varies substantially across different workers. The youth, poorly ed- ucated, and unskilled blue-collar workers are disproportionately represented among the informal workers. In Ukraine, for example, unskilled workers account for one-third of informal sector em- ployment, compared with less than one-fifth of formal sector employment. In Hungary, the un- skilled account for about 70 percent of informal workers, while they account for only 50 percent of formal sector workers. In pay, informal sector jobs are comparable to formal sector jobs, once due account is taken of individual and firm-specific characteristics. In other words, the observed wage differentials between formal and informal sector jobs are largely the result of selection of workers in the two sectors; low-skilled and less-experienced workers tend to be sorted into in- formal jobs, often in small firms with low-productivity potentials. At the same time, high-skilled workers in self-employment tend to earn more than similar workers in formal sector jobs, espe- cially when taking account is made of the fact that they tend to work longer hours. The analysis of the transition across the different types of jobs in the Region suggests a number of common patterns across countries: · In all countries, informal workers have a higher chance of moving into unemployment than into formal jobs. · At the same time, exit from unemployment is more likely to take place to an informal job than to a formal one. · Mobility into and out of self-employment is more limited than mobility into and out of infor- mality. · Informality and self-employment are also more likely entry points into the labor market than formal employment is. Individual characteristics not only affect the type of job that workers hold but also their chances of moving from one type of job to another. In all four countries, young workers are more mobile than adults and especially older workers. The unskilled are generally more mobile than skilled workers and are more likely to move into informality than skilled workers. Finally, women have less access to self-employment and more likely to be in formal employment than in informal jobs. However, they also have higher chances of moving out of the labor force if they lose their formal job. Source: Duryea and others 2005. 96 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ment plays the most important part in low income CIS countries, where (on average) it accounts for 50 percent of total employment. The share of self-employment is much lower in middle income CIS countries (17 percent) and in CEE (20 percent). By way of compari- son, in EU-15 countries, self-employment accounts for about 15 per- cent of total employment. What are the reasons behind the surge in self-employment? On one hand, people are pushed into self-employment by a lack of work opportunities. In this case, self-employment is the only available means of earning subsistence income. Sometimes employers also force workers into self-employment to lower hiring and firing costs. For example, in Bulgaria and Poland, some health care sector employ- ees were turned into independent self-employed contractors. This obviously inflated the number of self-employed, although there was no corresponding change in the nature of jobs. On the other hand, people can be pulled into self-employment if the expected benefits, mainly earnings, exceed those of wage employment. Self-employ- ment, then, is a voluntary welfare-maximizing choice among differ- ent alternatives. In such a case, self-employment often becomes a springboard for firm creation. In reality, self-employed workers make up both categories--aspir- ing entrepreneurs and those for whom self-employment is a last- resort grasping for survival where regular jobs are scarce and poorly paid. However, there is some evidence that in the Region, the push factors tend to prevail and that most of the self-employed resort to self-employment to avoid unemployment and earn at least subsis- tence income (Earle and Sakova 2000; Verme 2004). The recourse to subsistence agriculture (as has happened in Poland, FYR Macedonia, and Romania and in the most of the CIS) supports this point. Service Sector Jobs Replacing Manufacturing Jobs The reallocation of jobs and labor, which has been an integral part of the transition, entailed an increase in service sector employment and a decline in manufacturing employment. The expansion of the ser- vice sector has been more pronounced in CEE than in the CIS. For example, the share of market services in total employment during the 1990s went up by about 14 percentage points in Estonia, but only by a mere 2 percentage points in Russia (see chapter 4 for more details). Deindustrialization and the expansion of the service sector have been associated with the change in the skill profile of labor demand. The skill content of jobs has shifted upward, especially in the new sector. Jobs destroyed during the transition were disproportionately low-skilled, Main Labor Market Developments during the Transition 97 blue-collar jobs. In contrast, newly created jobs tend to require higher educational attainment and white-collar skills. Thus, job creation in the new sector has been biased against low-skilled workers, largely because of skill-biased technological change. For example, in Estonia, the share of nonmanual workers in total employment increased by 8 percentage points during 1990­2000, so that presently nonmanual workers account for 55 percent of total employment. In Russia, the share of workers with higher education increased by 6 percentage points during 1992­2000, and it currently exceeds 20 percent of the workforce. A similar increase occurred in Poland and other CEE economies (Peter 2003). Clearly, job destruction has fallen mostly on manual workers, who are mainly unskilled (Commander and Köllő 2004). This mismatch between the skill content of jobs that are being destroyed in the old sector and those that are being created in the new sector contributes to structural unem- ployment in transition economies.30 Labor Market Outcomes: Disappointing during the Transition? In most cases, the labor market outcomes that have emerged during the transition were expected and have been an unavoidable conse- quence of the profound institutional changes and restructuring asso- ciated with the transition. Some unemployment, for instance, is a natural phenomenon in a market economy, in which workers con- tinuously move across jobs and the process is not frictionless. In addi- tion, the pace of industrial restructuring, which is a salient characteristic of the transition, is bound to contribute to unemploy- ment by creating structural mismatches. Persistent Unemployment and Growing Labor Market Disparities: Sources of Concern There have been, however, a few unpredicted, negative outcomes. The first is persistently high unemployment and the long duration of unem- ployment periods. High unemployment was expected to be transient. It was assumed that a fast-growing private sector would soon absorb work- ers released from the shrinking public sector, so that workers would experience only relatively short periods of unemployment. This has proved not to be the case. Workers who lost their jobs found it extremely difficult to find new ones because the pace of job creation lagged behind the pace of job destruction for a long time (see Chapter 4). As a result, unemployment has become a stagnant pool. Only in those countries such as the Baltic States, where the rate of job creation has picked up 98 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union and caught up with the rate of job destruction, have outflows from unemployment increased and unemployment started to decline. The second outcome has been the sharp increase in, and high level of, wage inequalities. Again, some increase in wage inequality was expected and was a desirable phenomenon, improving labor market incentives. However, in many countries, wage inequalities have reached extremely high levels.31 This may undermine the sense of fairness and negatively affect worker morale and motivation. More- over, wage inequalities translate into income inequalities and thus can have a negative effect on poverty. The third unanticipated, negative outcome has been the develop- ment of dual labor markets: the often shrinking formal sector offering considerable employment protection, and the growing informal sec- tor promising little job security. This can be mainly traced to high labor taxes and to strict labor regulations. But while for some work- ers, mainly in the poorer CIS countries, the informal sector is the employer of last resort, for others, mainly in richer CEE countries, it is a preferred alternative to work in the formal sector, offering better earning opportunities and more scope for entrepreneurship. The fourth unexpected outcome was the significant growth of regional labor market disparities. This largely reflects strong regional concentration of job creation and high unemployment coinciding with low wages. A few centers of job creation and good employment prospects have emerged, surrounded by depressed regions where few new jobs are being created and employment opportunities are scarce. These negative labor market outcomes--in particular, the high and persistent unemployment along with high wage inequality--have a number of important welfare implications:32 · The underutilization of labor resources translates into lower out- put and therefore lower welfare. · Unemployment and wage disparities translate into even larger income inequalities. · Large income inequalities imply, all else being equal, higher poverty rates. · Long-term unemployment and persistent poverty often lead to social exclusion. · Unemployment, inequality, and eventually poverty carry high eco- nomic, social, and political costs. For all these reasons, improving labor market performance is criti- cal for poverty reduction and the political sustainability of market- oriented reforms. Main Labor Market Developments during the Transition 99 A widespread assumption concerning labor market outcomes during the transition was that they would initially deteriorate because of inten- sive restructuring, but then improve as the reforms started to bear fruit. Put differently, one could hypothesize that changes in labor market outcomes are U-shaped. Is this hypothesis borne out by the evidence? Although the transition has been under way for more than 10 years, it is too early to tell. It has taken longer for labor markets to adjust to the transition shock than initially assumed, and the process of transforma- tion is still far from complete, at least from a microperspective (see below). Therefore, one currently observes a dynamic process whereby key labor market variables, such as employment and unemployment, continue to adjust to various shocks associated with the transition. Pre- sent levels of employment and unemployment are thus unlikely to rep- resent steady-state values. Some countries, where reforms have been slow, are still on the downward slope of the U curve. That is, employ- ment is relatively high and open unemployment low, as for example in Moldova or Ukraine. Other countries are close to the bottom of the U (that is, unemployment is high and employment low, such as in Bul- garia or Poland). But it seems that some countries have already started to move up the right-hand side of the U curve (that is, employment has started to increase and unemployment to fall). Examples include the Baltic States of Estonia, Latvia, and Lithuania, as well as Bulgaria and Hungary. Particularly in the Baltic States, the employment rate, after being well below the EU average, is presently quite close to it. The gap is narrowing. It is noteworthy that the EU average employment rate is significantly below the one that was prevailing under central planning. This suggests that there is probably no return to the extraordinarily high employment rates observed before the transition (at least in the short to medium term). But reaching the current EU average employment rate should not be seen as the ultimate objective. After all, this rate is considered too low by the EU, as illustrated by the Lisbon target of raising the employ- ment rate to 70 percent of the working-age population from the current level of 65 percent. Therefore, even in the most successful transition economies, there is room for improving labor market performance. Summary: Key Stylized Facts on Labor Market Transition in the Region The main features of labor market transition in the Region can be summarized as follows: · Transition has led to a substantial underutilization of labor, negatively affecting social welfare and growth prospects. However, the adjust- 100 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ment to the transition shock has taken different forms in various sub- groups of the Region. In the European transition economies, the job fall has translated into open unemployment. Within this group, there are differences in labor market conditions associated with the pace of enterprise restructuring and the rate of job creation. In the CIS, the fall in labor demand has led largely to underemployment (hidden unemployment). Many workers hold low-productivity jobs in unre- structured and unprofitable enterprises, in the informal sector, and in subsistence agriculture, which is the employer of last resort. · Limited job opportunities have also led to discouragement and massive labor force withdrawal, especially among younger and older cohorts, as well as women. The combined effect of unem- ployment and labor force withdrawal was a substantial fall in the employment-to-population ratio. This fall was stronger in Euro- pean transition economies than in CIS countries. · The low open unemployment and high employment rates that prevail in many CIS countries hide significant problems in their labor markets. They often indicate delayed enterprise restructuring with persistent overstaffing and--especially in low income CIS countries--to the dominance of low-productivity jobs in the infor- mal sector to earn subsistence income. The latter feature is typical of developing countries, where social protection is lacking and so for most workers unemployment is not an affordable option. · Coping strategies of displaced workers differed between European and CIS transition economies, too. In European transition economies, many displaced and discouraged workers took advan- tage of relatively generous nonwork benefits (early retirement and disability pensions, unemployment benefits, and social assistance). In most CIS countries, these benefits are less available and many laid-off workers have moved to subsistence agriculture, self- employment, and casual work in the informal sector. · Although the emergence of unemployment in the wake of the tran- sition was expected, its persistence is a source of major concern. In particular, outflows from unemployment to jobs have been low in many cases, leading to a buildup of a large pool of long-term unem- ployed, with a negative effect on their employment prospects. · Real wages fell sharply during the early phase of the transition, but have rebounded since the mid-1990s, following the resumption of economic growth. The fall in wages was much more pronounced in the CIS economies (where enterprises restructured slowly and Main Labor Market Developments during the Transition 101 were reluctant to lay off workers) than in European economies (where enterprises restructured by shedding redundant labor). The uniform wage growth during the second stage of the transition (when economic growth resumed), coupled with limited, if any, employment growth, implies that output and associated produc- tivity growth are translated into higher wages, rather than into higher employment. · Wage inequalities have increased substantially across the Region; however, the increase has been more moderate in European than in CIS transition economies. As a result, the current level of earn- ings inequality is still relatively modest in CEE and SEE, while it is very high by international standards in most CIS countries. One important driving force behind the growth in wage inequalities has been the increase in returns to education and high (white-collar) skills. However, wage inequalities also reflect firm-specific charac- teristics, such as profitability, ownership, industry affiliation, loca- tion, and so forth. · The segmentation of the labor markets has been an important fea- ture of the transition in the Region's countries. Although the infor- mal sector has become sizable in European transition economies (by European standards), it often accounts for the bulk of the pri- vate sector in some low income CIS countries, resembling devel- oping countries in this respect. In European transition economies, the informal sector is mainly driven by tax evasion, as well as by the avoidance of strict regulations. In contrast, in CIS countries, the informal sector is largely the employer of last resort. The infor- mal sector there is concentrated in agriculture, while in European transition economies, it is concentrated in the expanding services sector. The growth of the informal sector is associated with the increased incidence of casual jobs, as well as with self-employ- ment. · Regional disparities in labor market conditions across the Region are large and persistent. Job creation and employment prospects are heavily concentrated around large urban agglomerations with diversified economic structures and, in particular, an expanding service sector. These growth poles are surrounded by economically depressed regions, often one-company towns, where job opportu- nities are scarce and unemployment is high. Equilibrating forces are too weak to alleviate the imbalances. Labor mobility in the Region is relatively low, resulting from (among other things) an underdeveloped housing market. Although regional wages tend to 102 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union respond to regional unemployment, this is not enough to entice the entry of new firms and investment, which are prerequisites for job creation. · Labor market transition has created both losers and winners. The former include less-skilled blue-collar workers in declining indus- tries and regions, among whom unemployment is high and wages are low and falling in relative terms. The latter constitute well-edu- cated white-collar workers who find employment mainly in the expanding services sector and who command high wages. The dra- matic shift of labor demand away from low-skilled labor--giving rise to high unemployment and the fall in relative wages--is one of the most notable features of labor market transition in the Region. · There are signs of an emerging divide between labor markets in the transition economies of Europe and those of Central Asia. Labor markets in European transition economies in many respects resemble those in developed economies of Europe, in both positive (for example, productivity growth) and negative aspects (for example, high and stagnant unemployment). In contrast, labor markets in low-income CIS countries seem to become similar to those in other low-income countries, with typical characteristics such as the dominant informal sector, underemployment, and low- productivity employment. Notes 1. The study focuses on the transition economies of Eastern Europe and the Former Soviet Union. The transition economies of Eastern Europe and the Former Soviet Union, plus Turkey, are all part of the World Bank's Europe and Central Asia (ECA) Region. Labor market developments in Turkey are discussed in a separate box (box 2.2). Turkmenistan is not covered in this chapter because of lack of reliable data. 2. Self-employment and informal sector employment have acted as shock absorbers in many transition economies, providing a source of earnings and helping the displaced workers to avoid poverty. At the same time, informal sector jobs in the CIS are often of low productivity and casual in nature. 3. See annex 1.1 for the country grouping within the Region. 4. The scope of this chapter is limited to the presentation of basic labor mar- ket trends and patterns in the Region. A comprehensive overview of labor markets during the transition is provided in Svejnar (1999). Useful analysis of labor market outcomes during the transition is also provided in Boeri and Terrell (2002), Riboud and others (2002), and Haltiwanger, Scarpetta, and Vodopivec (2003). Main Labor Market Developments during the Transition 103 5. Turkey's GDP per capita at PPP is close to that of the average for SEE and middle income CIS. 6. This report uses internationally comparable indicators of unemployment based on the labor force surveys. For some countries, however, these data are not available, and the report refers to administrative data on unemployment as collected by labor offices. These latter data are not fully comparable with those of other countries and may also vary over time because of changes in unemployment benefit systems that change the incentives for the unemployed to register with the labor offices. 7. However, employment has started to grow in Hungary since the early 2000s, while it has continued to decline in Poland, so the employment rates in both countries have diverged. 8. Female labor force participation rates continue to be high in most transi- tion economies. Before the transition, they were extraordinarily high by international standards, so their fall may represent a movement toward an equilibrium. The fall in the youth labor force participation, in turn, partly reflects the increased demand for higher education, which has led to a dra- matic increase in college enrollments. For example, in Russia, the number of college students increased more than two times during the transition. 9. Outflows from unemployment to jobs account for about two-thirds of all outflows from unemployment (the other being to inactivity), so dou- bling the outflows-to-jobs rate (holding the outflows-to-inactivity rate constant) results in shortening the average duration of unemployment and thus lowering the unemployment rate by one-fourth. 10. Russia does not fully fit into this pattern because the labor market there is quite vibrant, providing jobs to migrants from other CIS countries. Still, many jobs are in the informal sector and of low productivity. For example, according to Goskomstat (Russian Federation State Statistics Committee) data, more than 40 percent of firms in Russia were report- ing losses in 2002. 11. Measurement of employment and its changes is not straightforward, and various sources, using different definitions, show somewhat different trends. In particular, employment data coming from household-based surveys (for example, labor force surveys and household budget sur- veys) differ from those coming from establishment-based surveys. The former cover self-employment and informal employment in addition to regular employment and thus use a broader definition of employment than the latter, which cover only regular (registered) employment. 12. In the OECD countries, the ratio of the maximum to the minimum regional unemployment rate is around 3. The coefficient of variation in most cases is in the range of 25 to 30 percent (OECD 2000). These values should be regarded as illustrative only because they depend on the num- ber and size of regions. In general, the greater the level of aggregation (that is, the larger the regions), the less is the variation in unemployment rates. 13. Regional variations in labor market conditions in Russia are less pro- nounced and similar to those in CEE if one weighs regional indicators by the labor force. 14. Regional disparities also tend to be persistent in many Western European economies (Italy being the best-known example), but much less so in the United States. A high degree of persistence in transition economies is 104 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union somewhat surprising, given the large-scale reallocation of resources asso- ciated with the transition. 15. Recent econometric evidence by Römisch (2001) for EU-accession can- didate countries and new member states, by Profit (1999) for the Czech Republic, and by Solanko (2003) and Granberg and Zaitseva (2002) for Russia, supports this hypothesis. They find that divergence has been accompanied by an increased polarization of regions. 16. Fidrmuc (2004), comparing internal migration in the Czech Republic, Hungary, Poland, the Slovak Republic, and Slovenia with that in Ger- many, Italy, the Netherlands, and Spain, concludes that migration rates are largely ineffective in reducing regional disparities in the new EU countries. Ederveen and Bardsley (2003) find evidence that, after con- trolling for methodological and data construction differences between studies, migration in the new EU countries is less reactive in particular to differences in unemployment rates. In addition, Andrienko and Guriev (2003) state that overall migration in Russia is low, although Russia is the only country--aside from Hungary--where migration rates approach European levels, but clearly fall short of U.S. levels (EBRD 2003). 17. Fidrmuc (2004) for the big central European candidate countries, Haz- ans (2004) for the Baltic states, Kallai (2003) for Romania, and Andrienko and Guriev (2003) for Russia all find this decline in migration rates to be a stylized fact of the transition period in the countries they analyze. 18. Caution is required in interpreting these finding as these studies may dif- fer in methodology and definitions. The cross-country comparability of migration rates is limited as these depend on the size and degree of homogeneity of regions and these vary across countries. 19. Kallai (2003) and Andrienko and Guriev (2003) provide some evidence on the importance of liquidity constraints in shaping migration in Russia and Romania. Bornhorst and Commander (2004) argue that housing market imperfections are an important aspect. 20. The substantial real-wage growth despite high unemployment reflects a combination of market and institutional forces: employers maximizing profits by using efficiency wages (that is, wages set above the market- clearing level so as to motivate workers), union pressures, and govern- ment wage policies. 21. The purchasing power of wages under central planning was overesti- mated because prices of many consumer goods were set below the equi- librium level, as attested by widespread shortages. Therefore, the fall in real wages during the initial stage of the transition, when prices were lib- eralized, was most probably overestimated. 22. One reason why productivity gains have been translated into higher wages rather than employment is that they have been largely achieved through defensive restructuring by firms--that is, by eliminating over- staffing and shedding redundant labor (see chapter 4). High hiring and firing costs may have contributed to this process (see chapter 6). 23. There is some econometric evidence that in Russia the relationship between a firm's wage rates and its performance (profitability or produc- tivity) is particularly strong--stronger than in CEE countries. This points Main Labor Market Developments during the Transition 105 to a less-competitive labor market in Russia because, under perfect com- petition, wages are independent of firm performance (Gimpelson and Kapelyushnikov 2005). 24. Background calculations for the World Bank (2001), using data from the firm-based 1998 Survey of Earnings. 25. Rutkowski 1996. 26. The contribution of wage inequality to overall income inequality is sub- stantially larger in CEE, where wages account for a high share of total income, than in the CIS, where wages account for a much lower share (World Bank 2000). 27. Earnings dispersion and thus the incidence of low pay may be overesti- mated as a result of the practice of underreporting wages to avoid taxes on labor. 28. The estimation of the so-called wage curve, which is the relationship between individuals' wages and regional unemployment rates (control- ling for human capital and other characteristics), shows that for a sam- ple of CEE countries, wages are at least as responsive to regional unemployment rates as in most developed market economies (Huber 2004). 29. Informal employees are identified as those who do not have a written contract or who are not affiliated with the social security system. 30. In principle, the skill-mismatch problem can be addressed by training programs. However, in practice, the potential of training programs to successfully address large skills mismatches is limited. First, if the skill profile of displaced workers is much different from the skill profile of new jobs, then successful retraining can be extremely, perhaps prohibi- tively, expensive. Second, for retraining to be successful, there need to be unfilled vacancies for which the training participants can apply. If, how- ever, job openings are few, as is often the case in transition economies, then training is of little help in finding a new job. Existing evaluation of training programs indeed confirms that training is not an efficient tool for addressing mass unemployment resulting from large-scale industrial restructuring (Betcherman, Olivas, and Dar 2004). 31. It is an open question whether existing wage differentials reflect a labor market equilibrium or are still of a disequilibrium nature. 32. The term "unemployment" is used in its broad meaning, which covers workers of working age who withdraw from the labor force, as well as different forms of underemployment. CHAPTER 3 Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? This chapter looks at the links between economic growth and employment trends in the countries of the Region during the transi- tion to a market economy and assesses the possible roles of macro- economic policies in shaping these links. Output and employment fell significantly in the early phases of the transition in all countries of the Region. In the subsequent phases of output recovery, employment has been somewhat slow to grow, giving rise to the concern that the transition countries were developing symptoms of a jobless growth, or at least of low-employment-content growth. The macroeconomic performance of transition economies is marked by two major episodes of output loss, which have also affected the links between subsequent output and employment. The first was the widely documented transitional recession; the second was the recession following the 1998 Russian financial crisis. During the first episode, the output loss brought a decrease in employment and an increase in unemployment. Yet, such labor market adjustment was smaller in the CIS countries than in CEE countries, partly because the former group of countries experienced large wage flexibility. This chapter argues that the labor market response to the second episode was different in CEE countries that had progressed the most in the transition. In those countries, the output recovery was not associated with job creation. This chapter assesses whether jobless 107 108 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union growth in those CEE countries can be accounted for by a particular set of macroeconomic conditions. It suggests that the positive link between employment and output growth was potentially still pres- ent, but it was counterbalanced by a combination of high real interest rates and a loose underlying fiscal policy. In the years to come, low real interest rates, possibly linked to euro adoption, and strong emphasis on fiscal discipline may prove beneficial for the job-creation potential of CEE countries. This chapter is organized as follows: Section 1 analyzes the rela- tionship between output and labor market variables during the dif- ferent transition periods. Section 2 discusses the possible channels through which macroeconomic policies might have lessened the responsiveness of employment to economic growth. Section 3 pres- ents the employment outlook in CEE, in view of EU accession, and in the CIS, given the delayed restructuring. The Employment-Output Link during the Different Phases of the Transition The pace and depth of restructuring and the macro- and micropolicy stances largely influence labor market developments in transition economies. The noticeable feature of the transition in most of the countries of the Region is that output recoveries have not been asso- ciated with major rebounds in employment growth. A good starting point is to describe the various contributions of the labor market to per capita growth. To this end, figure 3.1 presents output per capita growth decomposed into labor-productivity growth, changes in the employment rate (employment divided by the working-age popula- tion), and changes in demography (working-age population divided by total population). The Response of Employment to Output Has Changed. Two main phases can be identified, each characterized by a different response of the labor market to output performance. The first phase refers to the transitional recession, while the second phase starts with the 1998­99 financial crisis in Russia and extends to the most recent years.1 Not surprisingly, when output declined in the first phase, employment fell and was only partially compensated by increases in productivity. But even when output growth resumed in the second phase, it was largely driven by productivity growth, with further declines in employment in most countries. Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 109 FIGURE 3.1 Output per Capita Growth Is Largely Driven by Productivity Growth 8 6 4 % 2 rate, 0 Growth­2 ­4 ­6 ­8 1990­ 1997­ 1994­ 1997­ 1992­ 1997­ 1993­ 1997­ 1994­ 1997­ 1994­ 1997­ 1993­ 1997­ 1990­ 1997­ 1990­ 1997­ 1990­ 1997­ 97 2002 97 2002 97 2002 97 2002 97 2002 97 2002 97 2002 97 2002 97 2001 97 2002 Bulgaria Czech Rep. Hungary Poland Romania Slovak Rep. Slovenia Estonia Lithuania Latvia Output per worker (labor productivity) Ratio of employment to working age population Ratio of working population to total population Output per capita growth Source: Bank staff calculations. The (quantity-based) labor market response to output losses in the first phase was larger in CEE than in the low income CIS countries, Russia, and Ukraine. This resulted from the more intense process of economic restructuring in the former countries, which inevitably involved quantity adjustments. But it also resulted from the greater downward rigidity in real wages in CEE countries, compared with that in Russia, Ukraine, and other CIS countries. The labor market response to the second episode was different in the CEE countries. Although the output decline associated with the 1998­99 Russian financial crisis again led to job losses, the subsequent recovery did not generate marked net job creation in CEE countries. What has been the relative importance of macroshocks in shaping key labor market outcomes? How pervasive are jobless growth and high structural unemployment? And what is the relationship between the labor market outcomes and the macroeconomic policy stance? Until 1997, Output and Employment Moved in Tandem in CEE Countries Table 3.1 presents correlations between the key macroeconomic vari- ables (GDP growth and inflation) and labor market outcomes 110 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union (changes in unemployment, employment, and real-wage growth). In the transitional recession period, GDP and employment changes were strongly and positively correlated in CEE, suggesting that there were employment losses during the early recession while employment started to improve as growth gained momentum. A similar picture emerges when one looks at the correlation between changes in out- put and unemployment. In other words, during the initial phase of the transition, employment was responsive to changes in output. In the CIS countries, the elasticity of employment in relation to output was insignificant in the first stage of the transition, while the wage output correlation was strong. This suggests that in the initial period of the transition, price (real-wage) adjustments were more important than quantity (employment) adjustments in the CIS. In Recent Years, Employment Seems to Be Less Responsive to Output The labor market response during the second phase of the transition was different. This phase began with a balance-of-pay- ments crisis in the Czech Republic and the Russian financial crisis and covered the 1998 recession, with the subsequent strong output recov- ery in CEE and in Russia, Ukraine, and the low income CIS countries. However, consistent with a weaker employment output correlation (table 3.1) compared with the first stage of the transition, the subse- quent output recovery was not associated with job creation in a sub- set of CEE countries. Jobless growth was not uniform across CEE countries. There were some obvious exceptions, such as Hungary (which experienced only one year of jobless growth), the Czech Republic (which experienced two episodes of more than 3 percent growth with modest job destruc- tion), and the Slovak Republic. Conversely, countries such as Bulgaria, Lithuania, Poland, and Romania had several years of jobless growth. At the same time in the CIS group, the employment output corre- lation has become stronger in the more recent period, while the wage output correlation has weakened. It is also noticeable that inflation and changes in unemployment were positively correlated during the first phase of the transition. This suggests that inflation was associated with the start of economic restruc- turing, the absorption of monetary overhang, and an overall collapse of the centrally planned economy inherited from the previous system and that such inflation was close to the hyperinflation limit. By the mid- 1990s, inflation had declined everywhere below the 20 percent thresh- old (figure 3.2). Noticeably, the reduction in inflation was accompanied Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 111 TABLE 3.1 Significant Changes in the Correlation between Macroeconomic Variables and Labor Market Indicators (Contemporaneous Correlations between GDP Growth and a Set of Macro Variables) First Episode: 1990­97 Real GDP growth Inflation Real-wage growth Employment growth CEE countries Real GDP growth 1.00 -- Inflation ­0.59 1.00 0.00 -- Real-wage growth 0.60 ­0.37 1.00 0.00 0.01 -- Employment growth 0.46 ­0.20 0.19 1.00 0.00 0.16 0.25 -- Change in unemployment ­0.55 0.10 ­0.29 ­0.60 0.00 0.43 0.03 0.00 CIS countries Real GDP growth 1.00 -- Inflation ­0.28 1.00 0.01 -- Real-wage growth 0.44 ­0.37 1.00 0.00 0.02 -- Employment growth ­0.01 ­0.12 0.19 1.00 0.94 0.41 0.32 -- Change in unemployment 0.05 0.13 ­0.01 ­0.02 0.75 0.43 0.95 0.90 Second Episode: 1998­2003 Real GDP growth Inflation Real-wage growth Employment growth CEE countries Real GDP growth 1.00 -- Inflation ­0.36 1.00 0.00 -- Real-wage growth 0.30 ­0.18 1.00 0.02 0.17 -- Employment growth 0.29 0.03 ­0.01 1.00 0.03 0.80 0.96 -- Change in unemployment ­0.40 ­0.11 ­0.11 ­0.39 0.00 0.38 0.42 0.01 CIS countries Real GDP growth 1.00 -- Inflation ­0.15 1.00 0.19 -- Real-wage growth 0.30 ­0.56 1.00 0.06 0.00 -- Employment growth 0.17 0.08 0.22 1.00 0.22 0.56 0.21 -- Change in unemployment ­0.19 0.01 0.09 ­0.32 0.21 0.98 0.64 0.04 Sources: Bank staff calculations; Boeri and Garibaldi 2004. Note: p-value in italics; -- = not available. 112 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union by a convergence in macroeconomic developments within each group of countries. Differences across groups were still large by the mid-1990s and reflected differences in the initial trajectories: early inflation was much more moderate in CEE than in the CIS countries. Real Wages and the Pace of Restructuring Affected the Output-Job Links Why was the quantity adjustment larger in CEE countries than in the other transition economies? There are different but interrelated explanations for this. When CEE countries started the transition, they had in place effective safety nets, notably relatively generous unem- ployment benefit systems and social assistance of the last resort schemes, which constrained wage adjustment from below.2 This FIGURE 3.2 Inflationary Pressures Have Declined over Time in Most Countries 142 147 inflation inflation GDP GDP ­1 ­1 1990 1992 1995 1998 2000 2002 1990 1992 1995 1998 2000 2002 CEE Low income CIS 1761 114 inflation inflation GDP GDP 5 ­17 1990 1992 1995 1998 2000 2002 1990 1992 1995 1998 2000 2002 Middle income CIS SEE Sources: World Development Indicators; Boeri and Garibaldi 2004. Note: Inflation (GDP deflator) 1990­2003. Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 113 induced more job destruction than in the CIS, where wages ended up bearing most of the costs of adjustment (figure 3.3). Within-group variation in the levels of unemployment in CEE can also be explained in institutions moving the labor force participation margins (see chap- ter 6). In some countries, the decline in employment rates was mainly accompanied by the growth of unemployment; in other countries, it was the rise of inactivity that mainly absorbed employment declines (see chapter 2). Finally, the pace of sectoral reallocation largely affected (and was influenced by) the possibility of using wage adjust- ments instead of labor mobility to accommodate the transition shocks. In countries (mainly CEE--see chapter 4) where sectoral transforma- tion proceeded speedily, labor had to be relocated across firms, indus- tries, and locations, and the price adjustment could not substitute for that. At the same time, the greater willingness of workers to move because of the presence of safety nets probably contributed to speed- ing up the sectoral transformation. The behavior of real wages is also consistent with this interpreta- tion. Indeed, as suggested by figure 3.4, real-wage adjustments in CEE countries were much less marked than in the CIS.3 In the latter group of countries, the cumulative decline in real wages during the transitional recession was on the order of 60­70 percent, compared with 20­25 percent in CEE. Later on, the CIS also experienced a rapid recovery, interrupted only in 1999, while real-wage evolutions in CEE flattened out at one-digit rates. FIGURE 3.3 Employment Adjustment Has Been More Marked in CEE than in CIS Countries LAT 10 4 CRO CRO SLO SLK CRO HUN KAZ LAT GEO LIT BUL ROM SLO RUS SLO LAT 2 POL SLK BUL SLK LAT growth POL HUN EST HUN EST 5 UZB CZE POL POL KYR POL LIT ROM SLO HUN EST SLK RUS SLO CZE SLO KAZ LIT CZE UZB CRO KYR CZE SLO HUN HUN SLK 0 BUL HUN BUL ROM CZE ROM KYR SLO EST SLK LAT AZE UKR ROM LIT ROM RUS MOL KAZ HUN CZE SLK LIT CZE CZE UZB UZB KYR BEL UZB GEO KYR KYR RUS EST POL KYR AZE UKR ARM UKR KYR KYR BEL AZE KYR RUS AZE MOL AZE PO BUL CZE POL SLK UZB AZE BEL AZE KAZ SLO 0 UZB AZE UZB KYR UKR GEO AZE AZE BEL ROM EST LAT AZE KAZ ARM KAZ AZE KAZ BEL Employment­2 HUN HUN LIT ROM CZE BUL KAZ POL BEL KAZ BEL RUS BEL ARM MOL UKR BEL BEL EST LIT EST LAT UZB UKR GEO KAZ ARM LAT SLO KYR RUS AZE GEO CRO SLK POL BEL BEL ARM ARM EST CRO ARM BUL BEL UKR LIT RUS ARM RUS ­4 POL LIT BUL LIT ARM RUS EST ­5 1990 1995 2000 2002 1990 1995 2000 2002 CEE CIS Sources: ILO (LABORSTA database); Boeri and Garibaldi 2004. Note: Employment growth in CEE and low and middle income CIS countries, 1990­2003. 114 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 3.4 Real-Wage Adjustments Have Been More Marked in CIS than in CEE Countries % 20 20 10 10 manufacutring, 0 0 ­10 ­10 growth, ­20 ­20 wage Real 1990 1993 1996 1999 2002 1990 1993 1996 1999 2002 CEE CIS Sources: ILO (LABORSTA database); World Development Indicators; Boeri and Garibaldi 2004. Note: Growth in real manufacturing wage 1990­2003 for CEE and low and middle income CIS countries. Any Role for Macropolicy to Influence the Employment-Output Link? The uncertainties associated with the transition, the major reforms it implied, and the associated economic volatility affected firms' ability to do business (see chapter 5). Such a climate was probably not con- ducive to carrying out further investment or employment creation. In recent years, such policy uncertainties have faded, and although fur- ther reforms are needed, employers are likely to face less economic and policy uncertainty in the coming years, especially in countries that have joined the European Union. The significant weakening of the employment-output link in CEE countries in the second phase of the transition was primarily the result of the large overstaffing that existed in most firms in the central-plan- ning period. This overstaffing largely enabled firms to promote growth without hiring many more workers. (This explanation is developed in more detail in chapter 4.) Changes in monetary and fiscal policy may also have contributed to the observed jobless growth. In particular, this section argues that higher real interest rates and changes in the fiscal stance are possible factors that contributed to jobless growth. Real Interest Rates Have Increased in Recent Years Real interest rates have been positive and sizable since 1995, mainly as a reflection of the tightening in monetary policy necessary to reduce inflationary pressures. Figure 3.5 also highlights a remarkable increase in real interest rates following the 1998­99 recession. Such Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 115 behavior is linked to the financial turmoil that hit the Region in the aftermath of the 1998 Russian crisis. There are various channels through which an increase in real inter- est rates could contribute to jobless growth.4 One important channel is the complementarity between new capital investment and job cre- ation. High real interest rates tend to reduce capital investment, and transition economies certainly need large quantities of capital for com- pleting the restructuring process.5 In particular, the adoption of new technologies and production processes, as well as the development of new activities in the service sectors, requires the hiring of workers. If real interest rates are high, this process of developing new activities can be slowed. Instead, firms are pushed toward defensive restructuring and tend to increase output through more-productive use of existing capital and labor. Especially in transition economies, output and pro- ductivity growth could also be achieved by further rationalizing current production techniques and eliminating the still large labor hoarding. As an illustration, figures 3.6 and 3.7, respectively, show gross fixed capital formation for countries of the Region and of East Asia and a decomposition of gross fixed capital formation for the Czech Republic, Korea, Poland, and the Slovak Republic. Consistent with the idea developed above, figure 3.6 shows that investment has been (on aver- age) lower in the transition countries than in East Asian countries over the 1990s. More important, some evidence exists that productive investment (measured as GFCF minus investment in housing) has been consistently lower in transition countries (figure 3.7). FIGURE 3.5 Real Interest Rates Have Increased in Recent Years in CEE Countries 20 CRO LAT POL 15 SLK CRO SLK LIT SLO POL LAT POL SLK POL SLO LIT SLO CRO POL 10 HUN LIT CRO SLO LAT SLK LAT LAT BUL CRO LAT LAT CRO LAT SLO HUN rate HUN SLO CRO SLK CRO LIT LAT LIT POL SLK PO BUL HUN EST LAT EST CZE SLK BUL SLK LAT SLO CZE CRO SLO SLK 5 BUL HUN HUN CZE SLO BUL SLO CZE LIT POL POL CZE BUL HUN HUN CZE HUN HUN EST CZE POL HUN EST CZE EST EST HUN SLK BUL EST CZE HUN interest POL POL BUL LIT 0 HUN LIT LIT BUL CRO CZE POL BUL Real CZE ­5 EST BUL EST ­10 EST SLK LIT ­15 1990 1993 1996 1999 2002 Sources: World Development Indicators; Boeri and Garibaldi 2004. 116 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 3.6 Share of Gross Fixed Capital Formation as a Percentage of GDP (1990­2003 Average) Korea, Rep. of Malaysia Thailand Slovak Rep. Czech Rep.* Azerbaijan Estonia Indonesia Belarus Bosnia & Herzegovina Taiwan, China* Slovenia Lithuania Kazakhstan Hungary* Ukraine Uzbekistan Latvia Poland Armenia Russian Fed. Romania Croatia* Moldova Albania* Macedonia, FYR Kyrgyz Rep. Bulgaria Tajikistan Georgia* Serbia & Montenegro 0 5 10 15 20 25 30 35 40 % of GDP Countries in the Region Countries outside the Region Source: World Development Indicators 2004. * 1990­2002 Fiscal Policy Has Been Loosened Expansionary fiscal policy is a usual response in times of economic downturns; fiscal expansion is expected to counteract the associated decrease in employment. The fiscal loosening that followed the 1998 recession can be interpreted as such a policy response (figure 3.8). However, in CEE countries, there could have been a connection between such fiscal loosening and jobless growth. Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 117 FIGURE 3.7 Share of GFCF and Productive GFCF as a Percentage of GDP (1990­2003 Average) 40 35 GDP 30 in 25 20 investment of 15 Share 10 5 0 Korea, Rep. of Slovak Rep. Czech Rep. Poland Productive GFCF Other GFCF Source: National accounts of OECD countries (OECD 2003). Note: GFCF = gross fixed capital formation. Changes in the fiscal stance can affect employment if they alter the credibility and expectations of private employers about future changes. Fiscal loosening in CEE countries implies a likely future increase in interest rates and possibly in the tax burden. The anticipa- tion of these adverse effects is akin to a negative capitalization effect in the private sector. In other words, forward-looking entrepreneurs identify fiscal loosening as presaging a future increase in interest rates (and possibly in taxes) and so reduce their hiring upfront. This is par- ticularly likely in CEE countries, which, as mentioned above, have known many years of policy uncertainty, significantly affecting the way enterprises do business, and which have already seen real inter- est rates increase from those of the mid-1990s. A second channel that links fiscal loosening to jobless growth is a crowding-out effect induced by an increase in labor costs. As shown in figure 3.9, real wages in the public sector increased in the past five years at a higher pace than those in the private sector in several CEE countries. In addition to worsening the fiscal balance, high wage growth in the public sector may have negative effects on private sec- tor job creation, crowding out private sector employers from hiring skilled workers and possibly creating a wage-drift effect from the pub- lic to the private sector. How important have these changes in macroeconomic policy set- tings been in driving the jobless growth observed in some CEE coun- tries? To shed light on this question, a simple econometric analysis is 118 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 3.8 Loosening of the Fiscal Stance in CEE Countries in Recent Years LAT BUL BUL HUN HUN GDP BUL HUN BUL of CZE HUN % ROM BUL EST CRO HUN BUL HUN BUL SLK EST BUL HUN 2 HUN SLO EST SLK SLO LAT SLO POL SLO ROM EST BUL BUL ROM BUL POL SLO LAT ROM POL ROM POL SLO SLO ROM EST ROM 0 CZE POL POL HUN SLO SLO LIT SLO CZE EST LAT SLO CRO POL ROM CRO CZE SLO POL ROM surplus, POL EST LIT LAT CZE EST CRO LAT CRO CZE SLK LAT LIT LIT ROM SLK EST SLK LIT ROM LAT LAT ­2 HUN LAT HUN ROM LAT CRO SLK CZE CZE CZE SLK POL HUN POL CRO ROM SLK LAT CRO POL SLK LIT SLK CZE budget POL LAT LIT EST ROM CRO CZE HUN LIT LIT CRO SLK CRO ­5 POL LIT HUN CRO ­7 LIT Primary SLK 1990 1993 1996 1999 2002 CEE Sources: World Economic Outlook Database; Boeri and Garibaldi 2004. performed.6 Its results suggest that GDP growth had a strong effect on employment and unemployment in the second phase of transi- tion. However, during the second phase, the impact of growth on the labor market was mediated by adverse macroeconomic policy changes. The rise in real interest rates and the loosening of the fiscal stance contributed to weakening the potential effect of GDP growth on employment and unemployment. In other words, employment and unemployment would have been more responsive to changes in GDP growth had the macroeconomic environment not changed at the same time (see box 3.1). Summing Up: Employment Prospects in CEE and CIS Countries The Employment Prospect Is Likely to Improve in EU Transition Countries and Accession Countries For the CEE countries that have joined the EU, macroeconomic pol- icy is likely to evolve in the years to come in a favorable way for the labor market. It is likely that interest rate premiums will decline fol- lowing EU accession, reflecting a lower investment risk. Therefore, lower interest rates in the new EU transition economies may promote investment in new technologies and production processes and in job Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 119 BOX 3.1 An Empirical Investigation of the Possible Links between Employment, Output, and Macroeconomic Policy This box provides an attempt to link employment growth to economic growth and a set of poli- cy variables. It is illustrative of the possible linkages between these factors (see Boeri and Garibaldi [2004] for details). The analysis focuses on CEE economies, where there is clearer ev- idence of jobless growth. If Dlneit is employment growth in country i between time t-1 and time t and Dlnyit is GDP growth during the same period, the basic regression is as follows: Dlneit = ai + bDln yit + dpost97t + b1post97tDln yit + g1rit + g ps + eit2 it (3.1) where rit is the real interest rate and psit is the primary surplus. The employment equation 3.1 also features country-fixed effects (ai where the indicator "i" denotes different countries) and a post97t dummy aimed at capturing the effects of employment growth of the second episode. Notice further that the regression includes an interaction term between GDP growth and the post-1997 dummy. That term should capture any change in the employment-output relationship that occurred during the same episode. Finally, eit is a white noise error term. The time period used was 1992­2002 for nine countries. Because of missing observations in some countries and time periods, the total number of observations is 75. The results of the econometric exercise are reported in the table below. The poor employment performance since the 1998 recession seems also associated to the worsening of the fiscal stance, as indicated by the significant effect of the primary budget variable. An important result also concerns the size and statistical significance of the interaction term between GDP growth and the post-1997 dummy. The coefficient b1 clearly indicates that in the employment growth regression, the employment-to-output elasticity did not change significantly over time, but the partial effect of output on employment was undone by the worsening of the fiscal stance. The second column reports the results of the unemployment regression. The results suggest that high real interest rates had adverse effects on the increase in unemployment. Remarkably, the second column of the table suggests that unemployment became more sensitive to GDP growth in the aftermath of 1998, although the coefficient is estimated with a large standard er- ror. According to those estimates, the pickup in growth since the year 2000 should have induced CEE countries to experience a sizable reduction in the unemployment rate. The fact that this re- duction in unemployment did not take place is largely attributed to adverse effects of the inter- est rate hike. (continues on the following page) 120 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 3.1 (continued) Panel Regressions on Employment Growth and Change in Unemployment CEE employment growth change unemployment gdp growth 0.19 ­0.13 0.03 0.01 post97 0.03 0.93 0.97 0.087 gdp growth*post 97 0.04 ­0.17 0.78 0.11 primary surplus 0.22 ­0.77 0.07 0.31 real interest ­0.03 0.03 0.22 0.05 Observations 79 89 R^2 0.19 0.22 FIXED Effect Years 92­02 92­02 Source: Boeri and Garibaldi 2004. Note: P values are in italics. Various robustness tests (not reported) suggest that other macrovariables, notably the current account deficit, the investment-to-GDP ratio, and the amount of foreign direct investment in dol- lars, do not appear to be significant in this multivariate analysis. creation, finally pushing firms beyond the defensive restructuring strategy (see chapter 4). An early adoption of the euro may reinforce this effect. Indeed, real interest rates in the Euro Area are much lower than those experienced by most CEE economies, and adopting a com- mon currency will provide a low interest rate environment.7 There could also be an additional beneficial effect on the labor market stemming from a change in the fiscal stance. For CEE coun- tries that are now members of the EU, there is an obligation to avoid deficits in excess of 3 percent of GDP, whether or not they take part in the Economic and Monetary Union (EMU).8 Although tightening fis- cal policy may reduce resources for needed infrastructure develop- ment, it may also improve expectations of private firms about the ability to contain interest rates, thereby promoting investment and job creation. Similarly, if fiscal prudence leads to wage moderation in Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 121 As an illustration, the figure below shows the expected relationship between output growth and employment growth in CEE countries over the 1998­2002 period had macropolicies remained unchanged (see the corrected correlation line). The figure suggests that employment would have increased more had the macropolicy setting not changed. 6 4 2 growth 0 Employment ­2 ­4 ­6 ­5 0 5 10 15 GDP growth, 1997­2003 Output-employment Output-employment relationship One country year relationship (no correction) with correction for macro policy in the Region Source: Bank staff calculations, using ILO (LABORSTA database) and World Development Indicators. The corrected output-employment relationship is obtained by assuming no change in macro- economic policies over the observed period. the public sector, then this may also have positive spillover effects on wage dynamics in the private sector, stimulating employment. CIS Countries Are Still Lagging Behind in Restructuring, and More Employment Adjustment Is to Be Expected The job prospects in CIS countries largely depend on their progress in structural reforms. Evidence suggests that there is a strong link between output and growth, but the latter has been modest for most of the transition period. Moreover, as discussed below, structural changes, including the reallocation of resources to more-productive activities, have been relatively lacking to date. And even if this mod- est progress has not resulted in very high levels of unemployment or low participation, it has negatively affected growth and the living 122 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 3.9 RealWagesinthePublicSectorHaveIncreasedMoreRapidlythaninthePrivateSectorinCEECountries Hungary Poland 170 135 130 150 125 120 130 115 110 110 105 100 90 95 1995 1996 1997 1998 1999 2000 2001 2002 1995 1996 1997 1998 1999 2000 2001 2002 Bulgaria Czech Republic 200 115 180 160 110 140 120 105 100 100 80 60 95 1996 1997 1998 1999 2000 2001 2002 1995 1996 1997 1998 Slovak Republic Slovenia 135 125 130 120 125 115 120 115 110 110 105 105 100 100 95 95 90 90 1995 1996 1997 1998 1999 2000 2001 2002 1995 1996 1997 1998 1999 2000 2001 2002 Romania Lithuania 140 200 130 180 120 160 110 140 100 120 90 80 100 70 80 1995 1996 1997 1998 1999 2000 2001 2002 1995 1996 1997 1998 1999 2000 2001 2002 Latvia 150 140 130 120 110 100 90 80 1995 1996 1997 1998 1999 2000 2001 2002 All sectors Agriculture Manufacturing Wholesale & retail trade Public administration & defense Public administration Source: LABORSTA database. Macroeconomic Policy, Output, and Employment: Is There Evidence of Jobless Growth? 123 standards of the population. These issues will be addressed in the next chapter of this study. Notes 1. The cutoff point between the first and second transition phases is some- what arbitrary. In the aftermath of the Russian financial crisis, most of the transition countries experienced a significant slowdown in economic activities; but the size of the slowdown differed significantly, being stronger in the CIS countries than in most CEE countries. 2. This hypothesis is proposed by Boeri and Terrell (2002) and Garibaldi and Brixiova (1998). 3. The data refer to manufacturing because of measurement problems in other sectors. 4. Several theoretical and applied papers have also suggested a possible role for real interest rates in influencing unemployment. In particular, Phelps (1992 and 1994), in the "customer market" model of pricing, suggests that an increase in real interest rates lowers incentives to invest in expanding market shares. Thus, the increase in marginal production costs resulting from a rise in interest rates is likely to be followed by a rise in price markups that, in turn, should have negative effects on employ- ment. Moreover, in an intertemporal model, if workers have nonwage income, an increase in the rates of interest may reduce the expected util- ity of being employed. 5. Microdata on labor demand of firms in transitional economies confirm a strong complementarity between capital and labor (Svejnar 1999; Boeri and others 1997). 6. The econometric analysis includes regressions of both employment growth and changes in unemployment rates against a set of variables: GDP growth, the primary surplus, the real interest rate, a post-1997 dummy, and a variable capturing the change in coefficient in the GDP growth variable post-1997. 7. Yet, real interest rates are not under full control of the monetary author- ity, and the real interest rate is partly a market-determined variable. Indeed, some increase in the real interest rate in the aftermath of the Russian financial crisis was probably inevitable, and such country risks would not be fully eliminated by early euro adoption. 8. This obligation is not obviated by the fact that members states not par- ticipating in EMU are not subject to the financial penalties for noncom- pliance. CHAPTER 4 Restructuring, Productivity, and Job Creation Faced with radical institutional and policy changes, enterprises in all transition countries have been forced to adapt their behavior. Some have been able to seize new opportunities, but many others have simply tried to survive in a market to which they were unaccustomed and ill-prepared. The ensuing process of restructuring has involved shifts in technology, suppliers, and customers; abandoning old pro- duction lines and introducing new ones; and hiring and (more often) laying off workers--and has resulted in improved productivity of some firms, but also in the death of many others. The pace and depth of enterprise restructuring and the restructur- ing and reallocation of resources across firms are determined by the extent to which previously nonexistent control mechanisms of enter- prise efficiency have emerged during the transition. A number of key factors are responsible for these efficiency increases, including improvements in corporate governance (for example, changing of managers), brought on by privatization, and stronger competition in product markets, not least because of the opening to foreign markets. The systemic environment of firms (that is, the investment climate) has also improved, with the liberalization of conditions for entry, the reduction of state paternalism via the hardening of the budget con- straint, price liberalization, and macroeconomic stabilization. All these factors have taken different forms in CEE, CIS, and SEE, with 125 126 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union very different implications for output, productivity, and employment growth. How successful have countries been in reallocating resources to more productive uses? Is the process over, or are countries still strug- gling with restructuring their economies? Moreover, is there a role for policy to influence the pace and nature of the process of labor reallocation? And are different approaches to reallocation and restructuring associated with different labor market outcomes? The Required Transformation of the Transition Economies and Progress So Far Economic Structures Were Often Far from Those in Market Economies One way to assess the size of the structural gap that transition economies inherited from the centrally planned period is to compare their employment distribution by sectors with those prevailing in market-oriented economies of similar income per capita.1 This exer- cise offers only an illustration of the potential scope of structural changes expected in the transition economies in their process of con- vergence toward market-based economic systems, not least because the latter vary a great deal across market economies. Bearing in mind these caveats, table 4.1 shows the distribution of employment by sec- tor for transition economies at the beginning of the transition and in 2002.2 In the table, the sectoral deviations from the market bench- mark are also aggregated into a synthetic distortion index, which measures the deviation from the market benchmark (see box 4.1).3 Taken at face value, the distortion index suggests that at the begin- ning of the transition period, CEE countries had an economic struc- ture that was further from the market economy benchmark than that of the CIS countries. Given their level of income per capita, CEE coun- tries already had a relatively small agricultural sector and a largely oversized industrial sector. During the transition, all countries experienced a sideways-U shape change in the employment structure, driven by the evolution of output, with an initial large decline followed by a recovery. By 2002, CEE countries had managed to move closer to a market-based structure of the economy. The European CIS countries have also shown some sign of convergence toward the market benchmark in recent years, but are still very far away from it and would require major adjustments in the years to come. The other CIS countries have Restructuring, Productivity, and Job Creation 127 TABLE 4.1 The Employment Structure in CEE and CIS Countries CEE countries CIS countries Share in employment 1989­90 2002 1989­91 2000 Agriculture Actual 17.8 14.5 30.9 39.0 Benchmark 23.2 22.3 32.1 43.1 Industry Actual 42.0 32.4 30.3 18.3 Benchmark 26.0 26.2 23.8 20.4 Market-oriented services Actual 20.9 29.7 13.6 17.9 Benchmark 28.5 28.9 24.3 19.4 Nonmarket services Actual 18.8 23.3 23.3 23.0 Benchmark 22.4 22.7 19.8 17.0 GDP per capita (1995 US$) 7,268 7,627 4,603 2,706 Distortion index 17.0% 12.1% 13.7% 13.0% Source: Bank staff calculations. Note: The distortion index is defined as half the sum of the absolute value of (s ­ s*) where s is the actual share of employment in a sector and s* is the benchmark share. The benchmark shares are calculated as described in the text. seen increases in agriculture and little progress in the development of a modern service sector. They have moved closer to the market benchmark by becoming poorer rather than by progressing toward modern economies (see figure 4.1 for the example of the Kyrgyz Republic; see Scarpetta and Vodopivec [2005] for the other coun- tries). It could be argued that with the removal of subsidies from the Former Soviet Union, these countries have reverted to the income levels that are more consistent with their comparative advantages and underlying economic characteristics. Shifts to Market Services in CEE Looking at changes in the sectoral composition of employment sug- gests that all transition countries experienced a process of deindustri- alization, but this was associated with significantly different patterns across them. In most CEE countries, except for Romania, deindustri- alization was associated with a reduction in employment in agricul- ture. In 2002, the share of agriculture in total employment in most CEE countries was close to what is seen in many EU countries. By contrast, the CIS countries--and, within the group, the poorest coun- tries in particular--saw a rise in relative and even absolute employ- ment in agriculture (see figure 4.1 and Scarpetta and Vodopivec 128 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 4.1 Economic Development and the Employment Structure Using large cross-country data sets, a number of stylized development patterns have been iden- tified (Chenery and Taylor 1968; Rowthorn and Ramaswamy 1997). The share of agriculture in GDP and employment tends to fall as economies grow richer. The share of industry in GDP and employment tends to increase with income per capita, but the relation is nonlinear (that is, at higher levels of income per capita, employment in industry shifts toward services). Finally, the share of services and, in particular, market-oriented services rises monotonically with income per capita. Raiser, Schaffer, and Schuchhardt (2003) propose a market-economy benchmark based on regression analyses on a cross-section of 50 countries. For each sector, the share in employment is regressed on the log of GDP per capita and its square. (The results are shown in the table below.) Because all sectoral regressions fit the data quite well, this section uses the fit- ted curves as benchmarks in the analysis of the distortion in economic structures and the evo- lution of these structures over the transition. It should be stressed that similar conclusions con- cerning the distance from the market-based economic structures can be obtained using real benchmarks, such as the EU average or the North EU and the South EU benchmarks (see Aigin- ger and others 1999 and Landesmann 2000). More important, benchmark comparisons are generally performed to compare underlying fea- tures of economies at their steady state. Transition economies are in fact evolving rapidly, and the observed employment structures--even in most recent years--should not be taken to rep- resent their new steady state equilibrium. Benchmarking Regressions (standard errors in parentheses) Dependent variable: share of employment in Independent variables Agriculture Industry Market service Nonmarket services Log GDP per capita ­0.38390 0.32560 0.08198 ­0.03740 (0.036) (0.019) (0.576) (0.808) (Log GDP per capita)2 0.01082 ­0.01600 0.00058 0.00548 (0.293) (0.042) (0.945) (0.533) Constant 2.78970 ­1.37035 ­0.49013 0.12316 (0.001) (0.023) (0.441) (0.853) R2 0.883 0.4544 0.7141 0.4784 F(2, 47) 176.4 19.57 58.71 21.55 Number of observations 50 50 50 50 Source: Raiser, Schaffer, and Schuchhardt 2003. [2005]), often as the employer of last resort, given the lack of job openings in other sectors. Further downsizing in industry could be expected in CEE countries if they are to move toward a market economy industrial structure. Moreover, despite some significant shifts in employment toward mar- Restructuring, Productivity, and Job Creation 129 FIGURE 4.1 Different Patterns of Labor Reallocation across Transition Economies The structure of employment in the Czech Republic Evolution over the transition and with respect to benchmark Agriculture Industry 20 45 . 1989 Benchmark 1990 1992 . . 1996 40 . 15 .. 2000 . 2002 1990 1989 35 employment . employment 10 1992 30 total Benchmark . total . of . 1996 2000 2002 of 5 . .. % . % 25 9000 9500 10000 10500 11000 11500 9000 9500 10000 10500 11000 11500 GDP per capita/PPP terms GDP per capita/PPP terms Market Services Nonmarket Services 26 32 Benchmark . .. 2002 2000 . Benchmark 1996 24 . 30 . 2002 . 2000 . 22 employment 28 . . employment 1992 . . 1996 . . total 26 1992 total 20 . 1990 of of 1989 % 24 . 1989 18 1990 % 9000 9500 10000 10500 11000 11500 9000 9500 10000 10500 11000 11500 GDP per capita/PPP terms GDP per capita/PPP terms The structure of employment in the Kyrgyz Republic Evolution over the transition and with respect to benchmark Agriculture Industry 55 30 . 2000 1990 50 25 . 1989 . . . 1996 45 Benchmark 1992 . 20 Benchmark employment . employment . 40 . . total 1992 total 15 1996 35 . of 1990 of 1989 . . % % . 30 10 2000 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 GDP per capita/PPP terms GDP per capita/PPP terms Market Services Nonmarket Services 25 26 . 1992 1990 24 . 1989 Benchmark . 20 22 . 1996 employment . . employment 1996 20 . .. 2000 15 . 2000 total . . total 18 of . 1992 . 1990 of Benchmark % % 16 10 1989 2000 2500 3000 3500 4000 2000 2500 3000 3500 4000 GDP per capita/PPP terms GDP per capita/PPP terms (continues on the following page) 130 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 4.1 (continued) The structure of employment in Romania Evolution over the transition and with respect to benchmark Agriculture Industry 45 45 2000 1989 1990 40 40 35 1992 employment 1996 employment 2002 1996 30 2002 35 1992 total Benchmark total 2000 of of 25 % % Benchmark 30 1990 1989 20 3500 4000 4500 5000 3500 4000 4500 5000 GDP per capita/PPP terms GDP per capita/PPP terms Market services Nonmarket services 25 20 Benchmark Benchmark 18 20 16 2002 employment employment 1992 2002 2000 total . 14 1996 2000 1996 total of of % 15 1990 % 12 1992 1990 1989 1989 3500 4000 4500 5000 3500 4000 4500 5000 GDP per capita/PPP terms GDP per capita/PPP terms Source: Authors' calculations. ket services (trade, finance, transport, and communication), they still have only 30 percent of employment in these activities, and Romania has an even smaller market service sector. At the same time, most CEE countries have shares in employment in nonmarket services (public administration and social services) that are greater than what would be expected in market economies at a similar level of develop- ment, with significant implications for public finances. Backward Shifts to Agriculture in Many CIS Countries For the middle income CIS countries, there are signs of positive con- vergence (that is, associated with income growth) in the most recent years. The size of the required employment reallocation, however, remains daunting. Major shifts can be expected away from agricul- ture and possibly industry, especially toward market services. This will imply major changes in skill requirements and in work habits, as well as changes in the location of workers away from declining regions and toward expanding areas. Restructuring, Productivity, and Job Creation 131 The poorer CIS countries, by restructuring the least, are paradoxi- cally quite close to the market-based benchmark. The lack of reforms has been associated with major declines in income per capita, which has in turn brought them closer to low-income market economy benchmarks. Sustained economic growth in these countries is likely to trigger major pressures to reallocate labor along the lines of their Euro- pean counterparts. But there are greater uncertainties as to whether the poorer CIS countries have the economic and institutional capabil- ities to pursue an aggressive transformation of their economies. What Is the Role of Firm Restructuring and the Entry and Exit of Firms for Job Creation? The analysis of shifts in employment across sectors hides a greater dynamism taking place within sectors, because of new firms entering the market and displacing obsolete units and because of within-firm restructuring in the struggle to survive and possibly expand market share. In this context, it is of importance to first assess the size and characteristics of job flows (job creation and destruction) and then link them to productivity and output growth. Job Creation and Destruction Have Surged Rapidly Not surprisingly, job reallocation (the sum of job creation and destruction) increased dramatically in response to transition: from less than 10 percent of the workforce in the late 1980s to more than 20­25 percent in the 1990s (see box 4.2). Job creation and job destruction rates in transition countries exceededthoseindevelopedcountriesorwereataboutthesamelevel,and they lagged slightly behind the average job flows in the sample of develop- ing countries (figure 4.2). Once the initial years of transition are excluded, the average rates of job reallocation and excess job reallocation for the tran- sition countries were 25 and 21 percent, respectively, thus exceeding the average of developed countries with 24 and 20 percent, respectively, but falling behind developing countries. It is quite remarkable that it took only a few years for the job flows in transition countries to increase from very low levels to the levels of flow present in mature market economies. In Some Countries, Job Creation Has Lagged behind Job Destruction The timing of job creation and destruction has played an essential role in shaping labor market outcomes. In all transition economies, job 132 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 4.2 A Consistent International Firm-Level Database Available data at the firm level are usually compiled for fiscal and other purposes, and (unlike macroeconomic data) there are few internationally agreed-on definitions and sources, although harmonization has improved over the years. The data used in this chapter are based on a har- monized firm-level database for 24 OECD, transition, and emerging economies. The data set used in the study was collected in various stages. Most recently, the firm-level project organized by the World Bank collected indicators for 14 countries (Estonia, Hungary, Latvia, Romania, and Slovenia; Argentina, Brazil, Chile, Colombia, Mexico, and Venezuela; Indonesia, Republic of Ko- rea, and Taiwan [China]). An earlier OECD study collected indicators based on information on firms from Canada, Denmark, Germany, Finland, France, Italy, the Netherlands, Portugal, the United Kingdom, and the United States (see Bartelsman, Haltiwanger, and Scarpetta [2004] for details). The main source of the data and the period covered for the five transition economies in- cluded in the sample are presented in the table below. The analysis of firm demographics is based on business registers, censuses, social security databases, or employment-based regis- ters containing information on both establishments and firms. Data for the analysis of produc- tivity growth come more frequently from business surveys. Using these data, time-series indi- cators on firm demographics were generated for disaggregated sectors for each country. Basic Characteristics of the World Bank Firm-Level Database Firm demographics and survival Labor productivity Country Source Period Threshold Source Period Threshold Sectors Estonia Business Register 95­01 Emp 1 Business Register 95­00 to 96­01 Emp 1 All Hungary Fiscal register (APEH) 92­01 Emp 1 Fiscal register (APEH) 92­96 to 97­01 Emp > 1 All Latvia LURSOFT and State LURSOFT and State Social Insurance Social Insurance Agency 96­02 Emp 1 Agency 96­01to 97­02 Emp 1 All Romania Business register 92­01 Emp 1 Business register 95­98 to 96­99 Emp 1 All Slovenia Statistical Office of Statistical Office of Slovenia, Accounting Slovenia, Accounting data. Business data. Business registry data 92­01 Emp 1 registry data 92­97 to 97­01 Emp > 1 All Source: Bartelsman, Haltiwanger, and Scarpetta 2004. destruction surged first in the early 1990s, but the response of job cre- ation differed across countries. Many CEE countries, which advanced rapidly in the transformation of their economies, saw job creation rap- idly catching up with job destruction (see, for example, Estonia and Slovenia in figure 4.3), giving rise to balanced (synchronized) job flows. In other countries lagging behind in economic transformation, job creation continued to be lower than job destruction for prolonged Restructuring, Productivity, and Job Creation 133 Definition of Key Concepts Entry rate is defined as the number of new firms divided by the total number of incumbent and entrant firms in a given year. Exit rate is defined as the number of firms exiting the market in a given year divided by the pop- ulation of origin (that is, the incumbents in the previous year). Labor productivity growth is defined as the difference between the rate of growth of output and that of employmenta and, whenever possible, controls for material inputs. Job creation rate equals employment gains summed over all plants that expand in a given year, divided by the average employment in the period. Job destruction rate equals employment losses summed over all plants that contract in a given year, divided by the average employment in the period. Job reallocation rate is the sum of all plant-level employment gains and losses that occur in a giv- en year. Comparability Issues Two prominent aspects of the data have to be borne in mind while comparing firm-level data across countries: Unit of observation. The data used in this study refer to "firms" rather than "`establishments." Firm-based data are likely to more closely represent entities that are responsible for key aspects of decision making, compared with plant-level data. Nevertheless, business registers may de- fine firms at different points in ownership structures; for example, some registers consider firms that are effectively controlled by a "parent" firm as separate units, while others record only the parent company. Size threshold. Although some registers include even single-person businesses, others omit firms smaller than a certain size, usually in the number of employees, but sometimes in other measures (such as sales). Data used in this study exclude single-person businesses. Source: Bartelsman, Haltiwanger, and Scarpetta 2004. a. Available data do not allow the control for changes in hours worked, nor do they distinguish between part- and full-time employment. periods (see, for example, Russia and Romania in figure 4.3), giving rise to either large increases in unemployment or falls in participation. The main differences across countries are in job creation rather than job destruction. Although countries lagging behind on reforms were able to contain job destruction in the early phases of transition, job destruction later took off to high levels. At the same time, how- ever, delaying macro- and microreforms prevented new private initia- 134 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 4.2 Large Job Flows in Transition Economies 25 20 15 10 employment 5 0 total of ­5 % ­10 ­15 ­20 Slovenia Hungary Estonia Latvia Romania Developed Developing countries countries Job creation: firm entry Job creation: continuers Job destruction: firm exit Job destruction: continuers Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Estonia: 1997­2000. Hungary:1994­2000. Latvia: 1994­99. Romania: 1994­2000. Slovenia: 1993­2000. Developed countries: West Germany 1978­99, USA 1989­97, Italy: 1988­93, U.K. 1987­98, Portugal 1984­94, Finland 1989­98, France 1991­96, Canada 1985­97, Denmark 1982­94, Netherlands 1994­97. Developing countries: Argentina 1997­2001, Mexico 1987­2000. tives from emerging and creating new jobs. Thus, job creation did not catch up with job destruction, leading to net job losses. In other words, slowing down the required transformation of the economy cannot prevent job destruction, but risks reducing incentives for job creation, with a consequent buildup of unemployment or nonparticipation. Job Creation Largely Depends on the Ability of New Firms to Enter the Market and Hire Workers An essential channel for job creation in transition economies is the creation of new firms. As an example, in Russia, before the transition, firm turnover (entry and exit of firms) accounted for less than 20 per- cent of overall job turnover. During the transition, the contribution of firm turnover to job flows increased strongly. In the countries for which firm-level data are available, during the initial phases of the transition, the entry of firms contributed from about 40 percent of total job creation in Estonia, Latvia, and Hungary to more than 70 percent in Romania and Slovenia. Moreover, there is a stronger (positive) correlation between the number of entrant firms in a given country and industry and the over- all job creation in that country and industry in transition countries than in OECD countries. By contrast, there is almost no correlation between exit of firms and job destruction in transition countries, while Restructuring, Productivity, and Job Creation 135 FIGURE 4.3 Unsynchronized Job Creation and Destruction Can Give Rise to Unemployment or Underemployment Synchronized job flows Less synchronized job flows Estonia Russian Federation 14 14 12 12 10 10 8 8 employment employment 6 6 total 4 total 4 of of % 2 % 2 0 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Slovenia Romania 14 20 18 12 16 10 14 8 12 10 employment 6 employment 8 total 4 total 6 of 4 2 of % % 2 0 0 1993 1994 1995 1996 1997 1998 1999 1994 1995 1996 1997 1998 1999 Job creation Job destruction Net employment losses Net employment gains Source: Bartelsman, Haltiwanger, and Scarpetta 2004. in OECD countries this correlation is stronger. In other words, entry plays a significant role for job creation, but the exit of obsolete firms, while promoting productivity, does not strongly contribute to job destruction, which largely comes from downsizing of surviving firms. Not surprisingly, the contribution of firm entry to overall job cre- ation declined over the transition period (see figure 4.4). After having filled the pretransition void in certain activities, the characteristics of job flows converged toward those observed in market economies: job creation and destruction come increasingly from within firm adjust- ment, rather than from the entry and exit of firms. In this context, Russia and Ukraine--two laggard reformers--saw the reverse pat- terns, with an increasing role of firm entry to total job creation in the second half of the 1990s. Notably, the contribution of firm closure to job destruction fol- lowed the opposite trend than that of firm entry and job creation. 136 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 4.4 Job Flow Rates, Selected Transition Countries, 1990­2001 (percentage) Total economy Hungary Latvia 20 25 15 20 employment) employment) or 15 10 or 10 (firms (firms 5 total total 5 of of % % 0 0 1994 1995 1996 1997 1998 1999 1994 1995 1996 1997 1998 Romania Slovenia 20 25 20 15 employment) employment) 15 or or 10 10 (firms (firms 5 total total 5 of of % % 0 0 1994 1995 1996 1997 1998 1999 1992 1993 1994 1995 1996 1997 1998 1999 Manufacturing Russian Federation Ukraine 20 20 15 15 employment) employment) or or 10 10 (firms (firms 5 5 total total of of % % 0 0 1989 1991 1993 1995 1997 1999 2001 1994 1995 1996 1997 1998 Job creation Job destruction Firm entry Firm exit Sources: Bartelsman, Haltiwanger, and Scarpetta 2004; Brown and Earle 2004. Restructuring, Productivity, and Job Creation 137 Competitive pressures have increased over the transition process, and in all the countries for which data are available, the share of job destruction resulting from firm exits increased later in transition. How does the share of job flows attributable to firm turnover com- pare with that of nontransition countries? For the sample of countries included in the analysis, the entry of firms contributed (on average) 38 percent to the total job creation rate (ranging from 25 percent in Latvia to 50 percent in Romania), exceeding this share in OECD countries and developing countries, where it amounted to slightly more than 30 percent (on average). This suggests that the entry of firms indeed played an exceptionally important role in job creation in transition countries. In contrast, at 28 percent, the average share of firm exits in job destructions was below that in the group of OECD and developing countries, where it was around 35 percent. What Is the Role of Firm Restructuring and the Entry and Exit of Firms for Productivity and Output Growth? If restructuring and creative destruction are important for employ- ment, they are also essential for promoting productivity and output growth. At least during the initial phases of the transition, these two objectives--employment growth and productivity--have been con- flicting because firms have attempted to promote productivity by reducing overstaffing. In Manufacturing, Productivity Growth Is Largely Resulting from Within-Industry Restructuring The effects of restructuring and reallocation of labor on productivity are analyzed at different levels for a group of CEE countries, Russia, and Ukraine.4 This is first done by exploring the reallocation across manu- facturing industries (see box 4.3). Then it is done by looking at how productivity growth is accounted for by within-firm growth, shifts in employment across existing firms, and the entry and exit of firms. In CEE economies, despite major changes in relative prices and revealed comparative advantages, most productivity growth over the past decade has come from changes within each manufacturing industry, rather than from shift of resources from less- to more-pro- ductive industries (figure 4.5). The effect of reallocation of labor from less- to more-productive industries is positive, albeit small. The results of the decomposition also suggest that those industries that con- tributed the most to aggregate labor-productivity growth in manufac- 138 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 4.3 Assessing the Impact of Labor Reallocation on Productivity Growth The effects of shifts in sectoral shares on aggregate productivity growth can be calculated using different techniques. In all cases, it is crucial to consider the shift of employment not only from sectors with low-productivity growth to sectors with high-productivity growth but also from sec- tors with low-productivity levels to those with high-productivity levels. The reason is that the positive contribution to aggregate productivity of the high-growth sectors may be offset by their lower-than-average productivity levels. One approach (equation 4.1) is to express the productiv- ity for the economy as a whole as the sum of the productivity level of each sector weighted by the sectoral employment shares: Ym n Yj Lj n Pm=Lm =  * =  Pj * Sj j=1 Lj Lm (4.1) j=1 where Y is output, L is employment by sector (j = 1.....n) and the total economy (m), P is labor productivity (Y/L), and S is the sectoral employment share. In a discrete time perspective, the expression (equation 4.1) can be rewritten as follows: n n n t 0 0 t 0 t 0  (Pj -Pj )*S0j  Pj *(Stj - S0j )  (Pj -Pj )*(Stj - S0j ) Pm - Pm = j=1 j=1 j=1 + + (4.2) Pm 0 n n n  Pj S0j 0  Pj S0j 0  Pj S0j 0 j=1 j=1 j=1 for a current year t and a base year 0. The first term on the right-hand side is the within industry contribution to overall productivity growth. The second term can be defined as the net shift effect (that is, the contribution com- ing from changes in the sectoral composition of employment). The third term is derived as a residual and represents the joint effect of changes in employment shares and sectoral pro- ductivity: it is positive if sectors with above-average productivity growth increase their share of total employment; it is negative if either expanding sectors have below-average productiv- ity growth or sectors with high-productivity growth are also declining in their shares of total employment. The data used for the productivity decomposition in this study refer to 14 manufacturing indus- tries in the manufacturing sector of 11 CEE countries. They refer to 1989­2002, although for a number of countries data coverage is more limited. Restructuring, Productivity, and Job Creation 139 turing were also shrinking in size, while those with below-average productivity growth gained--in relative terms--in employment shares. Industries Have Boosted Productivity by Reducing Overstaffing The dominance of within-industry changes in total productivity growth is a common result for OECD countries, especially in recent years. But in previous decades, when structural changes were more marked, even in OECD manufacturing, productivity was also largely driven by substantial shifts in employment from less- to more-pro- ductive industries. Why is this not occurring on a large scale in CEE economies, despite massive changes in the patterns of demand, trade, and the use of production factors? Most industries have experienced job losses: first because of the decline in demand, and then to maintain or reinforce productivity by reducing the large overstaffing of the centrally planned era. The fact that the industries with above-average productivity growth are those that have shed relatively more labor confirms the hypothesis advanced above: productivity growth has been driven by a process of "defensive restructuring" rather than by reallocation of production factors to the most-productive uses (strategic restructuring).5 Also, many low-productivity industries--being sheltered from competitive pressure--have managed to contain job destruction, but may still have to go through a period of downsizing and restructuring, with further effects on employment and unemployment rates. Wage Growth Did Not Vary as Much as Productivity Growth across Industries Have wages responded to different productivity performance of industries and firms during the transition? Table 4.2 presents the average growth rates of labor productivity, real wages, and unit labor costs in manufacturing during the early phases of the transition and the most recent period. It also shows the within-manufacturing dis- persion of each of these variables (coefficients of variation), as well as the correlations between labor-productivity growth and real-wage growth. The picture is rather heterogeneous across transition countries and time periods. Two of the countries (Romania and Slovenia) even had a negative interindustry correlation between real-wage growth and labor-productivity growth, and in most of the other countries the cor- 140 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 4.5 Decomposition of Labor Productivity Growth, CEE Countries Czech Republic Hungary 2 15 2 15 growth 10 10 productivity 1 1 5 5 labor 0 0 manufacturing 0 0 productivity in ­5 ­5 total of ­1 ­10 ­1 ­10 growth Aggregate 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Shares Latvia Lithuania 2 15 2 15 growth 10 10 productivity 1 1 5 5 labor 0 manufacturing 0 0 0 in productivity ­5 ­5 total of ­1 ­10 ­1 ­10 growth 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Aggregate Shares Poland Romania 2 15 2 15 productivity 10 10 growth 1 1 5 5 labor manufacturing 0 0 0 0 in total ­5 ­5 productivity of ­1 ­10 ­1 ­10 growth Shares 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Aggregate Slovak Rep. Slovenia 2 15 2 15 growth 10 10 productivity 1 1 5 5 labor 0 0 manufacturing 0 0 in ­5 ­5 productivity total of ­1 ­1 ­1 ­10 growth 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Shares Aggregate Within industry Net shift effect Interaction effect Average annual labor productivity growth (%) Source: Bank staff calculations. Restructuring, Productivity, and Job Creation 141 relation was positive, but not very strong. Most likely, industries that were more exposed to competitive pressures and privatization com- bined downsizing with wage moderation to survive, while industries sheltered from competition were able to resist restructuring and maintained real-wage growth despite poor productivity performance. The second phase of the transition saw a significant increase in the interindustry correlation of real wages and productivity, indicating that market mechanisms played their role in a more effective way. The closer association between real-wage and productivity growth in the more recent years is also related to the fact that it is often difficult to reduce nominal or even real wages to compensate for negative or only modest productivity growth, as observed in the early phases of the transition. Another interesting element emerging from the table is that across manufacturing industries, many countries had a lower dispersion in interindustry real-wage growth than in labor-productivity growth in the second phase of the transition. The result is that differences in labor-productivity growth also showed up in changes in relative unit labor costs (that is, industries with above-average productivity growth also improved their relative position in relative unit labor costs). CEE Countries Are Gaining Comparative Advantages in High-Technology Industries These changes in unit labor costs are modifying the comparative cost advantages of industries in most transition countries. At the beginning of the transition, most countries had comparative advantages with respect to Western European countries in low-technology industries (textiles, clothing, footwear, and leather products) and in resource-intensive industries (wood, chemicals, nonmetallic minerals, and so forth). Over time, however, one can observe a general pattern of catching up of productivity by medium- and high-technology indus- tries such as machinery, electrical equipment, and transport, outpac- ing productivity increases of low-technology industries where relative productivity growth has been stagnant or very small (Landesmann 2000). By contrast, real-wage growth has been similar in these indus- tries to that observed in medium- and high-technology industries that experienced major productivity improvements. This wage drift between industries implies that countries are gaining comparative (unit cost) advantages in the medium- and high-technology industries or alternatively that they are losing comparative advantages in those sectors that are labor-intensive and where transition countries were expected to gain significant shares of EU market shares. 142 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union TABLE 4.2 The Evolution of Productivity, Wages, and Unit Labor Costs Manufacturing 1992­95 1996­02 Average1 Corr. coef.2 Corr. Average1 Corr. coef.2 Corr. Bulgaria Avg. labor productivity growth 0.69 0.60 Avg. real wage growth (3) 2.66 ­0.16 Avg. unit labor cost (EURO) growth (4) 7.83 1.43 Correlation labor productivity and real wages (5) 0.61 Croatia Avg. labor productivity growth 5.78 1.61 Avg. real wage growth (3) 2.66 1.23 Avg. unit labor cost (EURO) growth (4) 0.10 0.18 Correlation labor productivity and real wages (5) 0.48 Czech Rep. Avg. labor productivity growth 0.33 ­0.37 3.90 0.55 Avg. real wage growth (3) ­1.25 ­0.80 3.81 3.17 Avg. unit labor cost (EURO) growth (4) 4.48 1.95 4.34 0.94 Correlation labor productivity and real wages (5) 0.51 0.84 Estonia Avg. labor productivity growth 6.28 10.02 2.88 Avg. real wage growth (3) 5.20 4.52 0.58 Avg. unit labor cost (EURO) growth (4) 2.58 1.90 Correlation labor productivity and real wages (5) 0.58 Hungary Avg. labor productivity growth 14.91 2.05 8.28 0.96 Avg. real wage growth (3) 0.23 ­0.13 3.40 2.97 Avg. unit labor cost (EURO) growth (4) ­8.83 ­1.56 ­1.00 0.32 Correlation labor productivity and real wages (5) 0.40 0.35 Latvia Avg. labor productivity growth 1.57 ­0.08 7.22 1.49 Avg. real wage growth (3) 7.70 0.46 2.02 1.11 Avg. unit labor cost (EURO) growth (4) 4.03 0.72 Correlation labor productivity and real wages (5) 0.12 0.23 Lithuania Avg. labor productivity growth 5.78 0.98 Avg. real wage growth (3) 5.06 1.97 Avg. unit labor cost (EURO) growth (4) 14.40 2.80 Correlation labor productivity and real wages (5) 0.71 Poland Avg. labor productivity growth 10.67 2.39 8.92 3.28 Avg. real wage growth (3) 3.13 1.27 2.94 1.84 Avg. unit labor cost (EURO) growth (4) 1.03 0.37 0.83 0.35 Correlation labor productivity and real wages (5) 0.39 0.41 Romania Avg. labor productivity growth 0.90 0.09 5.18 0.93 Avg. real wage growth (3) ­6.18 ­2.64 ­1.09 ­0.16 Avg. unit labor cost (EURO) growth (4) ­4.66 ­0.42 0.96 0.42 Correlation labor productivity and real wages (5) ­0.44 0.24 Slovak Rep. Avg. labor productivity growth 1.82 0.24 7.68 1.46 Avg. real wage growth (3) 2.45 1.13 3.19 2.51 Avg. unit labor cost (EURO) growth (4) 13.48 3.04 0.87 0.50 Correlation labor productivity and real wages (5) 0.78 0.75 Slovenia Avg. labor productivity growth 2.56 0.60 2.92 ­0.07 Avg. real wage growth (3) 3.08 1.68 4.41 2.03 Avg. unit labor cost (EURO) growth (4) 3.68 0.49 3.80 0.56 Correlation labor productivity and real wages (5) ­0.31 0.73 Source: Authors' calculations. Note: 1. Weighted average across 14 manufacturing industries (in percent) 2. Unweighted average divided by the standard deviation (in percent) 3. Average monthly wages, real (defl. CPI) annual changes in percent. 4. Unit labor cost (EURO) - annual in percent 5. Correlation between average growth rates in real wages and labor productivity growth. Restructuring, Productivity, and Job Creation 143 Firm Creation and Destruction Is Also Vital for Promoting Productivity Growth One interesting result emerging from the previous section is that produc- tivity growth was largely driven by performance within each manufactur- ing industry, rather than by reallocation of resources across manufacturing industries. To shed more light on this result and proceeding to the level where the core action of restructuring is taking place, this section analyzes firm-leveldatatoassesshowrestructuringofexistingfirmsandtheprocess of creative destruction (that is, the entry of new firms that displace old and obsolete units) contribute to productivity growth. There are a number of ways in which aggregate productivity can be decomposed into a within-firm component and other components related to the reallocation of resources across firms. The approach used in this sec- tion distinguishes five different components of productivity growth: the within component, accounting for productivity growth taking place within firms; the between component, capturing the increases in aggregate produc- tivity that come from high-productivity firms gaining market shares or from low-productivity firms losing market shares; the covariance or cross component, a term that combines changes in market shares and changes in productivity (it is positive if enterprises with growing productivity also experience an increase in market shares); and components attributable to entry and exit of firms (see box 4.4 for further details). BOX 4.4 The Decomposition of Productivity Growth Using Firm-Level Data One approach (equation 4.3) used to decompose productivity growth is from Foster, Halti- wanger, and Krizan (2001). It uses base-year market shares as weights for each term of the de- composition:  DPt = qit-kD pit +  Dqit(pit-k - Pt-k)+  DqitD pit Continuers Continuers Continuers +  qit(pit - Pt-k)-  qit-k(pit-k - Pt-k) (4.3) Entries Exits where D. means changes over the k-years' interval between the first year (t ­ k) and the last year (t); qit is the share of firm i in the given industry at time t (it could be expressed in output or em- ployment); pi is the productivity of firm i, and P is the aggregate (that is, weighted average) pro- ductivity level of the industry. The first term is the within component; the second term is the be- tween component; the third term is the so-called covariance or cross component; and the fourth and fifth terms are the entry component and exit component, respectively. 144 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 4.6 Contribution of Reallocation to Russian Labor-Productivity Growth, 1986­2001 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 ­4 ­2 0 2 4 6 8 % growth Exit Entry Cross Between Source: Brown and Earle 2004. Note: between = productivity growth resulting from reallocation of labor across existing firms entry = productivity growth resulting from entry of new firms exit = productivity growth resulting from exit of firms cross = changes in market share and changes in productivity Firm-level data for a sample of CEE countries plus Russia and Ukraine suggest that since the beginning of the transition, reallocation of employment across firms and industries has gained a stronger role in promoting productivity growth (figures 4.6 and 4.7).6 Nevertheless, efforts to promote greater efficiency within each firm also play an Restructuring, Productivity, and Job Creation 145 important role. Also, confirming the findings obtained with more aggregate data, there is little evidence of labor reallocation across exist- ing firms from the least- to most-productive units. If anything, firms experiencing an increase in productivity were also losing employment shares (that is, their productivity growth was associated with restruc- turing and downsizing, rather than expansion) (figure 4.7). The net entry effect (entry plus exit) is generally positive in most transition countries, accounting for between 20 percent and 50 per- cent of total productivity growth. In particular, there is clear evidence that the exit of obsolete firms released resources that could be used more effectively by new or existing firms. Although lack of experience and small size often make new firms less productive than the average incumbent in OECD countries, new firms in transition economies are (on average) more efficient than the incumbents. They have been able to fill in new market niches and adopt new and more efficient tech- nologies, thereby contributing to productivity and output growth. FIGURE 4.7 Sources of Productivity Growth in Transition and Emerging Economies Labor productivity decomposition shares: manufacturing, three-year differencing, real gross output 150 100 50 total of 0 % ­50 ­100 Argentina Chile Colombia Estonia Hungary Indonesia Latvia Romania Slovenia 1993­2001 1983­99 1985­98 1998­2001 1995­2001 1993­5 1999­2002 1998­9 1995­2001 Within firm Between firms Cross Entry of new firms Exit of firms Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Within = within firm productivity growth between = productivity growth resulting from reallocation of labor across existing firms entry = productivity growth resulting from entry of new firms exit = productivity growth resulting from exit of firms cross = changes in market share and changes in productivity 146 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union In Medium- and High-Technology Industries, New Firms Are Better at Harnessing New Technologies Dividing manufacturing industries into a low-technology group and a medium- and high-technology group suggests important differences in the sources of productivity growth. Defensive restructuring seems to be largely concentrated in low-technology industries, where the contribution of new firms to productivity growth is modest (Scarpetta and Vodopivec 2005). By contrast, the entry of new firms plays a strong role in boosting productivity in medium- and high-technology industries. This is consistent with the idea that in areas where there are greater opportunities for adopting newer and better technologies, new firms play an essential role. New Firms Also Promote Productivity of Incumbents by Raising Market "Contestability" The role of entry and exit of firms is also important in reinforcing competitive pressure on incumbents. The risk that new firms may steal market shares from incumbents and the risk of failure both act as strong disciplinary devices to promote better performance by incumbents. A first look at such contestability effects is provided in figure 4.8, which presents the correlation between the contribution of net entry in a country or sector (using time averages from the FIGURE 4.8 Relationship between Net Entry Contribution and Productivity Growth of Incumbents Firm entry and productivity growth of incumbents: five-year differencing, real gross output, country and industry time averages % 1.5 growth, 1.0 0.5 productivity 0.0 Incumbents' ­0.5 ­1.0 ­0.5 0.0 0.5 1.0 1.5 Net entry productivity growth, % Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Note: Correlation coefficient: 0.58, statistically significant at 1 percent. Each dot is one country-industry observation. Ex- cludes Brazil and República Bolivariana de Venezuela. Outliers excluded. Restructuring, Productivity, and Job Creation 147 country-sector-year data) and the productivity growth from incum- bents (the within component in the productivity decomposition). Interestingly, a strong positive and statistically significant correlation exists between the net entry contribution and the productivity growth of incumbents. This finding is suggestive that there is a rela- tionship between the creative destruction and within-firm sources of productivity growth.7 In sum, while existing firms have largely responded to the need to improve performance by reducing overstaffing--especially in tradi- tional sectors--the creative destruction process has boosted produc- tivity growth, both directly by the entry of new, more-productive firms that have displaced less-productive units and indirectly by rein- forcing market contestability. What Drives Restructuring of Existing Firms? Restructuring of Existing Firms Has Been Largely Influenced by Privatization Privatization has often been the main trigger of firm restructuring, and transition countries have used different privatization methods, have transferred ownership at different paces, and have left different roles for the state as shareholder in privatized firms. All these factors have played a role in fostering the reduction in inefficiency and in promoting firm development. Transition countries have generally privatized small firms quickly and completely, while the privatization of large firms has proceeded more gradually. Among transition economies, CEE countries priva- tized both small and large firms faster and more comprehensively than SEE or CIS countries. The dispersion of both small- and large- scale privatization progress has been by far the largest in the CIS, while it has been relatively uniform in CEE countries, particularly for small-scale privatization. Privatization Has Promoted Firm Performance Privatization strongly improved enterprise performance, but its effects vary substantially by type of privatization and by groups of countries. The recent "quantitative" survey of Djankov and Murrell (2002), based on meta-analysis of 93 empirical studies from various transition countries, finds that outside owners have been more effective in improving performance than insiders and that privatization methods 148 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union generating concentrated ownership--particularly by foreigners-- improved performance the most. Firms with diffused outsider owner- ship performed virtually the same as state firms, and worker-owned firms performed the worst. Productivity effects of privatization also vary strongly across countries. In Central and Eastern Europe, these effects have been more statistically significant and greater in size (often adding several percentage points to enterprise growth) than in the CIS (Djankov and Murrell 2002).8 A more in-depth assessment of the privatization process in four countries (Hungary, Romania, Russia, and Ukraine) sheds further light on these issues (figure 4.9). It suggests that the effect of foreign privatization on performance is strikingly similar across countries, in contrast to the effects of domestic privatization. This suggests that for- eigners' advantages in accessing finance, new technologies, the latest managerial techniques, and world markets far outweigh any disad- vantages resulting from unfamiliarity with local conditions and weak political connections. The privatization methods used in Hungary and Romania also resulted in concentrated ownership from the time of privatization, whereas mass privatization in Russia and Ukraine led to dispersed ownership by employees and small outside investors.9 Secondary trading has resulted in more-concentrated ownership in Russia and Ukraine, but this has taken time. This raises the possibility that if ownership concentration is a necessary condition for privatization to have a positive effect, then the effect may have appeared sooner in Hungary and Romania than in Russia and Ukraine. FIGURE 4.9 Effects of Foreign and Domestic Privatization on Multifactor Productivity Growth (MFP) Dynamics of Foreign Privatization Effects Dynamics of Domestic Privatization Effects 1.2 Privatization Privatization 1.8 1.0 0.8 growth Years before Years after Years before Years after 1.3 0.6 . 0.8 0.4 productivity state-owned 0.2 0.3 vs. 0.0 ­0.2 ­0.2 Multifactor ­8 ­7 ­6 ­5 ­4 ­3 ­2 ­1 0 1 2 3 4 5 6 7 8 ­8 ­7 ­6 ­5 ­4 ­3 ­2 ­1 0 1 2 3 4 5 6 7 8 Years before/after privatization Years before/after privatization Russian Federation Ukraine Hungary Romania Source: Brown and Earle 2004. Restructuring, Productivity, and Job Creation 149 Positive Effects on Job Creation Are Found among Foreign- Owned Privatized Firms Privatization has also had significant effects on workers (see figure 4.10). Evidence from these four countries suggests that foreign priva- tization has increased employment and wages in all countries, though the effects are not always large. Domestic privatization has been less kind to workers, however, because wages suffered in all four coun- tries and employment rose only in a few instances. But there are also significant differences across types of workers. Thus, employment composition and relative wage changes have been significantly biased toward white-collar employees. This is consistent with the idea that foreign privatization, by promoting the upgrading of production processes with new technologies, tends to be skill-biased. It also suggests that although foreign privatization has improved the welfare of workers overall, it has also increased inequality. How Many Firms Enter and Exit the Market in Transition Countries? Much Creation and Some Destruction of Firms As emphasized earlier, firm dynamics--especially firm entry--have been an essential driver of job creation and an engine for productiv- FIGURE 4.10 Effects of Foreign and Domestic Privatization on Productivity, Employment, and Wages 45 40 35 30 25 employment 20 in 15 10 change 5 % 0 ­5 ­10 Hungary Romania Russian Federation Ukraine Foreign Domestic Source: Brown and Earle 2004. 150 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ity growth. How many firms enter and exit the market in transition economies? What are their characteristics, compared with those of the incumbents? A common fact in all market economies is that many firms enter and exit most markets every year. This is particularly true in transi- tion economies where 20­25 percent of firms (with at least one employee) have entered or exited the market (on average) annually in the past decade (figure 4.11). Moreover, in transition economies, firm entry largely outpaced firm exit. Obviously, this is related to the process of transition and is not sustainable over the longer run. Still, it points to the fact that new firms not only displaced obsolete incum- bents in the transition phase but also filled in new markets that were either nonexistent or poorly populated in the past. This is also reflected in the discrepancies between firm entry and exit across firm size. Indeed, the entry of firms is particularly large among microunits (20 or fewer employees). During the centrally planned system, there were relatively few of these microfirms; however, they exploded dur- ing the transition in most business service activities. Even Successful Entrants Do Not Expand Significantly A high entry rate is in itself a signal of market dynamism, and espe- cially in transition economies, it is essential to develop new private activities. However, in most market economies, high entry sectors are also characterized by high exit rates, involving many young busi- nesses that are censored by the market. So, how much do new firms displace old, obsolete units, and how high are failure rates among newcomers in the first years of their life? FIGURE 4.11 How Many Firms Enter and Exit the Market? Entry and exit rates for all firms with 1 or more employee(s), 1988 and beyond 25 20 firms 15 total 10 of % 5 0 Italy States France Latvia Denmark Germany Finland Estonia Canada Mexico Argentina Portugal RomaniaHungary Slovenia W. E. Germany United Netherlands Entry rate Exit rate Source: Bartelsman, Haltiwanger, and Scarpetta 2004. Restructuring, Productivity, and Job Creation 151 Compared with the OECD economies, new firms in transition countries seem to have higher survival rates (Scarpetta and Vodopivec 2005). Except for Romania, more than 70 percent of entrant firms are still in the market after four years, while the proportion is often around 60 percent in the OECD countries. However, the expansion of successful new firms is limited, and new firms remain of relatively small size. This is common to many European countries and largely differs from other markets, includ- ing the United States, where successful entrants double in size in the early years of life.10 These results point to the fact that the per- formance of new firms may be affected not only by entry conditions but also by factors that affect the decision to expand and hire more workers. These include access to finance, labor adjustment costs, and stability of the economic and regulatory environment (see chapter 5). Summing Up: Entry Conditions and Incentives to Create Jobs Are Essential for Improving Job Creation in the Region At the onset of transition, the presence of many inefficient state- owned firms offered a unique challenge for countries in the Region, but also a major opportunity for vast productivity improvements and sustained ameliorations of living standards. Because of large misallocation of resources, reallocation of labor and other resources--across firms, sectors, and locations--was perceived as one of the foremost tasks of transition. New control mechanisms of firm efficiency were put in place, dictating the pace and depth of restructuring. The process of creative destruction (that is, the exit of unprofitable firms, paving the way for the creation of new, more- productive ones), invigorated by privatization and competition, was expected to be exceptionally important in raising productivity in transition countries. To a large extent, these expectations were borne out. Transition brought a remarkable surge of job flows. Firms destroyed low-pro- ductivity jobs and created high-productivity ones. The entry of new private firms in the market has been exceptionally large by interna- tional standards, contributing to the creation of many new jobs. But the exit of obsolete firms also accounted for a large share of job destruction. Because of restructuring and ensuing resource reallo- cation, the sectoral structure of employment strongly changed, with the increase of market services, deindustrialization, and reduction of employment in agriculture being the main outcomes. However, 152 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union this general picture hides large differences across transition economies. For example, reflecting harsher adjustment conditions and slower progress in transforming the institutions to market stan- dards, the share of agriculture in total employment increased in most CIS countries. The reallocation of resources also proved to be efficiency-enhanc- ing: although productivity increases have largely been driven by within-firm forces, resulting in defensive restructuring, firm entry has strongly increased productivity. In particular, not only have many new firms been relatively more productive than existing ones in tran- sition--a fact that is uncommon in OECD countries--but their entry has also contributed to igniting competitive pressure on incumbents promoting their restructuring. Investment climate conditions are also of great importance to stimulating the expansion of successful firms, often constrained by lack of access to finance or uncertainties about the institutional and market environment in which they operate. And by continuing to provide budgetary support to ailing firms (albeit often not as direct subsidies), many transition countries have not provided consistent and credible enforcement of firm exit (the share of job destruction resulting from exit of firms has been below that in nontransition countries, and increasing it would quite possibly be productivity- and welfare-increasing). These countries have therefore underuti- lized the potential exit of firms as a disciplinary device, as well as the actual exit of firms as a means of freeing resources to other, more- productive uses. The main engine to enterprise restructuring has been the transfer of ownership to the private sector. Most pretransition firms have been privatized in Eastern Europe and the Former Soviet Union, although the method of privatization and the governance of the new private entities vary a great deal across countries of the Region. Most of these privatized firms have pursued defensive restructuring, with large lay- offs. Their ability to move toward more strategic restructuring--with new investment and job creation--depends on the same investment climate conditions affecting newly created firms. The evidence that foreign-owned privatized firms have had better productivity and employment performance than domestically owned firms suggests possible difficulties in accessing new technologies, in exploring foreign markets, and in management. The stance of the investment climate in transition economies and its impact on firm performance and job creation are reviewed in the next chapter of this report. Restructuring, Productivity, and Job Creation 153 Notes 1. Previous studies have used two types of benchmark to assess how far away transition countries were from market economy structures and how much they had progressed over the transition periods toward these market structures. The first approach uses the GDP and employment structure of the neighboring Western European countries as a reference, using Northern and Southern European countries as two possible bench- marks (Aiginger and others 1999; Landesmann 2000). Alternatively, other studies have used regression analyses based on cross-country data for a large number of nontransition market economies (Raiser, Schaffer, and Schuchhardt 2003). The first approach is more relevant for accession countries, whose level of development is relatively close to their Western European counterparts; for these countries, both approaches yield simi- lar conclusions. The second approach, however, is more adequate to assess CIS countries, whose market-structure benchmark is probably to be found in low- or low-to-middle­income developing countries and is used in this study. 2. The data refer to 1989­2002, although for a number of countries data coverage is more limited. 3. The distortion index is calculated as the sum of the observed distortion in each macrosector of the economy. The market benchmarks are obtained through regressions of sectoral employment shares on GDP per capita levels and its square. In this respect, the distortion index is a measure of the overall distance of an economy from a market economy with the same per capita income. See also Raiser, Schaffer, and Schuchhardt (2003) for more details. 4. Given data availability, the decomposition is limited to 14 manufacturing industries in 11 CEE countries. This limits the scope of the analysis because, as stressed above, most of the reallocation of labor has been driven by shifts away from manufacturing into business services, rather than within manufacturing. Still, it helps to assess whether market forces have promoted reallocation of labor toward those industries with greater growth potentials. 5. See Grosfeld and Roland (1996) for a discussion of defensive versus strategic restructuring. 6. The analysis focuses on Estonia, Hungary, Latvia, Romania, and Slove- nia, and it is based on business register and enterprise survey data. For more details, see Bartelsman, Haltiwanger, and Scarpetta (2004). 7. It might be, however, that the correlation is resulting from the impact of technological advances for both continuing firms and for the creative destruction process. With technological advances, it can be observed that incumbents who survive increase productivity and that entering busi- nesses (which presumably adopt the latest advances) are more produc- tive than the exiting businesses. To focus on the contestability hypothesis more directly, Bartelsman, Haltiwanger, and Scarpetta (2004) also exam- ine the relationship between the firm turnover rate and the productivity growth of incumbents. They again find a positive and statistically signif- icant correlation. This latter finding provides even more direct evidence 154 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union of a connection between the competitiveness or contestability of markets and the productivity growth within incumbent firms. 8. Nearly all the studies covered by Djankov and Murrell (2002) are based on small surveys conducted soon after privatization. Since then, a few studies have used comprehensive manufacturing firm data with long time series to provide more rigorous tests of the effects of privatization. Orazem and Vodopivec (2003) find no effect of either domestic private or foreign ownership on total factor productivity in Slovenia during 1994­2001, but a strong effect of market competition. Lizal and Svejnar (2002) find a clear positive effect of foreign ownership on firm produc- tivity and employment growth in the Czech Republic during 1992­98, but no effect of domestic ownership. The insignificant domestic owner- ship effect on productivity is surprising at first glance, given that the postprivatization ownership structure was concentrated in the hands of investment funds. These investment funds were poorly regulated, how- ever, and many were at least temporarily owned by state banks. 9. Romania used management-employee buyouts early in the privatization process, but employees voted as a group, unlike in Russia and Ukraine. 10. Among the few CEE countries for which data are available, Romania is an exception: not only are failure rates higher than in the other coun- tries but even successful entrants have very limited opportunities of expanding. CHAPTER 5 The Investment Climate and Job Creation This chapter analyzes the link between investment climate and job creation. It shows that differences in the investment climate account for a significant part of the variation of employment outcomes in the Region. It thus argues that improvements in the investment climate are critical for the creation of more and better jobs in the Region. Although the investment climate varies considerably in the transi- tion countries, it is generally worse than in both developing and developed countries with the most hospitable business environments. For example, it takes 11 days to start a business in Latvia and more than 100 days in Azerbaijan. In CEE, the average time to start a busi- ness is 50 days, compared with only 4 days in Denmark or the United States. The cost to register a business in low income CIS countries is significantly higher than in Thailand, for example. In all of the transition countries, there is room for reducing the risks, costs, and barriers associated with business operation, although the nature and intensity of constraints vary across countries. In low income CIS countries, the dominant problem is underdeveloped insti- tutions of a market economy (including security of property rights), economic and regulatory policy uncertainty, arbitrary and selective application of the regulations, corruption, and poor infrastructure. Middle income CIS countries, among them Russia, have a higher eco- nomic potential and perform better in some dimensions of the invest- 155 156 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ment climate than do the poorer low income CIS countries. However, the challenges are still profound. Corruption is widespread. Burden- some regulations and bureaucratic harassment hamper business growth, as do underdeveloped financial markets. In SEE, the playing field for firms is not level because of various-- formal and informal--constraints on competition. The legal system does not function effectively, making contract enforcement difficult. Administrative barriers in some areas (for example, customs and trade) are still high, inhibiting firm growth and fuelling corruption. In CEE, institutions of a market economy are more developed and mar- ket-oriented reforms more advanced, easing some of the constraints. Yet, much can be done to lower the cost of doing business. High taxes, including taxes on labor, inhibit investment and hiring. Access to credit, although much easier than in other parts of the Region, is still seen as difficult by entrepreneurs. In some CEE countries, onerous regulations, including labor regulations, are a significant impediment to firm expansion. Finally, in some CEE countries, employers find it difficult to hire workers with necessary skills. How to foster job creation in the Region? In low income CIS coun- tries, the key issue is to develop institutions of a market economy and lower the risk associated with doing business. Property rights need to become more secure and infrastructure more reliable. In middle income CIS countries and SEE, focus should be placed on lowering the administrative barriers. Deregulation, less discretionary power in the hands of bureaucrats, product market competition, and a more effective legal system would address the most binding con- straints. Finally, in CEE, priority should be given to addressing the demand for skilled workers, improving labor market flexibility, and lowering the monetary cost of doing business. One key measure is to reduce taxation, particularly taxes on labor, which are especially high in CEE. This requires carrying through difficult public expendi- ture reforms, particularly reforms of the pension and health care sys- tems (see chapter 6). But obviously, investment climate reforms need to consider country- specific obstacles to business growth. These can be identified by using investment climate surveys and focus groups of employers and--most important--by instituting authentic dialogue with the business com- munity, including the newly established small private firms. The rest of the chapter is organized as follows. Section 1 shows the link between the investment climate and employment. Section 2 presents major obstacles to firm operation and growth in the Region, as perceived by the employers. Section 3 provides empirical evidence The Investment Climate and Job Creation 157 of the investment climate's impact on job creation in the Region and shows that differences in employment outcomes are to a large extent accounted for by differences in the investment climate. Section 4 compares the investment climate in the transition countries to that in other regions and uses benchmarking to identify areas where there is room for improvement. Section 5 summarizes the main investment climate constraints in the Region and its subgroups of countries. Importance of Investment Climate for Job Creation As stressed in the previous chapter, the birth of new firms and the expansion of existing firms are prerequisites for employment growth and the reduction in unemployment. The importance of new firm creation as a source of new jobs is particularly large in a situation of massive reallocation of workers and jobs across sectors of the econ- omy. New firms in the expanding sectors need to absorb workers released from firms in the declining sectors.1 During the past decade, the Region's economies suffered a large displacement of workers (see chapter 4). Many SOEs went out of business once they were exposed to hard budget constraints and competition. Others have downsized. They reduced labor hoarding and shed redundant labor to reach the economically efficient size. As a result, the average firm size decreased and comes closer to that in market economies. For example, presently, the average size of a Polish firm is roughly 40 percent of that at the beginning of the transition. But if many firms downsize, as is the case in the Region, then the number of firms needs to grow for employment to increase. A Sound Investment Climate Is Essential to Firm Creation and Expansion and Thus Employment Growth Firm entry depends on the costs of starting a business and also on the expected benefits and costs of operating a business (Klapper, Laeven, and Rajan 2004). Similarly, firm growth depends on the rate of return to investment, which, in turn, is influenced by competitive pressures in the product market, internal governance, and access to new tech- nologies. Institutions, policies, and regulations that affect firm entry, survival, and growth are referred to in this chapter as the "investment climate." The rest of this chapter examines the investment climate factors that hamper job creation in the Region through limiting firm entry and growth. 158 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Employers' Views on the Major Obstacles to Firms' Operation and Growth in the Region One way to assess the investment climate is to consider the entrepre- neurs' opinions about their major constraints in running and expand- ing their businesses.2 Additional valuable insights can be gained from focus group discussions (see box 5.1). Entrepreneurs See Instability and Uncertainties as Major Obstacles to Business Expansion across the Region Economic and regulatory policy uncertainty, along with macroeco- nomic instability, are top business concerns in the Region. One firm in three views these factors as major or severe obstacles to its opera- tion and growth (figure 5.1). To a large extent, high business risk is an inherent feature of the transition, but bad governance often adds to this natural uncertainty. This includes frequent changes in regula- tions, lack of regulatory consistency and clarity, arbitrary interpreta- tion of regulations and their selective enforcement, and (last, but not least) corruption. Although business is always associated with some BOX 5.1 Small Entrepreneurs Complain about the Business Environment in Bulgaria A few quotes from the survey of small entrepreneurs illustrate the problems they face in open- ing and running a business: "If I had to start again, I would not even think of opening a business." "I cannot even remember how many times I went for each permit. It just takes insanely long." "During the inspections, they pick on every single thing. When they decide to pick up your money, there is no way out. They always find something to pick on." "You must hold a law degree to be able to open a cafeteria." "Instead of thinking how to be more efficient, we spend 60 percent of our time thinking how to cope with tax authorities and inspectors." "There should be rules of the game, but clear ones and equally applicable to all." Source: World Bank 2002a. The Investment Climate and Job Creation 159 risk, reducing policy-induced uncertainty and improving governance are keys to improving the investment climate in the Region. Uncertainty and macroeconomic instability, albeit important in all countries, are--as could be expected--deemed less severe not only in those CEE countries that are members of the EU but also (somewhat surprisingly) in the low income CIS countries. In contrast, uncer- tainty and instability are very frequently perceived as major con- straints in SEE and middle income CIS countries. Taxation, Access to Finance, and Corruption Also Severely Impede Business Activity Firms in the Region frequently complain about taxation. This is natu- ral, because taxes raise costs and lower returns on investment. At the same time, taxes are necessary to finance public goods and services. So setting tax rates is a difficult balancing act. However, in the transition countries, taxes are often unduly high because of inefficiencies in ser- FIGURE 5.1 Most Frequently Reported Major Obstacles to Firm Operation in the Region Economic & regulatory policy uncertainty Macroeconomic instability Tax rates Tax administration Cost of financing Corruption Access to financing Anticompetitive informal practices Customs & trade regulations Legal system/conflict resolution Crime, theft & disorder Business licensing & operating permits Skills & education of available workers Electricity Access to land Labor regulations Telecommunications Transportation 0 5 10 15 20 25 30 35 40 45 % of firms in the Region that report the above as major or severe obstacles Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individ- ual countries or subgroups of countries in the Region. 160 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union vice delivery, poor tax collection capacity, and a narrow tax base (result- ing from a large informal sector), so there is room for reducing tax rates by improving efficiency without compromising important social objec- tives. One example is to reduce payroll taxes by reforming the pension and health care systems. This pertains especially to CEE and SEE coun- tries, where payroll taxes are particularly high (see chapter 6). An aspect of taxation that is more amenable to direct policy action is tax administration, which is seen as a major or severe obstacle by close to 30 percent of firms in the Region. Tax administration is a syn- onym for bureaucratic harassment (for example, disruptive and costly tax inspections), arbitrary interpretation of obscure tax regulations, selective enforcement, and associated extortion. As such, inefficient tax administration--an element of poor governance--contributes to both costs and risks faced by firms and so discourages investment. Tax administration is seen as a particularly severe issue in CIS countries, making it a policy priority. Another top investment climate issue in the Region, as perceived by employers, is the high cost and difficult access to finance. 3For more than one in four firms in the Region, the cost of credit is a major obstacle to firm growth. For more than one in five, difficult access (for example, high collateral requirements) is a major constraint. Evi- dence suggests that small firms are particularly hurt by poor access to credit, which limits their growth prospects (Beck, Demirguc-Kunt, and Laeven 2005). Improving access to finance, especially for small firms, is thus a priority in those cases where high real interest rates (or interest rate spreads) reflect an underdeveloped and noncompetitive banking system, which is particularly the case in low income CIS countries (De Nicoló, Geadah, and Rozhkov 2003). Finally, firm growth in the Region is inhibited by widespread cor- ruption and anticompetitive informal practices. Roughly, one firm in five sees these as major or severe obstacles to their operation and growth. Corruption and government interference are closely associ- ated with numerous administrative barriers, red tape, and discre- tionary power in the hands of bureaucrats. These governance-related issues are particularly pronounced in SEE, while less so in the new EU-member countries. Labor Regulations Are Seen as a Less Severe Obstacle in the Region Labor regulations are not among top business concerns in the Region. According to the investment climate surveys, they are ranked 14 out of 18 obstacles.4 They are seldom viewed as a major obstacle, but The Investment Climate and Job Creation 161 many firms see them as a moderate obstacle. Altogether, labor regu- lations pose a moderate, major, or severe obstacle to one firm in four in the Region, which is by no means a negligible fraction. Labor Regulations Still Matter Why are labor regulations perceived as less important in the Region? One reason is lax enforcement, especially in countries with poor administrative capacity. Another reason is that they are less impor- tant compared with other constraints. Labor regulations may con- strain business operation, but other regulations and conditions may be much more constraining. In less-advanced economies, they are dwarfed by the overall uncertainty and instability, poor access to cap- ital, corruption, and so forth. Another potential reason may be cultural or historical: in many countries of the Region, employers take historically established labor regulations for granted. This is especially likely in countries where the interests of private businesses have not yet fully taken hold. The com- parison between Croatia and Poland is illustrative. In Croatia, employ- ment protection legislation is very strict, yet employers hardly complain about it. In contrast, in Poland, employment protection legislation is significantly more liberal; nonetheless, it ranks high among key busi- ness concerns. At the same time, the pace of enterprise restructuring has been much faster in Poland than in Croatia (Rutkowski 2003a). Therefore, the difference in perceptions is likely to reflect the fact that Polish employers are more aware of the role of labor regulations as an impediment to restructuring. This explanation is supported by the fact that restructuring firms (those that actually shed labor) more often report labor regulations as a significant obstacle than do firms that have not gone through downsizing (Pierre and Scarpetta 2004a). Although labor regulations are not the top business concern, enhancing labor market adaptability is still essential for improving labor market outcomes in the Region. There are at least three reasons why the efficiency of the labor law needs to be improved:5 · The fraction of firms adversely affected by labor market regulations is significant in most the Region's subgroups of countries, and in some countries it is high. For example, in Poland, every second firm sees labor regulations as at least a moderate obstacle, and in Serbia and Montenegro and the Slovak Republic, labor regulations are seen as an obstacle by about one firm in three. In addition, as mentioned earlier, labor regulations tend to hurt firms that are most dynamic and are the source of growth and jobs. 162 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union · In many countries--mainly in the CIS--labor regulations are not seen as an obstacle only because they are not enforced (see chap- ter 6). But flexibility through nonenforcement is not an optimal solution; it undermines the rule of law and contributes to the uncertainty faced by both firms and workers. · The need for greater efficiency of labor regulations is particularly pronounced in the Region, given the paramount challenge of restructuring and productivity catch-up (see chapters 4 and 6). Worker Skills Can Constrain Firm Strategic Restructuring Skill deficiencies are a significant obstacle to firm growth, especially in CEE, where restructuring has proceeded at a higher speed and is more advanced. For example, in the Czech Republic, Hungary, Latvia, and Poland, 35 to 40 percent of employers see the skills of the available workforce as a significant obstacle to the growth of their firms. In Estonia, this proportion reaches 65 percent. A special survey of labor demand in an intensively restructuring region of Azerbaijan found that the lack of workers with adequate skills is the third most impor- tant reason that prevents firms from hiring more workers (World Bank 2005a). One important source of the skills problem is a strong bias against workers with low educational attainment and skills, asso- ciated with enterprise restructuring and technological progress brought about by the transition (Commander and Köllő 2004). Entrepreneurs' Perceptions of the Most Pressing Problems Vary by Region Although many investment climate concerns are common across most transition economies, there is also substantial variation in their relative importance by country groups. In the transition economies that recently joined the EU (EU-8), labor-related issues, such as the skills of the available workers, labor regulations, and taxation, figure prominently in the surveys (figure 5.2, panel A). In other European transition economies, entrepreneurs highlight a combination of poorly functioning institutions (for example, ineffective legal system, industrial conflict resolution) and bad governance (for example, anti- competitive and informal practices, corruption) that limit firm growth and job creation (figure 5.2, panels B and C). In the middle income CIS countries, the predominant part is played by administrative bar- riers, such as numerous licenses and operating permits, burdensome tax administration (with associated bureaucratic harassment and The Investment Climate and Job Creation 163 extortion), and inefficient regulations (figure 5.2, panel D). Finally, in the low income CIS countries, the constraints are more basic and range from unreliable infrastructure to underdeveloped institutions of a market economy (figure 5.2, panel E). It is striking that labor market issues--employment regulations and worker skills--affect firm growth prospects in CEE, while they apparently play a negligible role in the Region's other subgroups of countries.6 One explanation is a faster pace of enterprise restructur- FIGURE 5.2 Obstacles to Business Operation and Growth Vary by Subgroup Panel A--CEE EU-Member Countries Skills of available workers Labor regulations Access to land Tax rates Crime, theft & disorder Anticompetitive/informal practices Telecommunications Transport Less important More important Legal system/conflict resolution Customs & trade regulations Licensing & operating permits Cost of finance Access to finance Tax administration Economic & regulatory policy uncertainty Corruption Electricity Macroeconomic instability Average Panel B--CEE EU-Accession Countries: Bulgaria, Croatia, and Romania Legal system/conflict resolution Access to finance (e.g. collateral) Corruption Anticompetitive/informal practices Licensing & operating permits Economic & regulatory policy uncertainty Cost of finance Tax rates Macroeconomic instability Less important More important Crime, theft & disorder Skills of available workers Labor regulations Transport Access to land Telecommunications Customs & trade regulations Electricity Tax administration Average (continues on the following page.) 164 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 5.2 (continued) Panel C--Southeastern European Countries Legal system/conflict resolution Anticompetitive/informal practices Customs & trade regulations Electricity Corruption Crime, theft & disorder Transport Cost of finance Less important More important Labor regulations Economic & regulatory policy uncertainty Macroeconomic instability Telecommunications Access to land Access to finance tax rates Licensing & operating permits Tax administration Skills of available workers Average Panel D--Middle Income CIS Countries: Belarus, Kazakhstan, Russia, and Ukraine Licensing & operating permits Tax administration Access to land Customs & trade regulations Skills of available workers Access to finance Crime, theft & disorder Macroeconomic instability Economic & regulatory policy uncertainty Less important More important Tax rates Cost of finance Labor regulations Corruption Telecommunications Transport Legal system/conflict resolution Anticompetitive/informal practices Electricity Average ing in CEE than in other parts of the Region, which affects the labor market in two ways. First, widespread job reallocation gives rise to skills mismatch because the newly created jobs differ in skill content from jobs that were destroyed. This explains why employers in CEE more often than in other regions view skills of available workers as a major constraint to firm growth. Second, labor regulations become an effective obstacle only if firms face competitive pressure and have to adjust the size and composition of their workforce to stay profitable. This in turn explains why employers in faster-restructuring CEE countries more often complain about labor regulations, although on the books they are less strict than in other subgroups. The Investment Climate and Job Creation 165 FIGURE 5.2 (continued) Panel E--Low Income CIS Countries: Armenia, Azerbaijan, Georgia, the Kyrgyz Republic, Moldova, Tajikistan, and Uzbekistan Tax administration Electricity Customs & trade regulations Access to finance Licensing & operating permits Access to land Telecommunications Cost of finance Transport Tax rates Corruption Less important More important Macroeconomic instability Crime, theft & disorder Skills of available workers Economic & regulatory policy uncertainty Labor regulations Legal system/conflict resolution Anticompetitive/informal practices Average Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Bars show the deviation from the regional and factor averages. Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individual countries or subgroups of countries in the Region. Smaller Firms Tend to Be Disproportionately Affected by Investment Climate Obstacles Small firms are more sensitive to the investment climate than are larger firms. Results of multivariate regression analysis (figure 5.3) show that, controlling for other characteristics such as firms' age, smaller firms tend to be more affected by the main constraints identified above.7 Smaller firms also tend to pay more bribes, and to pay bribes more often, than larger ones do (Gray, Hellman, and Ryterman 2004). Because smaller firms are the prime source of job creation, this implies that job growth is sensitive to investment climate changes (see chapter 4). Policies to Foster Job Creation Need to Go beyond the Labor Market Entrepreneurs' perceptions of the business climate suggest that the main constraints to job creation in the Region currently lie outside the labor market. It could also mean that labor market regulations are just a part of a broader institutional and policy framework that con- strains firms' ability to grow and create new jobs. This has important implications for policies to foster employment and reduce unemploy- ment. They cannot be limited to improving the workings of the labor 166 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 5.3 Smaller Firms Are More Constrained by the Investment Climate Probability of reporting the following as major obstacle by size of firm Labor regulations Conflict resolution Anticompetitive practices Corruption Cost of financing Access to financing Tax administration Tax rates Macroeconomic instability Policy uncertainty ­3 ­2 ­1 0 1 2 3 4 5 6 % change 20 to 100 employees Fewer than 20 employees Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individ- ual countries or subgroups of countries in the Region. market; instead, they should aim at improving the overall investment climate and focus on the most severe obstacles to firm growth. Given the complexity of investment climate reforms, there is a need to set priorities and focus on the most binding constraints. Macroeconomic instability and economic and regulatory policy uncertainty are seen as major obstacles for businesses across the Region. These two aspects catch various dimensions of the economic environment in which firms operate. In particular, they relate to the quality of institutions and governance. They are likely to encompass such elements as the unpredictability of policy changes, frequent reg- ulatory changes, discretionary interpretation of laws and their selec- tive enforcement, inconsistency in regulations, and discretionary power in the hands of bureaucrats (for example, the power to give permits) and bureaucratic harassment, associated with corruption and extortion. The Investment Climate and Job Creation 167 Therefore, improving labor market outcomes entails a set of com- prehensive and complex policy reforms to remove a wide range of constraints to business operation. This involves institutional reforms and changes in governance. Which particular issues should be given priority depends on country-specific circumstances (because key con- straints differ by country) and on the extent to which various ele- ments of the investment climate affect labor market outcomes. The Impact of Investment Climate on Job Creation in the Region This section attempts to shed light on two questions. Do countries with better investment climates have better labor market perform- ance? Which institutions are most important for a good employment record? In the rest of this chapter, labor market performance is meas- ured mainly through employment generated in the new private sec- tor. The latter is proxied through service sector employment, as explained in box 5.2. Differences in Investment Climate Explain a Substantial Part of Variation in Employment Outcomes in the Region The results of a multivariate regression analysis show that differences in market services sector (used as a proxy for the new private sector) employment across the transition countries to a large degree reflect differences in the investment climate.8 In other words, knowing var- ious components of the investment climate, one can predict private sector employment reasonably well.9 This supports the main claim of this section that a favorable investment climate is critical for job cre- ation and employment growth in the Region. Table 5.1 shows how different components of the investment climate contribute to private sector employment (proxied by service sector employment controlled for GDP per capita10).11 The first row of table 5.1 reflects the actual change in service sec- tor employment (corrected by GDP per capita). All countries have experienced an increase in service sector employment, above the one implied by economic development.12 The second row shows the change in service sector employment predicted by the model, given the impact of the time shock, country- specific effect, and investment climate in the country (see endnote 11). For the new EU member (first column), the model performs quite well, which means that the set of institutions constituting the 168 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 5.2 Service Sector Employment Rate as an Indicator of Job Creation Potential The focus of this report is the employment generated by the new private sector. It might seem obvious to consider a country's total employment as an indicator of job creation potential, but there are difficulties. Total employment is the sum of employment created by the public and private sectors. Coun- tries with large public sectors have, in general, a large total employment rate, but can still have a bad business environment. In essence, public sector employment does not respond to the state of business environment institutions. To overcome this difficulty, this report focuses on market service sector employment as a bet- ter indicator of an economy's job creation potential because in modern economies--and in tran- sition economies, in particular--most job creation occurs in the market services sector. There- fore, employment opportunities are better in economies where the market services sector is large and expanding. An index summarizing the government interference in the economy has been plotted against to- tal and market service sector employment rates (figures A and B). As can be seen, the negative impact of the investment climate indicator on market service sector employment is much larg- er than the impact on the total employment rate. A. High government interference in B. If service sector employment is plotted the economy is compatible with against government interference, high employment rate a clear negative relation is found 70 ROM 5 EST LVA CZE SVN CZE KAZ % (adjusted)* HUN BGR 60 rate, LVALTU HUN SVK SVK LTU UKR POLl GEO services 0 POLl BGR MDA 50 HRV ROM Employment HRV TUR GEO Employment SVN MDA 40 ­5 2 4 6 8 2 4 6 8 Less More Less More Government interference index Government interference index Source: Lopez-Garcia 2004. Note: To control for the economic development, a regression of the service employment rate was run against the GDP per capita. The residuals from that regression represent the service employment rate not explained by the development of the economy. The residual is what is called the "service sector employment rate controlled for GDP per capita." The Investment Climate and Job Creation 169 TABLE 5.1 Contribution of Investment Climate Components to Change in Service Sector Employment in the CEE EU Countries over the Past Decade Average: of which: Average: of Aver- CEE, EU- Slovak CEE, EU- which: age: member Lithuania Hungary Republic Poland Latvia Slovenia accession Romania Russia- countries (97­02) (92­02) (94­02) (94­02) (96­02) (93­02) countries a) (94­02) Ukraine Actual change in service sector employment 1.08 ­1.83 1.11 0.66 1.33 1.89 0.61 1.06 ­0.08 Predicted change in service sector employment Of which: 1.09 0.13 1.92 0.42 1.51 1.26 0.98 0.76 0.30 0.39 Time shock 0.97 0.10 1.42 1.08 1.08 0.44 1.33 0.81 1.08 0.81 Start-up costs ­0.13 0.03 ­0.57 ­0.16 ­0.26 0.17 0.02 0.23 0.53 ­0.16 Access to finance 0.44 ­0.02 0.70 0.02 0.52 0.39 1.03 ­0.50 ­1.75 ­0.25 Market regulation 0.08 ­0.01 1.00 ­0.36 0.06 0.14 ­0.53 0.15 0.06 ­0.35 Tax burden ­0.26 0.04 ­0.63 ­0.16 0.11 0.12 ­0.87 0.06 0.37 0.34 Source: Bank staff calculations. Note: See annex 5.1 for the presentation of the regression model and definitions of explanatory variables. Only statistically significant explanatory variables are shown in the table. On average (considering all countries in the sample), corruption and the quality of the legal system were not significant; hence, they are not included in the table. The averages for country groups are the unweighted average of the estimated contribution of each explanatory variable to the rise in market service employment within the group. Low income CIS countries and Estonia were not covered by the regression analysis because of lack of data. a) Bulgaria, Croatia, and Romania. business environment is able to explain the employment perform- ance in the private sector. The model is less powerful at predicting the actual service sector employment change in the next wave of acces- sion countries. This results from the fact that corruption and the qual- ity of the legal system play a far more important role in this group of countries than they did for the new EU members. Poor Access to Finance, Excessive Market Regulations, Administrative Barriers, and High Taxes Are the Most Important Factors Impeding Job Creation in the Region In CEE EU-member countries (EU-8), the most important driver of private sector employment growth has been by far the relatively good access to finance. To a lesser extent, employment growth has also been facilitated by relatively liberal market regulations.13 However, a high tax burden and high administrative barriers (proxied by start-up costs) hindered job creation. For example, a better-than-average access to finance can explain the relatively good private sector employment performance of the Czech Republic, Hungary, and Latvia. Slovenia, another country where the access to finance is quite easy for entre- preneurs, would have performed much better if it had not been for the 170 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union strict labor market regulations and high tax burden. The high tax bur- den has also hampered the employment outcomes in Hungary. In CEE EU accession countries (EUA-3), the main contributors to the private sector employment growth have been relatively liberal market regulations and relatively low administrative barriers to new business formation. At the same time, private sector employment growth in these countries has been slowed by poor access to finance. For example, Romania saw a service sector employment change smaller than what would be expected, given its GDP per capita. This is mainly the result of poor access to finance. Indeed, Romania scores last in the ranking of the access-to-finance index because of its very high real interest rates, high collateral requirements, and, above all, lack of protection to creditors. In Russia and Ukraine, growth in private sector employment has been achieved mainly thanks to a relatively low tax burden, but the poor invest- ment climate has impeded job creation. Specifically, heavy market regula- tions, high administrative barriers to firm formation, and poor access to finance have all slowed the pace of private sector employment growth. Access to finance stands out as the most important determinant of the growth in private sector employment in the Region. Access to finance alone can explain about 40 percent of the whole increase in service sector employment predicted by the regression model. Improving access to finance--along with other complementary meas- ures to improve the investment climate--therefore carries the poten- tial to foster job creation. Figure 5.4 illustrates this close relationship between access to finance and private sector employment. However, other components of the investment climate matter too. Stringent market regulations discourage job creation (figure 5.5). Strict employment protection regulations may also discourage job creation, as has been the case in Croatia (box 5.3). Removing one particular constraint may therefore not be enough to engender employment growth. This is especially true, given the complementar- ities and synergies between various components of the investment climate. It is likely that countries will need to pursue a set of policies to improve the overall investment climate. Overall, countries with favorable investment climate in the Region enjoy better labor market outcomes. Lower risk and costs of doing business and fewer administrative barriers lead to a faster pace of job creation and employment growth. Using a set of indicators of the investment climate, one can successfully explain much of the varia- tion in private sector employment in the Region. The key determi- nants are access to finance, the tax burden, administrative barriers to firm creation, and market regulations. The importance of these com- ponents varies across subgroups. In European transition economies, The Investment Climate and Job Creation 171 FIGURE 5.4 Market Service Employment Is Higher in Countries with Easier Access to, and Lower Cost of, Credit Access to finance and service employment* 6 5 EST 4 GDP LVA KAZ CZE for 3 2 BGR HUN 1 SVK corrected LTU UKR 0 POL ­1 HRV ­2 ROM employment ­3 GEO SVN ­4 MDA Service ­5 1 2 3 4 5 6 7 8 Less More Access to finance index Source: Lopez-Garcia 2004. Note: The access to finance index is a summary of data on the cost of credit, collateral required, credit to private sector, and protection to creditors (for details, please refer to annex 5.1). * Controlled for GDP per capita. FIGURE 5.5 Excessive Market Regulation Hurts Job Creation Access to finance and service employment 6 5 EST 4 LVA GDP CZE for 3 2 HUN BGR 1 corrected UKR LTU SVK 0 POL ­1 HRV ­2 ROM employment ­3 SVN ­4 Service ­5 2.8 3 3.2 3.4 3.6 3.8 4 4.2 4.4 4.6 4.8 5 Fewer More Market regulation index Source: Lopez-Garcia 2004. Note: Market regulation includes credit and labor market regulations. The former refers to issues such as the percentage of deposits held in privately owned banks, the degree of competition from foreign banks and interest rate controls. Labor mar- ket regulation considers hiring and firing practices, impact of minimum wages, unemployment benefits, and union coverage. * Controlled for GDP per capita 172 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 5.3 Stringent Employment Protection Regulations May Forestall Job Destruction, but at the Same Time They Discourage Job Creation A simple comparison between Croatia and Poland illustrates the point. Until recently, dismissal costs had been extremely high in Croatia, whereas they are relatively low in Poland. On a scale from 0 to 5, the OECD index of the cost of collective dismissal was 5.0 in Croatia and 3.9 in Poland (late-1990s data). Expectedly, the job destruction rate was much higher in Poland (about 12 percent in 1999) than in Croatia (5 percent in 2001). However, Poland enjoyed a much higher job creation rate than Croatia (10 percent against less than 4 percent, respectively). As a result, the job reallocation rate, which is a measure of industrial restructuring, was significantly lower in Croatia. In Poland, more jobs were reallocated away from declining firms and industries toward expanding ones. A similar pattern--low rates of both job creation and destruction, reflecting strict EPL--also pre- vailed in the Slovak Republic. In contrast, high rates of job creation and destruction occurred in countries with relatively flexible EPL, such as Latvia and Lithuania. As shown in chapter 4, the low job reallocation rate carries a significant cost in forgone produc- tivity gains. Thus, policies to protect low-productivity, unsustainable jobs through high legislated dismissal costs are inefficient. An efficient approach is that of encouragement (that is, fostering job creation through enhancing labor market flexibility and improving the investment climate). Sources: World Bank 2001, 2002c, 2002d, 2003b, and 2003d. job creation has been supported by better access to finance (especially in the new EU-member countries) and more efficient market regula- tions. At the same time, high business and labor taxation and some administrative barriers have impeded job creation. In contrast, in Russia and Ukraine (which are likely to be representative of other CIS countries), taxation is less of a problem, but inefficient market regu- lations, administrative barriers, and poor access to finance have hin- dered job creation. Table 5.2 summarizes these differences. TABLE 5.2 The Importance of Determinants of Job Creation Varies by Subgroup in the Region Impact of investment climate components on job creation Foster job creation Impede job creation CEE, EU-member countries Access to finance Taxation, start-up costs CEE, EU-accession countries Start-up cost, market regulations Access to finance Russian Fed. and Ukraine Taxation Market regulations, access to finance, start-up cost Source: Bank staff calculations. The Investment Climate and Job Creation 173 Investment Climate: International Comparisons and Variations within the Region Perceptions from entrepreneurs, which were presented earlier in this chapter, are an important reality check of what matters to firms. How- ever, they are difficult to compare across countries because they closely depend on the environment in which entrepreneurs operate and the standards they use as references (see box 5.4). This section uses selected objective measures of the investment climate to compare the experience of the transition countries relative to other countries.14 The focus is on those aspects of the investment climate that in the previous section were found to have a strong effect on job creation. Analysis starts by comparing selected elements of the investment cli- mate in the Region to those in other regions. Then differences in the investment climate are examined across the Region's subgroups and across countries. All in all, the investment climate in the Region com- pares rather favorably to that in other regions, although in some key dimensions there is room for improvement. The study also shows that there are substantial differences in the investment climate among the Region's subgroups, as well as across countries. Generally, the cost of doing business is lowest in CEE and increases as one moves further east. The gap between the investment conditions in CEE and low income CIS countries is considerable. Investment Climate in the Region Compares Favorably to That in Other Regions, but There Is Substantial Room for Improvement On one hand, several elements of the investment climate in the Region compare well with those in other regions of the developing world. For example, the cost of market regulations (measured in management time spent dealing with them) is less than in other regions (figure 5.6). Unofficial payments to public officials to get things done (as a percentage of total sales) are smaller in the Region than in other regions (less than 2 percent of annual sales against more than 2 percent in all other regions).15 On the other hand, the comparison is less favorable in other dimen- sions of the investment climate. For example, property rights--an important element of the investment climate--are less protected in the Region than in some countries, and the costs of providing security are high (figure 5.7). The capital market is also less developed in the Region than in other areas of the world. This inhibits access to credit, which is likely, as shown earlier, to impede job creation (figure 5.8). 174 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 5.6 In Some of the Region's Subgroups, Time Spent Dealing with Government Regulations Is Still Substantial 12 10 spent 2002 8 time 6 regulations, management 4 with senior 2 of dealing % 0 Sub- East Asia Europe Latin South CEE SEE Middle Low Turkey Saharan & & America Asia income income Africa Pacific Central & CIS CIS Asia Caribbean World Bank Regions Eastern Europe and the Former Soviet Union Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individ- ual countries or subgroups of countries in the Region. FIGURE 5.7 Protection of Property against Crime Can Be Costly 3.5 3.0 2.5 2002 security, 2.0 sales, 1.5 providing total of of %1.0 Cost 0.5 0.0 Sub- East Asia Europe Latin Middle South CEE SEE Middle Low Turkey Saharan & & America East Asia income income Africa Pacific Central & & North CIS CIS Asia Caribbean Africa World Bank Regions Eastern Europe and the Former Soviet Union Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individ- ual countries or subgroups of countries in the Region. The Investment Climate and Job Creation 175 FIGURE 5.8 Firms in the Region Rely to a Lesser Degree on Capital Coming from Formal Institutions than Do Firms in Other Regions 50 from 40 coming 2002 30 capital institutions, 20 working formal total 10 of % 0 Sub- East Asia Europe Latin Middle South CEE SEE Middle Low Turkey Saharan & & America East & Asia income income Africa Pacific Central & North CIS CIS Asia Caribbean Africa World Bank Regions Eastern Europe and the Former Soviet Union Source: EBRD­World Bank Business Environment and Enterprise Performance Surveys, 2002. Note: Preliminary results from the new 2005 BEEPS are broadly consistent with those reported in this figure. However, some changes may have occurred for individ- ual countries or subgroups of countries in the Region. The Region's Subgroups Differ Substantially in Investment Climate The comparison of the Region with other areas of the world is illus- trative, but should not be taken too far. First, for more-developed countries of the Region, especially in CEE, the right comparators are developed market economies, rather than developing countries. For less-developed countries of the Region, especially the low income CIS countries, the appropriate comparators should be countries repre- senting best practice among low and middle income countries. Sec- ond, the Region is heterogeneous, with substantial differences in the investment climate among subgroups and countries. Thus, even if the Region compares favorably to other regions (on average), this is not necessarily the case for some subgroups or countries of the Region. Below, the study looks more closely at variations in the investment climate within the Region and benchmark countries of the Region against investment climate "leaders." In general, the investment climate is much better in CEE than in other parts of the Region. Consistent with earlier analysis, the cost of red tape in middle income CIS countries and SEE is much higher than in the rest of the Region and is also high by international standards (fig- 176 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 5.4 What the Official Data on Entry Barriers Do Not Show: Romania Measurement matters. Different ways of measuring the same concept can yield different con- clusions. The start-up costs in Romania are a good example. According to objective data from the World Bank, opening a business is easy in Romania: no more than six procedures are required to start a business, one can go through the process in less than a month (against 37 days on average in Europe), and the cost of the process and min- imum capital requirements are among the lowest in the transition countries. However, when one looks at similar indicators of start-up costs based on subjective opinions, in this case those of managers of large corporations, the situation is quite different. According to the evaluation given to the sentence, "Starting a new business is generally easy," by the Glob- al Competitiveness Report, Romania is one of the countries where starting a business is con- sidered the most difficult (figure A). Figure A: Subjective Perception of How Easy It Is to Start a Business 8 Easy 7 6 5 4 3 2 1 Difficult 0 Bulgaria Romania Ukraine Russian Lithu- Poland Czech Latvia Slovak Turkey EU-15 Croatia Slovenia Hungary Estonia United Fed. ania Rep. Rep. States Countries in the Region Countries outside the Region Source: World Economic Forum 2002. Note: The index is based on managers' evaluation of the sentence, "Starting a new business is generally easy." The index varies from 0 to 10, with high- er values indicating more favorable assessment. This example shows that indicators based on perceptions in some cases can be preferable to objective indicators, since they reflect de facto rather than de jure conditions. But their inherent subjectivity limits their validity. On the other hand, objective indicators, although not distorted by subjectivity, often give only a partial picture and do not reflect all the dimensions of the problem. Given the limitations of both approaches, it is best to treat them as complementary. The Investment Climate and Job Creation 177 ure 5.6). The cost of providing security (protecting firm property against crime) is especially high--also by international standards--in SEE and in the low income CIS countries (figure 5.7). Access to financing, as measured by proportion of working capital coming from formal insti- tutions, is particularly poor in the low and middle income CIS countries and SEE (figure 5.8).16 But, interestingly, it is still worse in Turkey--the only nontransition economy in the Region. Finally, corruption is a par- ticular problem in the low income and middle income CIS countries. Countries of the Region that are poorer and less advanced in market reforms still have a long way to go to improve the investment climate to make it more conducive to firm growth and job creation. On Most Key Dimensions of Investment Climate, the Transition Countries Lag behind World Leaders The multivariate analysis presented earlier showed that low business start-up costs; easy access to, and low cost of, finance; and liberal mar- ket regulations are keys to the growth of the market services sector, and thus for employment. In all those dimensions of the investment cli- mate, most of the transition countries lag behind world leaders. The gap is relatively small for the best-performing CEE countries, but is wide for most of the CIS and SEE countries. Below, the study bench- marks countries of the Region against selected developed and best-per- forming developing economies. The former are a natural benchmark for middle income countries of the Region, whereas the latter are a more adequate benchmark for low income countries of the Region. Opening a Business Is Costly, Especially in SEE and Some CIS Countries In most of the transition countries, the start-up costs are higher than in the United States (where it is least costly to start a business), the Euro- pean Union (EU-15), or the dynamic developing countries, such as Korea, Malaysia, or Thailand (see annex table A5.2).17 Among the tran- sition countries, start-up costs are (on average) lowest in CEE, especially in the Baltic States. Costs of opening a business are also moderate in most CIS countries, although Belarus and Ukraine are clear outliers. Start-up costs are high in SEE by international standards (figure 5.9).18 The number of start-up procedures is limited to three in industrial countries with a liberal approach to entrepreneurship, such as Ireland or Sweden. In the Region, the most liberal country in this respect is Hun- gary, where one needs to go through five procedures. But in Belarus, where opening a business is most cumbersome, there are 19 necessary 178 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 5.9 Starting a Business Is Not Easy in Many of the Region's Countries United States Romania Latvia Armenia Lithuania EU-15 average Kazakhstan Kyrgyz Rep. Russian Fed. Uzbekistan Bulgaria Low income CIS average Georgia CEE average Slovenia Moldova Croatia Macedonia, FYR Turkey Czech Rep. Slovak Rep. Middle income CIS average Serbia & Montenegro Poland Azerbaijan SEE average Albania Average of Hungary the Region Ukraine Bosnia & Herzegovina Belarus 0.0 2.0 4.0 6.0 8.0 10.0 12.0 High Low Start-up cost index, 2003 Low income CIS CEE Middle income CIS SEE United States and EU-15 Turkey Source: World Bank, Cost of Doing Business Database. Note: The index goes from 0 to 10, 0 being the most expensive. The start-up cost index was constructed by rescaling the four dimensions of start-up costs provided by the CODB database: number of procedures, total time, cost and minimum capital. Then a simple unweighted average of the four components was calculated. procedures. Similarly, in many developed and developing countries, there is no minimum capital requirement to start a business. Within the Region, only in Azerbaijan is no start-up capital required. In Ukraine, the minimum required capital is almost five times the gross national income (GNI) per capita, which can effectively limit firm creation.19 The Investment Climate and Job Creation 179 Hungary exemplifies the point that even in countries with overall good investment climate, there is room for improvement. Procedural costs associated with opening a business are the lowest in Hungary, compared with those in other countries of the Region. At the same time, the monetary costs associated with registering a business are the highest. Thus, although (on average) start-up costs are moderate in Hungary, the country would benefit from improving the incentives for firm creation by lowering the registration costs. To conclude, most of the transition countries can lower adminis- trative barriers to firm entry and thus foster job creation. Opening a business can be made easier even in CEE countries, but the biggest room for improvement is in SEE and in some of the CIS countries. Access to Finance Is Difficult, Especially in the CIS Access to finance is substantially more difficult in the Region than in the EU-15 or in developing countries with developed credit markets, such as Korea or Malaysia, although better than in Turkey. Within the Region, access to finance is easier in European transition economies (especially in CEE) and more difficult in the CIS (especially in the low income CIS countries) (figure 5.10). But even in CEE, the credit mar- ket is notably less developed than in the EU-15. Domestic credit to the private sector (relative to GDP) is much less in CEE than in EU- 15, and the interest rate spread is significantly larger.20 To illustrate, domestic credit to the private sector represents more than 100 percent of GDP in the European Union and some develop- ing countries, such as Korea and Malaysia, but less than 30 percent in CEE. In middle income CIS countries, the ratio goes down to about 16 percent, and in low income CIS countries, it barely exceeds 10 per- cent.21 This comparison shows that there is substantial room for developing the credit market and improving access to credit in all of the transition countries, but especially those in the CIS. Better access to credit, in turn, is going to support firm entry and growth and, ulti- mately, job creation. Moreover, better access to credit will provide incentives for firms to move from the informal to the formal sector, which in turn may lead to productivity gains. Regulations Often Add Heavy Burden to Firms' Operation There is still much room in the transition countries to make markets more liberal. In all of these countries, markets are more regulated than in the United States, and most of them are more regulated than in the EU-15. Again, European transition economies, especially new EU mem- 180 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 5.10 Job Creation in the Region Is Likely to Be Hampered by Difficult Access to Credit EU-15 average Latvia Slovenia Macedonia, FYR Estonia Hungary Poland Bulgaria Czech Rep. CEE average Azerbaijan Slovak Rep. Croatia Lithuania Ukraine Moldova Russian Fed. Middle income CIS average Belarus Low income CIS average Armenia Kazakhstan Kyrgyz Rep. Average of the Region Turkey Georgia Romania 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Low High Access to finance index, 2002 Low income CIS CEE Middle income CIS SEE EU-15 Turkey Source: World Development Indicators. Note: The index goes from 0 to 10, 0 being the worst. The index is the combination of the following variables: ratio of do- mestic credit provided by deposit money to GDP, interest rate spread and real interest rate, ratio of deposit coverage to GDP and the index of creditors' protection (from the World Bank's Doing Business Database). bers, tend to have less-regulated markets than the CIS and also Turkey do (figure 5.11). Thus, given that excessive market regulation has been shown to limit job creation, lessening the regulatory burden on business activity in the Region is yet another way to encourage job creation. The Investment Climate and Job Creation 181 FIGURE 5.11 Markets in the Region Tend to Be Overregulated Russian Fed. Turkey Average of Albania the Region Slovenia Slovak Rep. Lithuania Ukraine Poland Romania Bulgaria Czech Rep. Latvia EU-15 average Hungary Croatia Estonia United States 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 Most Least Market regulation index, 2001 CEE Middle income CIS SEE United States and EU-15 Turkey Source: Economic Freedom of the World 2003. Note: The indexes go from 0 to 10, 0 being the most regulated. The market regulation index is an unweighted average of the credit market and labor market regulation indexes provided by the Economic Freedom of the World's Annual Report (Gwart- ney and Lawson 2003). Corruption Adds Up to Heavy Regulations to Raise Costs of Doing Business Corruption is closely related to excessive regulations, and it has been shown to be a severe obstacle to business formation and growth, as well as to job creation. Given that corruption looms large in the Region, its nega- tive impact on employment is likely to be substantial.22 Within the Region, 182 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union the least corrupted are the CEE countries that are EU members. But even there, corruption is much more widespread than in the EU-15. In other European transition economies, corruption is a severe problem, and it is apparently rampant in the CIS countries (figure 5.12). But corruption is also a major issue in Turkey, which is a nontransition economy. Fighting corruption and its sources is thus critical to encouraging job creation. FIGURE 5.12 Corruption Is High in the Region Azerbaijan Moldova Kazakhstan Low income CIS average Georgia Ukraine Albania Romania Russian Fed. Uzbekistan Turkey Middle income CIS average Czech Rep. Latvia Slovak Rep. Croatia Average of the Region Bulgaria Poland CEE average Belarus Lithuania Hungary Estonia Slovenia EU-15 average United States 0.0 2.0 4.0 6.0 8.0 10.0 Less More Perceived corruption index, 2002 Low income CIS CEE Middle income CIS SEE United States and EU-15 Turkey Sources: Transparency International 2002; Bank staff calculations. Note: The index goes from 0 (best) to 10 (worst). The country ranking responds to the degree of corruption perceived to ex- ist among public officials and politicians. It draws from 15 different polls and surveys from nine different institutions. The Investment Climate and Job Creation 183 Summing Up: Promoting a Better Investment Climate to Foster Job Creation Despite major structural reforms, the investment climates in many transition economies still suffer from significant shortcomings that negatively affect firms' incentives to invest and create more and better jobs. Therefore, better employment outcomes in the Region depend on further progress in structural reforms affecting the quality of the investment climate. Many investment climate challenges are common across the Region, although their importance varies by subgroup and country. The three challenges that stand out are the following: · Strengthening macroeconomic stability and reducing economic and regulatory policy uncertainty. · Lowering taxes, especially on labor, and simplifying tax adminis- tration. This is associated with public expenditure reforms, partic- ularly with reforms to the pensions and health insurance systems. It also includes cutting programs that are not cost-effective and improving the efficiency in the delivery of public services. In addi- tion, tax administration needs to be improved to reduce the bur- den on firms (for example, the costs associated with tax inspections) and to reduce the scope for corruption. · Improving access to, and lowering the cost of, credit. This is associ- ated with the reforms of the banking system, such as through enhancing competition. Investment climate issues are also more pronounced in some regions than in others. The key elements of a reform agenda by sub- group are summarized here. The Investment Climate Reform Agenda in CEE Investment climate in the CEE EU-member countries is by far the most favorable among the transition countries. Nonetheless, even the most-advanced CEE countries lag behind world leaders in many dimensions of the investment climate, and there is significant scope for improvement. For example, starting a business in CEE is still con- siderably more difficult than in the United States or in the EU-15. Three issues are more of a constraint to firm growth in CEE EU-mem- ber countries than in other subgroups: · High tax rates · Labor regulations 184 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union · Skills of available workers In the EU-accession countries, costs of doing business tend to be higher than in CEE EU-member countries. Additional investment cli- mate issues that need to be addressed include the following: · Corruption · The efficiency of the legal system and conflict resolution · Licenses and operating permits The Investment Climate Reform Agenda in SEE SEE countries are less advanced in investment climate reforms than are other European transition economies. There is considerable scope for reducing the risk and costs of doing business, as well as lowering administrative barriers and cutting the red tape. Reform priorities include addressing the following: · Security of property rights · High costs of starting a business · Excessive regulation of business activities, including customs and trade regulations · Anticompetitive or informal practices · The quality of the legal system and conflict resolution · Corruption The Investment Climate Reform Agenda in the CIS In most dimensions of the investment climate, the CIS countries, both middle income and low income, lag behind European transition economies. The institutions of a market economy are less developed and market-oriented reforms are less advanced than in European transition economies. Therefore, the risks associated with doing busi- ness are substantially higher, as are administrative barriers and asso- ciated bureaucratic harassment. Corruption is widespread, adding to the costs of doing business. To bridge this divide and to foster the cre- ation of more-productive, formal sector jobs, the middle income CIS countries need to focus on addressing the following obstacles: · Excessive market regulations, administrative barriers to business activity (for example, licensing and operating permits), bureau- The Investment Climate and Job Creation 185 cratic harassment (for example, frequent and lengthy inspections), and associated corruption · Business start-up costs (especially in Belarus and Ukraine) · Underdeveloped credit market and access to finance Investment climate constraints in the low income CIS countries are still more profound. In addition to the ones listed above, these countries need to overcome more-fundamental obstacles, such as the following: · Security of property rights · Underdeveloped infrastructure, including access to electricity · Administrative barriers, including cumbersome customs and trade regulations Reforms to Foster Job Creation Require Identifying Key Constraints to Firm Growth and the Active Involvement of the Business Community Improving the investment climate is a complex task. Many of the reforms are interlinked and complementary and thus require appro- priate sequencing. For example, fighting corruption requires dereg- ulating business activities and lessening the discretionary power given to the bureaucracy. It also requires removing administrative barriers (for example, reducing the number of permits and licenses, simplifying customs procedures, and limiting the number of agencies that have the right to carry out inspections). Moreover, some reforms require substantial investments (for example, improvements in infrastructure, in the efficiency of the legal system, and in the insti- tutions responsible for enforcement). Given the scarcity of resources, governments need to select priorities and focus on addressing the constraints that are the most binding. Improving the investment cli- mate is a continuous process because the environment changes and new challenges emerge. Although investment climate reforms can be difficult and costly, the long-term social costs of abstaining from reforms are likely to be still higher. If firms' entry and growth is inhibited by high costs of doing business, job creation and employ- ment will suffer. The above-identified subgroup reform priorities define broad areas where the key constraints lie. Country-specific constraints and reform priorities need to be identified, using more-detailed information on the quality of business environment. Such information can be 186 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union obtained from specially designed investment climate surveys or from focus groups of employers.23 The most important channel for identifying obstacles to job cre- ation is ongoing government consultations and dialogue with the genuine representation of the business community, including the small private firms. Such dialogue has already been carried out in a number of the transition countries under different institutional forms. Bosnia and Herzegovina launched its so-called Bulldozer Initiative to involve the private sector in reforms. Latvia established a Steering Committee for Improvements of the Business Environment, which comprises representatives from key ministries and from associations of local firms, as well as foreign investors (World Bank 2003d). In Poland, a dynamic employers' confederation, representing the new private sector, is vocal in publicly articulating business concerns and actively takes part in a dialogue with the government on investment climate issues. Thus in many of the transition countries, the institutional struc- ture for voicing business concerns has already been created. The chal- lenge is to overcome bureaucratic inertia and to mobilize political support for these necessary reforms that hurt the interests of those who benefit from the status quo. Annex 5.1: Regression Analysis of the Impact of the Investment Climate on Employment in the Region To capture the interaction between shocks and institutions, the fol- lowing simple model (equation 5.1) was estimated: uit = ci + dt(1+ Âg Xi )+ OEit j j (5.1) j where u is the unemployment or employment rate in country i at time t, c are the country dummies, d are time dummies (that is, unidentified shocks assumed to be equal across countries24), and X is the time-invariant value over the period of the institution j in country i. What the model is trying to capture is the way institutions shape the response of countries to a similar shock. All transition economies of the Region have been hit by a similar shock at a similar time, but their labor markets have responded differently. The objective is to The Investment Climate and Job Creation 187 determine whether this is the result of the institutional settings of the countries, and if so, which institutions had the strongest impact. To answer these questions, the model uses as explanatory variables the interactions of various institutions with some proxy to the shock. This is the most general specification, because no specific shocks are imposed; it is only assumed that a number of shocks over time, of equal dimension across countries, have hit the economies. Hence, the model allows one to isolate the impact of the (different) institutions from that of the shocks. To estimate the coefficients corresponding to each of the institutions in the model, an unbalanced panel of 28 countries of the Region over 14 years (from 1989 to 2002) was used. Unfortunately, only one year of information for most of the institutional variables was available. The list of institutions (explanatory variables) considered in the regression model is as follows:25 · Start-up costs. A composite index, constructed by using the World Bank's Doing Business Database, 2004. · Exit costs. As above. · Access to finance. A composite index, using the following variables: ratio of domestic credit to GDP from World Development Indicators 2004 (World Bank 2004c), ratio of deposit coverage to GDP (IMF), and creditors protection index (Doing Business Database, 2004). · Quality of the legal system. A composite index, combining indicators of contract enforcement (Doing Business Database 2004) and sub- jective evaluation of the quality of the legal system provided by the Economic Freedom of the World 2003 Annual Report (Gwartney and Lawson 2003). · Market regulations. This is an unweighted average of the credit mar- ket and labor market regulation indexes, provided by the Economic Freedom of the 2003 World's Annual Report (Gwartney and Lawson 2003). · Corruption. Transparency International Corruption Perception Index 2003 (www.transparency.org). · Tax burden. Tax revenue as a percentage of GDP, from World Devel- opment Indicators, 2004 (World Bank 2004c). The results of the estimation of the regression are shown in table A5.1. 188 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union TABLE A5.1 Estimation Results Type of regression: Common shocks across countries assumed Service employment (corrected) Unemployment Variable Coefficients t-statistics Coefficients t-statistics Shocks (given average institutions) 0.8% 0.85% Start-up cost index ­0.29* (­1.45) 0.09 (0.26) Exit cost index ­0.003 (­0.28) ­0.006 (­0.24) Finance index 0.41** (2.95) ­0.18 (­0.42) Legal system index ­0.04 (­1.13) ­0.05 (­0.86) Mk. Regulation index ­0.64** (­2.45) 1.73* (1.53) Corruption index 0.06 (0.33) 1.57** (2.37) Tax burden ­0.08 (­1.11) 0.43** (2.35) Observations 106 104 Adjusted R-Squared 0.95 0.77 Source: Bank staff calculations. Note: Nonlinear least square estimation *significant at 15% **significant at 1% Annex 5.2: Investment Climate Indicators in Selected Transition and Developing Economies TABLE A5.2 Investment Climate Indicators in Selected Transition and Developing Economies Starting a business GDP per capita Domestic credit/ Cost Min. capital at PPP, 2002 GDP ratio Number of Duration (% GNI (% GNI (dollars) (2002) procedures (days) per capita) per capita) Armenia 3,230 6.9 10 25 7 4.5 Brazil 7,450 35.5 17 152 11.7 0 Hungary 13,070 35.3 6 52 22.9 86.4 Korea, Rep. of 16,960 115.6 12 22 17.7 332 Kyrgyz Republic 1,560 4.2 8 21 11.6 0.6 Malaysia 8,500 146.1 9 30 25.1 0 Moldova 1,600 17.6 10 30 18.6 24.6 Poland 10,450 28.8 10 31 20.6 237.9 Romania 6,490 8.3 5 28 7.4 0 Russian Federation 8,080 17.6 9 36 6.7 5.6 Serbia and Montenegro -- -- 11 51 9.5 120.3 Thailand 6,890 102.5 8 33 6.7 0 Ukraine 4,800 18 15 34 17.6 113.9 Vietnam 2,300 43.1 11 56 28.6 0 Sources: World Development Indicators (2004); Cost of Doing Business Database (2005). Note: -- = data not available. The Investment Climate and Job Creation 189 Hiring and firing workers Rigidity of Difficulty of Rigidity of Difficulty of employment Firing costs hiring index hours index firing index Index (weeks) Armenia 17 40 50 36 17 Brazil 67 80 70 72 165 Hungary 11 80 30 40 34 Korea, Rep. of 11 60 30 34 90 Kyrgyz Republic 33 40 40 38 21 Malaysia 0 0 10 3 74 Moldova 33 60 70 54 21 Poland 11 60 30 34 25 Romania 78 60 50 63 98 Russian Federation 0 60 20 27 17 Serbia and Montenegro 28 0 40 23 21 Thailand 67 40 20 42 47 Ukraine 33 80 80 64 94 Vietnam 44 40 70 51 98 190 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Notes 1. Lopez-Garcia (2002) shows that the contribution to unemployment of barriers to entrepreneurship doubled in a sample of OECD countries, when worker reallocation was taken into account. 2. Such information is gathered in the investment climate surveys. These are the result of a partnership between the World Bank and country counter- parts for the collection of data from large random samples of firms in more than 50 countries. The surveys collect objective quantitative data on meas- ures of the investment climate, plus information on firm performance. This allows linking the investment climate indicators with performance to understand their impacts on productivity, investment decisions, and employment decisions. The surveys were launched in 2001, with about 20 new surveys conducted each year. So far, more than 26,000 formal firms have been interviewed, half of them with fewer than 50 employees. 3. This is consistent with the results in Pissarides, Singer, and Svejnar (2003), which provides evidence that constraints on obtaining external financing and the high cost of this financing are two of the top five con- straints cited by managers of small and medium enterprises in Bulgaria and Russia. 4. Ranking is calculated according to the average score of each constraint. 5. Labor regulations are efficient if they support an efficient allocation of labor resources (see Estreicher and Schwab [2000], especially chapter 5). 6. This is also surprising, because on the books labor regulations are signifi- cantly more strict in the CIS and SEE than in CEE; however, the discrep- ancy between de jure and de facto strictness of employment regulations can be accounted for by differences in enforcement (see chapter 6). 7. Regressions are carried out using generalized ordered logit models. Per- ceptions for each type of constraint are estimated as a function of firms' characteristics (size, age, sector, innovating history, ownership, exports, imports, and foreign ownership) and country's income level. 8. Multivariate regression analysis was used to estimate the independent contribution of different components of the investment climate to employment. See annex 5.1 for the details, including the regression model, data used, and results. 9. The coefficient of determination (adjusted R2) is 95 percent, meaning that the investment climate variables used in the model explain a large portion of variation in the market service employment rate across the sample of the transition countries. 10. To control for the economic development, a regression of the service employment rate was run against the GDP per capita. The residuals from that regression represent the service employment rate not explained by the development of the economy. The residual is what is called the "ser- vice sector employment rate controlled for GDP per capita." 11. This analysis is based on the argument that the institutional setting of a country shapes the impact that otherwise-similar shocks have on its labor market performance. The first economists to propose this idea were Michael Bruno and Jeffrey Sachs (1985). But it was only after the publi- cation of the Blanchard and Wolfers (2000) model of equilibrium unem- The Investment Climate and Job Creation 191 ployment that the interaction of shocks with institutions became a pop- ular explanation among labor economists for the unemployment evolu- tion of developed countries. 12. Lithuania is an exception, but this result needs to be treated with cau- tion: comparisons of employment before and during the transition are inaccurate because of changes in methodology and data sources. 13. The indicator of market regulations is based on indexes of credit and labor market regulations. 14. One set of indicators corresponds to employers' answers to "objective" questions as part of the Investment Climate questionnaire (sample ques- tion: In a typical week, what percentage of senior management's time is spent dealing with requirements imposed by government regulations, including dealing with officials, completing forms, and so forth?). A sec- ond set of indicators reflects de jure conditions. 15. Based on the World Bank investment climate survey data. 16. Research on financial developments in the Region points to the poorly developed financial system in low income CIS countries as a major impediment to their sustained growth and emphasizes the "great divide" between the low income CIS countries and other subgroups, especially CEE countries (Nicoló, Geadah, and Rozhkov 2003). 17. Korea, Malaysia, and Thailand were chosen as comparators for the tran- sition countries because their level of GDP per capita is similar to that in the middle income countries of the Region. For example, Korea has GDP per capita similar to that in Hungary and Poland, Malaysia to that in Bul- garia or Russia, and Thailand to that in Romania or Ukraine (see annex table A5.2 for data on GDP and selected investment climate conditions). 18. Because of the space constraint, the figures do not show countries out- side the Region (except for the EU-15, which is a benchmark). Data for countries referred to in the text come from the same source as those used to construct the charts. 19. The data come from the World Bank Doing Business Database, 2003. 20. Interest rate spreads affect the cost of credit and thus investment. The level of spreads is determined by funding, operating and regulatory costs, rents accruing from banks' market power, and credit risk (Nicoló, Geadah, and Rozhkov 2003). 21. The data come from the World Bank (2004c). 22. The level of corruption is significantly negatively correlated with employment in the market services sector. However, the level of corrup- tion tends to decrease with the level of economic development, and for this reason, the corruption variable turned insignificant in the employ- ment regression reported in table 5.2. In other words, corruption matters for employment, but its impact is difficult to separate from that of eco- nomic development. 23. World Bank (2005a) provides an example of using focus groups and employer-based surveys for identifying constraints to job creation. 24. The first period is left out, so it becomes the constant. Therefore, the country dummies can be interpreted as the unemployment rate in the first period. 25. Detailed information on the construction of composite indices used in the regression is provided in Lopez-Garcia (2004). CHAPTER 6 Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation Enhancing job opportunities requires an adaptable labor market in which firms have the right incentives to create more jobs and work- ers receive support when coping with labor mobility. Improving the adaptability of the labor market requires interventions in several areas: wage determination and labor taxation, employment protec- tion legislation, and social safety nets for the unemployed. Although most transition economies of the Region have carried out significant labor market reforms over the past decade to improve labor market adaptability, there is substantial room for improvement. In most countries in the Region, employment relations are still over- regulated, discouraging hiring. Much of the labor market flexibility comes from nonenforcement. In many countries, especially in SEE and the CIS, wage bargaining is not conducive to swift wage adjust- ments to changing demand-and-supply conditions. High taxes on labor, especially in CEE, discourage labor demand. Income support to the unemployed can be provided in a more efficient way, to provide protection to displaced workers and to encourage job reallocation, which is of particular importance in the CIS. Further labor market reforms are thus necessary to enhance the ability of the transition economies to cope with various shocks. Also, the reforms of the pen- sion and health insurance systems need to be continued to make pos- sible the reduction in labor taxation. 193 194 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union The chapter is organized as follows. Section 1 provides a brief dis- cussion of the roles of labor market policies and institutions. Section 2 focuses on wage determination. Section 3 examines employment pro- tection legislation in the Region. Section 4 deals with labor taxation. Section 5 analyzes active and passive labor market policies. Section 6 suggests direction for labor market reforms for the Region's subgroups. The Role of Labor Market Policies and Institutions Labor market policy and institutions are important, both in their potential impacts on labor market adjustment and in the incentives that they give firms (see box 6.1). They have direct and indirect impacts on wages, which in turn can be expected to influence both the structural level of (un)employment and the speed of labor market adjustment to shocks. Policies and institutions can impact wage and employment adjustment in a variety of ways, including truncating the lower end of the wage distribution (with impacts on the level and structure of employment), affecting firm incentives for hiring and fir- ing, promoting (re)entry to employment for laid-off workers and new entrants, and affecting labor supply incentives. The interaction of dif- ferent labor market institutions in specific country settings further complicates the analysis of the impact on labor market outcomes.1 In a context of major economic restructuring such as in the transi- tion economies of the Region, the government has a role to play in balancing policies that facilitate the allocation of workers to more-pro- ductive jobs with policies that are designed to help workers cope with necessary adjustments. Most of the Region's countries have under- taken major reforms of labor market institutions during the transition, although the extent of reform has varied by subgroup and by specific element of the institutional framework. Despite reforms, excessive institutional rigidities appear to have contributed to unemployment among the low-skilled and in depressed regions and to have affected patterns of labor adjustment. However, they cannot be considered the primary cause of low labor market performance in the Region, which appears to be driven more by low labor demand, an unfavorable busi- ness environment, skills mismatch, and other factors. The Divergent Paths of Wage Determination during the Transition In market economies, a higher unemployment rate will have a damp- ening effect on wages: as outside options become less favorable, Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 195 BOX 6.1 The Role of Labor Market Policies and Institutions: Some International Evidence Social Dialogue and Wage Bargaining As illustrated by the experience of Western Europe and countries like Poland and South Africa, unions have a crucial role to play in promoting macro and structural reforms, as well as political openness and democracy. However, depending on the economic and social environment in which unions operate, collective bargaining has been found to affect the flexibility of wages and the performance of firms. Union wage markup (a premium over the wage of similar workers in nonunionized firms) is an important indicator of union bargaining power and ability to influence labor market outcomes. The evidence suggests that union wage markups are small in industrial countries, but high in countries that have weak competition in output markets and large rents. For example, wage pre- miums are high in Bulgaria and Malaysia (5­20 percent) and South Africa (10­24 percent), but low in Korea (2­4 percent). High union wage premiums obviously benefit union members at the expense of the unemployed; however, wages set at above the competitive level lower labor de- mand and lead to less employment. Unions also have other impacts. For example, in Mexico, unions have attempted to protect low- skilled jobs at the expense of higher productivity. In Guatemala, unionization is associated with lower productivity of coffee farmers. In Brazil, the improvement in productivity and profitability resulting from the greater participation of workers in certain aspects of company management was enhanced in unionized companies because unions facilitated communication between management and workers. The level at which negotiations take place (firm, sector, or country) is also important. Experi- ences of developing and industrial countries have shown that sectors (industries) do not coordi- nate their wage demands, decreasing aggregate and relative wage flexibility.a To alleviate these effects, some industrial countries have reinforced the coordination between the different levels of bargaining. For example, in Denmark, Ireland, and the Netherlands, the basic wage increase is fixed through nationwide agreements, while further increases consistent with a firm's per- formance are left to firm-level negotiations. Following the systems in countries like Australia, New Zealand, and the United Kingdom, others have chosen to reinforce firm-level bargaining, which is more likely to reflect firms' and workers' performance. Minimum Wages Minimum wages have been found to have disemployment effects, small in aggregate, but sig- nificant among low-productivity workers. For example, in industrial countries, where minimum (continues on the following page) 196 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 6.1 (continued) wages tend to be relatively low, they had a modest impact on low-technology firms and the em- ployment of low-productivity workers. In Colombia, for every percentage point rise in the mini- mum wage, employment has been estimated to fall by 0.15 percentage points. In Indonesia, the significant increases in provincial minimum wages led to a fall in employment in small firms, but not in larger firms, although it was also associated with an increase in the number of workers below the minimum wage. At levels close to the average wage (as in several low-income coun- tries), many private firms, especially those in low-technology activities, cannot afford to comply. In middle-income countries, the minimum wage is generally about half the average in the formal sector. Its coverage and enforcement tend to be low, but its impact on low-productivity firms and jobs can be large. In Latin America, for example, the largest proportions of workers who earn less than the minimum wage are found in countries where it is comparatively high: in Paraguay (where most workers earn less than two-thirds of the minimum wage), Nicaragua (40 percent of workers below the minimum), and Colombia (25 percent). Employment Protection Legislation The protection offered to permanent workers and the conditions for temporary employment vary considerably across and within regions. Countries in Latin America, Eastern Europe, and Central Asia tend to offer the most employment protection for permanent workers. By contrast, English-speaking industrial countries and East Asian economies have the lowest statutory pro- tection. Evidence from several countries that recently reformed their stringent labor regulations suggests beneficial effects. By moving their employment protection legislation closer to the standards of the European industrial countries in the 1990s, Colombia and Peru experienced a higher response of employment-to-output growth, with speedier employment adjustment but also positive employment effects. The reform in Colombia also contributed to increased com- pliance with labor legislation by lowering the costs of formal production. During the past decade, Italy and Spain have eased their restrictive firing regulations, which has led to sizable positive ef- fects on employment. Source: Pierre and Scarpetta 2004b. a. Each bargaining unit is strong enough to push wages above productivity, but at the same time, it is vulnerable to other units' wage strategies without being able to influence them. workers' bargaining power is decreased. Under Communism, except for the former Yugoslavia, unemployment was essentially nonexist- ent, and wages did not respond to firms' performance. The negative relationship between unemployment and wages, commonly called the "wage curve," appeared during the 1990s in several CEE countries (for example, Bulgaria, Hungary, Poland, and Slovenia), indicating that wages have become responsive to local labor market conditions.2 But no such relationship exists in countries Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 197 that have been less advanced in their market reforms (for example, Romania). A wage curve also exists in Turkey, except for women and young workers, pointing to the specificity of their labor force partici- pation and working conditions.3 In the central-planning period, because of the passive roles of unions and managers, there was no wage bargaining as such, but rather centralized wage setting, which mirrored price setting throughout the economy.4 Although the coun- tries of the Region began the transition with very high union mem- bership, the monolithic official unions exerted no independent influence on wages, being integrated into management structures and considered "transmission belts" for the authorities. Enterprise managers were also part of the planning apparatus, focusing on pro- duction and full employment objectives, rather than profits. The evolution from a common starting point of government wage setting and passive unions and management has been mixed across the Region's countries during the transition. What are the patterns and explanations for this? What is the evidence of the unions' impact on labor market outcomes? Falling Union Bargaining Power in Most Transition Countries Union density and bargaining coverage5 have significantly declined in nearly all of the countries of the Region during transition, leading to diminishing impact of unions on wages. Under Communism, vir- tually all employees were members of the unions, and this was the case in most countries of the Region at the onset of the transition. As a new private sector developed, union presence and impact on wages declined in all transition countries, although there are clear subgroup variations in the extent of the decline. Figure 6.1 provides available information on density and coverage rates for a range of countries of the Region in the early 2000s. The trends by subgroups are as follows: · Official density rates in the waged sector in Belarus, Russia, and Ukraine and other CIS countries remain high by OECD standards (estimated between 55 and 90 percent), although effective cover- age rates are likely to be lower, given large informal sectors and inflated membership numbers in most countries.6 · By contrast, average density and bargaining coverage rates fell more steeply in CEE, from more than 85 percent density and more than 90 percent coverage in most cases in 1989 to an average of around 30 percent density and less than 40 percent bargaining coverage by 198 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 6.1 Density and Bargaining Coverage, Early 2000s Poland Lithuania Estonia Hungary Latvia OECD average Albania Czech Rep. EU-15 average Turkey Slovak Rep. Slovenia Romania Macedonia, FYR Kazakhstan Serbia & Montenegro Kyrgyz Rep. Armenia 0 10 20 30 40 50 60 70 80 90 100 % of total workforce Density Coverage Sources: EIRO (2004); ILO (2002) for CIS; OECD (2004a); Micevska (2004) for Albania, Serbia and Montenegro, and FYR Macedonia. the early 2000s. There is also substantial variation within CEE, the Baltic States being down to around 20 percent coverage and Slove- nia maintaining 100 percent by the early 2000s.7 · SEE countries also witnessed substantial declines in union density and bargaining coverage, though estimates from different sources vary. Reasonable estimates of average union density are 22 percent in Albania, 45 percent in FYR Macedonia, and 51 percent in Serbia and Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 199 Montenegro.8 Estimates of coverage rates have proven challenging to obtain. In some countries, such as Bosnia and Herzegovina (BiH),9 they remain very high in the formal sector, because of the strong role of general and sectoral collective agreements. Although more research is needed, this pattern may be typical of former Yugoslav countries, reflecting the residual influence of worker self-manage- ment and the role of trade unions prevalent under the former regime. In both BiH and Serbia and Montenegro (SAM), automatic extension of collective agreements to all workers was only eliminated with the reform of labor legislation over the past five years. Despite declining membership, union bargaining strength in spe- cific sectors has remained relatively strong. Examples include natural monopolies, transport, heavy industry, and mining, which have been more likely to stay in public ownership. In contrast, expanding sec- tors comprising largely private enterprises, such as trade, business, and personal services, have tended to have much lower union pres- ence. Similarly, union presence in newly created small and medium enterprises is low. The scale of the decline in union bargaining power in the transi- tion countries reflects several influences: · Structural change in the formal sector--specifically the emergence of large private sectors (with typically much lower density and coverage rates) and the growth of service sectors at the expense of traditional industrial sectors (where unions are more strongly rep- resented) · A pattern--stronger in CEE--of increased segmentation and frag- mentation of unions and union confederations, often with con- flicting political and economic agendas10 · The increased level of informal sector activity The differential legal position of unions across the Region is an important factor in influencing their behavior. At one extreme, non- reformers, such as Turkmenistan, retain mandatory membership in the successor unions, with automatic union fee deduction from salaries. Others, such as Uzbekistan, inhibit the formation of inde- pendent unions through high membership thresholds for union reg- istration. Azerbaijan controls union activism through legal prohibition of political activity.11 Other countries, such as Estonia, have no such limitations on union formation or activity, but have eliminated the privileges for unions in bargaining that continue to exist in many other countries of the Region (see box 6.2). 200 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 6.2 Wage Bargaining in Estonia: A Radical Reformer Because of explicit policy choices and a more general rejection of the Soviet inheritance, Esto- nia has seen the most fundamental changes in wage bargaining of any country of the Region. Together with Latvia and Lithuania, Estonia has witnessed the most dramatic decline in union density and bargaining coverage in the Region. By 2003, union density had fallen to around 16 percent, with 22 percent coverage. The private sector had only 8 percent density and around 19 percent coverage, in contrast to density and coverage rates of 28 and 31 percent, respectively, in the public sector. Although some sectors such as banking, construction, and services have very low unionization, others such as light industry, energy, education, and transport have strong unions, with much higher density. The small enterprise sector, which accounts for around 90 percent of private sector employment, has very limited unionization. The low density and coverage, especially in the private sector, are partly the result of a lack of legal privileges for trade unions. This is reflected in the Collective Agreements Act of 1993, which abandoned the former union monopoly to conclude agreements and to represent work- ers at the enterprise level. There are no restrictions as to who may be elected or in which pro- portion to workers. This principle has been upheld not only for collective agreements, but also in later legislation on worker representation, such as the Trade Union Law of 2000. Estonian laws do not restrict union activity, but they do not stipulate any special bodies for codetermina- tion or representation that are reserved for unions, the right to demand the conclusion of col- lective agreements, or any obligation on employers to enter collective bargaining. Divergent Patterns of Change in the Wage-Bargaining Process across Subgroups Changes in unionization and in the role of unions in the economy have often radically affected the process of wage setting in transi- tion countries. As seen in chapter 2, all of the transition countries have experienced widening wage differentials resulting from transi- tion, suggesting that, in aggregate, wage determination has become more flexible. However, the aggregate patterns mask substantial cross-country variation in the evolution of wage-bargaining arrangements: · At one end of the spectrum, countries such as Belarus and Turk- menistan retain the inherited system, with wage setting in the for- mal sector being highly centralized and entirely controlled by the Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 201 Tripartite national bargaining began in 1992, with a Socio-Economic Council formed in 1999. Al- though the Council has a wide mandate to discuss economic issues, its only effective influence is on the minimum wage, the tax-free income threshold, and the level of unemployment bene- fit. For the public sector, negotiations focus only on the aggregate wage fund allocation, with substantial flexibility for employers to determine relative wages of workers beyond the mini- mum. The exception to date has been the education sector, where a more traditional wage co- efficient structure has been retained in the face of a strong union. However, unions and em- ployers accept the government view that budget and macroeconomic criteria are the primary determinants of wage outcomes. This can be seen in the large fluctuations in the real minimum wage during the second half of the 1990s and in the wage and pension freezes in the wake of the financial crisis in Russia. However, in 2001, there was a tripartite agreement on an increase in the ratio of minimum wage to average wage (MW/AW) from 32 percent in 2003 to 41 percent by 2008. Recent efforts of the government have focused on introducing performance-related pay, a move that has generally been resisted by the unions. Beyond the national-level negotiations, private sector bargaining in Estonia happens almost en- tirely at the firm level. Given low union density and limited rights of representation, individual employment contracts have a major role in determining the wages of workers. As a result, there are often high wage differentials within firms, and unions include many workers with limited in- dividual bargaining leverage (such as low-skilled or Russians whose language skills limit their work opportunities). Source: Patzwaldt 2004. government. A highly centralized model is also used in Slovenia, although the institutional mechanism is very different, leaving an important role for social partners in wage determination. · At the other end of the spectrum, wage bargaining in most CEE countries has become highly decentralized in the growing private sector, and there are efforts to introduce more flexibility into pub- lic sector pay determination. · Most of the other countries of the Region fall somewhere between these two extremes. Overall, Belarus, Russia, Ukraine, and other CIS countries still exhibit a strong legacy of socialist era wage deter- mination in the formal sector. SEE countries represent an interme- diate stage, with wage determination that is neither decentralized nor as dominated by government as in the CIS. 202 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Compression of the Wage Schedule in the Public Sector Key factors that distinguish the divergent evolution of wage bargain- ing in the transition countries include the size of public sector employment and the overall importance of government in economic decision making. The extent to which countries have reduced the role of the state in wage bargaining is thus the distinguishing feature of different subgroups of countries in the Region. To assess the divergent subgroup developments, it is necessary to look first at public sector wage setting, because this has been the base- line wage determination mechanism from which countries have diverged at different speeds. Most of the transition countries have retained a coefficient-based system of wage determination in the pub- lic sector, under which wage adjustments cascade from adjustments in the base. This remains the model in Belarus, Russia, Ukraine, and other CIS countries and in SEE, with the public sector wage schedules exerting strong influence on formal private sector wage determina- tion.12 The coefficient system also applies in CEE public sector wage determination, although Estonia has eliminated the tariff schedule above the minimum in the public sector (except in education). A key feature of such coefficient systems in most of the transition countries has been a compression of the wage schedule. Whereas the wage multiple from top to bottom of the public sector wage schedule in OECD countries has been around six to nine times, coefficient scales in the transition countries have typically been more com- pressed (for example, 2.3 in Romania, 2.8 in Moldova, 3.6 in Hun- gary, and so forth).13 This is an issue that many of the Region's governments have examined as part of civil service reforms, but as of the end of the 1990s, a significant share of countries still had high public sector wage compression.14 Spillover Effects in the Private Sector The primary significance of the coefficient system in economic terms in the transition countries is not its direct impact via public sector employment. The share of public sector employment in most of the transition countries is not very large--albeit higher than the mar- ket-benchmark for countries at the same level of income per capita (see chapter 4)--and public sector wages have remained fairly low, despite some recent increases, especially in the CEE countries.15 How- ever, the rigidities of the coefficient system assume wider significance in a labor market where there is a large state-owned sector that fol- lows the same system, or where the formal private sector mimics the Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 203 public sector wage-setting mechanism. A strict coefficient system with strong wage compression clearly limits the scope for adjusting wages to relative productivity levels and regional labor market characteris- tics. To the extent that this characterizes much of formal sector wage determination (public and private), the economic impacts could be expected to be negative. In practice, the coefficient system brings varying degrees of rigidity in wage determination across subgroups. The trends in wage deter- mination in different subgroups of the Region are strongly associated with the residual share of public sector and SOE employment in total wage employment and the overall degree of government interven- tion in the economy.16 · In CEE, greater wage flexibility has been mainly the result of a sub- stantial decentralization in formal private sector wage setting, cou- pled with product market liberalization and associated competitive pressure. The most notable exception is Slovenia, which retains 100 percent bargaining coverage and highly centralized wage set- ting, even in the private sector.17 In CEE countries, where wage- bargaining changes have progressed the most (for example, the Baltic States, the Czech Republic, Hungary, and Poland), enter- prise-level wage bargaining developed spontaneously, largely in line with growth in the private sector and in the face of rather weak trade union influence at the firm level. For most CEE coun- tries, wage bargaining takes place primarily at the enterprise level. The individual work contract is also a very important factor in the most-advanced reformers such as Estonia and Latvia.18 By con- trast, the role of sectoral bargaining is generally limited in CEE countries, with the exception of the Slovak Republic and Bul- garia--with around 40 percent of the workforce covered by sec- toral collective agreements--and Slovenia.19 · By contrast, in Belarus, Russia, Ukraine, and other CIS countries, the government remains a much more pervasive force in formal sector wage determination. The formal private sector has more closely followed the public sector's lead in wage determination. Unions have not (for the most part) evolved into independent forces in bargaining; they have retained a dominant role, but are almost entirely subject to the determination of wages by the gov- ernment. Employers often continue to place employment reten- tion above productivity gains, commonly mimicking public sector wage setting in the formal private sector. Although the situation varies across countries (for example, Kazakhstan has active inde- 204 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union pendent unions), the weakness of unions and employers is a shared feature of Belarus, Russia, Ukraine, and other CIS coun- tries. As a result, they remain price takers from government in relation to formal sector wages and focus their efforts on negotiat- ing soft budget constraints. At the same time, a large informal sec- tor (accounting for more than 40 percent of employment) has provided very flexible wage determination for a large share of the workforce and is an important source of large wage disparities in the CIS. In Russia, flexible wage setting coexists with the use of the coefficient system in the private sector. Although most firms use firm-level wage scales, a significant fraction of private firms has adopted the public sector coefficient system.20 · There is less information on wage-bargaining developments in SEE countries. Generally, in countries of the former Yugoslavia, unions have been relatively influential, in part because of the inheritance of self-management. At the same time, the role of the government in formal sector wage setting remains strong, either directly (as in BiH and SAM), or less directly through the wage pull of the SOE sector. This in part reflects a weak role of employers in central or sectoral wage bargaining in some SEE countries. In Bosnia and Herzegovina, the trade union confederations also have a strong influence on wage setting, in the face of still underdeveloped employer organizations and the strong influence of governments. Overall, the importance of general collective agreements in SEE continues to be strong, which both limits the impact of EPL reforms and prolongs wage-setting practices such as automatic seniority increases and large location bonuses. There is also a strong role for sectoral unions and bargaining, the presence of employers as an independent voice again being negligible. Serbia and Montenegro and FYR Macedonia provide examples of more-fragmented union movements, although a general collective agreement and sectoral collective agreements have a strong role in both countries. There is also a strong wage pull from the SOE sector to other sectors, so that one has yet to see the desired level of decentralization in wage set- ting to develop. Bargaining Structures in Transition Economies Are Still Evolving, and Their Labor Market Impact Is Underresearched Overall, the link between wage-bargaining systems and labor market outcomes in the Region remains underresearched, in particular in CIS and SEE countries. With this caveat in mind, several patterns Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 205 appear regarding bargaining arrangements and labor market out- comes in the Region: · Changes in union density per se do not have a major impact on employment and unemployment rates, which is consistent with evidence from OECD countries. This is confirmed in multivariate regressions for CEE, Russia, and Ukraine, which control for coun- try characteristics, including the depth of transitional reforms (using the EBRD transition index). The results for (un)employ- ment outcomes also hold for job flow outcomes.21 The finding is also confirmed in regression analysis on pooled data for OECD, CEE, and SEE countries for the second half of the nineties.22 · The typical negative correlation between bargaining coverage and earnings dispersion seen in OECD countries does not appear to hold at the cross-country level in transition economies. The most striking examples of this come from Russia and other CIS countries, which combine high union density rates with very large earning dispari- ties (see chapter 2). However, the conclusion must be interpreted with caution in light of the perhaps misleadingly high density rates from Russia, Ukraine, and other CIS countries. Further research is needed to consider country and subgroup fixed effects. · State-owned firms (which have higher union density) consistently have a greater share of firm surplus devoted to wages, reflecting insider bargaining power relative to both the privatized and de novo private firms. In the medium term, these firms may lack rein- vestment capital; this may lead in time to slower growth and even- tually lower employment. This pattern has been found in analyses of Bulgaria, Poland, and Russia.23 · Evidence from multivariate analysis for CEE, Russia, and Ukraine suggests that both highly centralized or coordinated systems and more decentralized systems are associated with higher employ- ment and lower unemployment.24 In other words, sectoral bar- gaining tends to deliver the worst employment outcomes, a pattern also observed in the OECD countries.25 Slovenia provides an example of a flexible system whereby firms are allowed to deviate from the minimum wages associated with skills categories to reflect the performance of the firm or individual workers or both. The resulting within-firm wage flexibility appears to reduce excess turnover of workers (that is, improve the stability of employment) and to improve the match quality of workers (that is, improve the degree of adequacy between workers' characteristics and the 206 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union requirements of their jobs).26 These findings suggest a microlevel trade-off between wage differentials and employment. Minimum Wages Vary a Great Deal across the Transition Countries The minimum wage (MW) is an important element of wage determi- nation. Particularly at the low end of the wage distribution, it may price out low-skilled, often young, workers. Minimum wages in the transi- tion countries exhibit a clear divide between CEE and SEE on the one hand, where they average close to 40 percent of the average wage (AW) in 2002, and the CIS (except Ukraine) on the other hand, where they are only around 20 percent (and for several countries less than 10 per- cent) of the average wage (see figure 6.2). These ratios place CEE at the high end of the range of minimum wages by international standards and CIS at the low-to-modest end. In Ukraine, where the minimum wage is comparatively high, there is evidence that it is not enforced (see figure 6.3 and World Bank [2005c]). In SEE, both BiH and FYR Mace- donia have high minimum-wage-to-average-wage ratios. FIGURE 6.2 Minimum-Wage-to-Average-Wage Ratio, 2002 (percentage) Lithuania Bosnia & Herzegovina Macedonia, FYR Ukraine Slovenia Hungary Slovak Rep. Czech Rep. Latvia Poland Romania Bulgaria Estonia Armenia Moldova Russian Fed. Belarus Azerbaijan 0 10 20 30 40 50 60 70 Ratio of minimum to average wage, % Sources: TransMonee and Bank staff calculations. Note: BiH = FBiH only. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 207 FIGURE 6.3 Minimum Wage in Ukraine Accounts for a High Percentage of the Market Wage of Low-Skilled Workers, but It Is Not Enforced Median 0.005 0.004 density 0.003 distribution 0.002 Wage 0.001 0 0 200 400 600 800 1000 1200 Hourly wage (hrywnas) Sources: Ukrainian Longitudinal Monitoring Survey, 2004; World Bank 2005c. Note: The vertical line indicates the minimum wage (205 hrywnas in June 2004). Low-skilled workers are workers classi- fied as "unskilled" according to the ULMS. Negotiated Minimum Wages Are Often Much Higher Than the Nationwide Minimum Wage Minimum wages for some sectors are higher--often considerably so--than the national minimum wage. This is typically a product of bargaining at the sectoral level. Examples indicate that the increment of sectoral MWs over general wage floors can be substantial: in Bul- garia, 52 percent of sectoral collective agreements provide for mini- mum wages higher than the economywide minimum, with negotiated minimums 10­80 percent higher than the national mini- mum; in the Federation of Bosnia and Herzegovina (FBiH), the min- imum wage in the 2004 telecommunications sector collective agreement was double the economywide minimum wage (itself a high 55 percent of the average wage) and equivalent to 110 percent of the economywide average wage.27 Although the principle of var- ied minimum wages is desirable, the potentially negative employ- ment effects of ratcheting MWs above already high national MWs need to be considered. Minimum Wages Sometimes Set Social Benefit Minimums In a number of the transition countries--most notably in Belarus, Russia, Ukraine, and other CIS countries--the role of the official min- 208 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union imum wage was perhaps more significant as the base around which certain social benefits were set. Many of these countries have abol- ished such linkages. Kazakhstan provides a case where this explicit welfare function of the minimum wage has been delinked, with a "base numerator" set instead as the base for determination of certain social transfers. The Czech Republic, however, provides a good example where such delinkage was done without sufficient consideration for the value of the minimum wage relative to base social welfare benefits, with negative implications for work incentives.28 From 1992 to 1998, various social benefits were linked to the minimum wage. Because of fiscal concerns, the government kept the minimum wage at a low level to avoid increased transfers obligations. However, this was changed in the late 1990s, with the introduction of a "subsistence wage" distinct from the minimum wage, which assumed the function of the benchmark for setting the level of the transfers.29 If low-income workers earn less than this amount, the state tops up their income to the subsistence wage level. With the link to the transfers system, peo- ple receive the subsistence wage whether they work or not. Such an example raises the issue of using uniform national minimums in social benefit determination when regional labor markets may have very different market wages on offer. What appears as a reasonable benefits numeraire in the capital city may well have significant work disincentive effects in depressed regions. Another notable feature of the minimum wage in some countries is its linkage to the revenue system. Croatia and FYR Macedonia con- stitute an interesting case. These countries do not have any econo- mywide minimum wage as such, but they require that social benefit contributions be paid on the basis of a set minimum amount. Although such an explicit link may be sensible from a public finance perspective, it is questionable whether it is prudent from a labor mar- ket perspective.30 In other countries, the relationship to public finances is not explicit, but disproportionately high shares of workers cluster on declared wages at or just above the minimum wage (with evidence of additional undeclared incomes above the minimum), cre- ating incentives to sustain a high minimum wage to sustain tax rev- enue. For example, in Latvia, almost one-third of private employees earn declared income within 20 percent of the official minimum wage, although the average wage is around three times the minimum and public sector wages exhibit a more regular distribution.31 This aspect of minimum wage policy in the Region requires further research. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 209 Minimum Wages Have Stronger Disemployment Effects in Backward Areas and among the Low-Skilled International evidence shows that high national minimum wages negatively affect employment chances of low-productivity workers, including workers with little labor market experience (for example, the youth), with low skills, and in backward regions (O'Keefe 2004). Large increases in minimum wage levels in the Czech and Slovak Republics (of 70 and 50 percent, respectively--but from very low lev- els) have led to limited job losses and to large increases in firm wages during 1999­2002.32 However, for specific subgroups of workers, such as the low-skilled or the youth, minimum wages represent a much higher proportion of the reservation wage.33 For example, in Poland, the minimum wage accounts for more than 80 percent of the wage received by low-skilled workers in high unemployment regions. Because the wage of low-paid workers is close to the minimum, they are the first to be affected by any increase in the minimum wage. Evi- dence shows that there is a negative relationship between MW/AW ratios and low-paid employment. This suggests that high minimum wages contribute to pricing marginal workers out of the workforce. Some evidence on the negative effects of minimum wage increases on employment rates among unskilled workers is found in regression analyses from Estonia and Hungary.34 It is also consistent with high youth unemployment in the transition countries, particularly given the general absence of separate youth minimum wages (except in Poland). Employment Protection Legislation Remains Strict despite Reforms, Although Enforcement Is Variable Employment protection legislation (EPL) is a key labor market insti- tution in ensuring secure and decent work. At the same time, it is important to balance its positive impacts on workers with jobs with the needs of the unemployed and those entering the labor market, whose prospects of being hired may be negatively affected by overly strict job-security provisions. There are also trade-offs from an eco- nomic viewpoint between the positive features of strict EPL (such as incentives to provide productivity-enhancing training to workers) and the need for sufficient ease of job turnover to promote efficient allocation of labor and contribute to macrolevel productivity improvements.35 Given the complex balance to be struck in the design and enforcement of EPL, it is perhaps not surprising that 210 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union research on its contribution to labor market outcomes, even in OECD countries, has reached contrasting findings.36 Analysis of EPL in the transition countries is further complicated by several factors, including the relatively fast pace of change in EPL during the tran- sition, the unusually high need for labor reallocation, and often weak compliance. Nearly All Transition Countries Have Liberalized Employment Protection Legislation, Especially Those Moving into the EU At the beginning of the transition, workers in the Region were among the most protected in the world. This was reflected in the socialist-era EPL, under which dismissals were very difficult, full-time permanent employment was the norm, and workers were entitled to a wide range of workplace benefits and protection. At the same time, protec- tions such as severance pay and notice periods upon dismissal were largely unnecessary (because dismissals were prohibited, except for severe misconduct). Former Yugoslavia, which had very generous provisions on both, was a notable exception. Nearly all transition countries subsequently carried out significant reforms of EPL, though at varying points during the transition. The direction of initial reforms has been toward greater flexibility in labor relations, including in hiring and firing (for example, lower direct dis- missal costs or removal of trade union veto on dismissals), promoting temporary and part-time employment, allowing for opting out of col- lective agreements, and so forth. Most CEE countries achieved their first wave of EPL reforms early in the transition, while other regions such as SEE and the CIS often delayed reforms until the late 1990s. Many countries have continued to adjust reformed legislation, although in some cases reducing flexibility in later revisions (for example, Romania's second wave of EPL reforms after 1998 and Hun- garian reforms effective from 1997). EU Transition Countries Have Focused More on Liberalizing Temporary Contracts Although the overall trend in EPL reforms is therefore clearly toward greater flexibility, the progression has not been linear. For 10 transi- tion economies, information has been gathered on the evolution of employment protection legislation since the beginning of the 1990s (figure 6.4). These data suggest that transition economies that were approaching entry into the European Union (for example, Hungary, Poland, and the Slovak Republic) have been more aggressive in liber- Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 211 alizing EPL than others whose accession was further in the future (for example, Bulgaria and Romania) or not foreseen in the short term (for example, Ukraine). This suggests that EU accession could have played a disciplining role in promoting reforms.37 It should be stressed, however, that most of the reforms have been concentrated on liberalizing temporary contracts by extending maxi- mum duration of fixed-term contracts or by legalizing temporary work agencies. In other words, following the experience of many continental European countries, EU transition economies have liber- alized at the margin, lowering the costs of temporary employment, with more modest reductions in the protection of permanent work- ers. This may have contributed to the observed duality in the labor market whereby vulnerable groups, including youth, women, and the unskilled, churn from one job to another without being able to build a proper career, while those who maintain a regular contract still enjoy a significant degree of job protection.38 Despite Reforms, Regulations Remain Fairly Restrictive in Most Countries Despite the importance of the many reforms, EPL in the Region remains (on average) stricter than in other regions in the world (fig- ure 6.5). In particular, the stringency of EPL for permanent workers in the Region is (on average) higher than that in any other region except South Asia. However, there is also considerable variation in the Region in the stringency of employment protection, particularly regarding temporary employment (figure 6.6A). There is no clear subgroup pattern to the relative levels of regular and temporary EPL (figure 6.6B). However, in some cases, the differ- entials are very pronounced, in both directions. For example, Bosnia and Herzegovina, Bulgaria, and Latvia mainly limit temporary employment and regulate permanent employment less, while Alba- nia, Estonia, Kazakhstan, Poland, and Slovenia provide significantly more protection to permanent employment. EPL Is Not Fully Enforced in Many of the Transition Countries There is evidence of weak enforcement of EPL in most of the transi- tion countries, except for CEE.39 This appears to be particularly the case in the CIS and a significant factor in much of SEE. This suggests that the effective rigidity imposed by EPL varies from that indicated by simple comparison of legislation (see also box 5.4). The point is supported by the available evidence on settlement of labor disputes 212 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 6.4 Employment Protection Legislation in EU-8 and Other Selected Countries during the Transition a. Employment Protection Legislation, EU transition countries 6 5 restrictive 4 More Employment Upper extreme of Legislation 3 75% 2 Median Protection 25% 1 restrictive Restrictiveness Lower extreme Less 0 1991 1996 2001 b. Employment Protection Legislation, Bulgaria, Romania, Russian Fed., Ukraine 6 5 restrictive 4 More Employment of Legislation 3 2 Protection 1 restrictive Restrictiveness Less 0 1991 1996 2001 Source: Haltiwanger, Scarpetta, and Vodopivec 2003. Note: The line in the middle of the shaded box represents the median. The shaded box extends from the 25th percentile to the 75th percentile, the so-called interquartile range (IQR). The lines emerging from the box extend to the upper and low- er adjacent values. The former is equal to the 75th percentile plus 1.5 times IQR, while the latter is the 25th percentile mi- nus 1.5 times IQR. Points more extreme than the adjacent values, if any, are outside values and are individually plotted. The employment protection legislation (EPL) index ranges from 0 (the least regulated) to 6 (the most regulated). The index measures regulations for both temporary and permanent employment contracts. (for example, in Russia) and, for some countries, in the differences between the legislative difficulty of hiring and firing and the ease of both reported by employers (for example, in Romania and Ukraine).40 In Belarus, Russia, Ukraine, and other CIS countries, there is widespread evidence of weak enforcement of EPL. It is seen Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 213 FIGURE 6.5 Transition Countries Have More-Stringent Regulations on Hiring and Firing than Do OECD Countries 0.60 0.50 regulated 0.40 Most index firing 0.30 and 0.20 Hiring regulated 0.10 Least 0.00 Sub-Saharan Eastern & South Asia Latin America Developed Middle East East Asia Africa Central & countries & & Europe Caribbean North Africa Pacific Temporary employment regulations Regular employment regulations Overall employment regulations Source: Bank staff calculations based on Doing Business Database (2005). A higher index indicates more-stringent regu- lations; see annex 6.1 for the construction of the indexes. most starkly in the massive scale of wage arrears in the subgroups during the 1990s and is supported by employer surveys. For example, in Moldova, despite strict EPL, employers identify labor regulations as their least important problem, which is consistent with the country's high job destruction rate. By contrast, CEE countries have less strict EPL (on average), but are considered more likely to enforce it (for example, evidence from Bulgaria, Lithuania, and Poland indicates that the costs of court challenges to dismissals act as additional disin- centives to firing workers). Finally, there is evidence (for example, from Lithuania and Poland) that EPL is more strongly enforced in large firms, which typically have a stronger union presence. Table 6.1 shows a tentative characterization of the transition coun- tries according to the strictness of EPL and its enforcement. It is indicative only, but illustrates that the effective impact of strict EPL in the Region may be less than comparisons of labor legislation would suggest, with only Slovenia having both strict EPL and strict enforce- ment. Although the characterization is crude, it indicates that coun- tries in the top right of the table need to focus more on credibly enforceable laws as opposed to "paper protections," which at best protect a limited share of formal sector workers. 214 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 6.6 There Are Significant Differences within the Region's Countries on EPL A. Across subgroups 0.70 0.60 regulated Legislation 0.50 Most 0.40 Protection 0.30 0.20 regulated Employment 0.10 Least 0.00 Low SEE CEE Middle Turkey income CIS income CIS B. Across countries Romania Ukraine Croatia Latvia Bosnia & Herzegovina Uzbekistan Moldova Azerbaijan Belarus Kyrgyz Rep. Macedonia, FYR Lithuania Slovenia Bulgaria Georgia Armenia Serbia & Montenegro Czech Rep. Hungary Russian Fed. Slovak Rep. Estonia Poland Albania Kazakhstan 0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Least regulated Most regulated Employment Protection Legislation Temporary employment regulations Regular employment regulations Overall employment regulations Source: Bank staff calculations based on Doing Business Database (2005). Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 215 TABLE 6.1 EPL and Enforcement Typology Flexible EPL More restrictive EPL Very rigid EPL Weaker enforcement Albania, Kazakhstan Armenia, Georgia, Azerbaijan, Belarus, Russian Federation, Bosnia and Serbia and Montenegro, Herzegovina, Kyrgyz Turkey Republic, Moldova, Ukraine, Uzbekistan Intermediate enforcement Bulgaria Croatia; Macedonia, FYR; Romania Stronger enforcement Czech Republic, Latvia, Lithuania, Estonia, Hungary, Slovenia Poland, Slovak Republic Source: Bank staff calculations. Note: (i) Strictness defined as whether country EPL index below world average EPL index in Doing Business 2005, up to 20% higher and more than 20% higher; (ii) strength of enforcement based on share of informal sector employment. The degree of enforcement is proxied by the size of the informal economy; the higher the informal economy the weaker the es- timated degree of enforcement (Schneider and Klinglmair 2004, and country studies where available). For Russia the EPL index was corrected toward stricter EPL following comments from Russian experts. Even with Varying Enforcement, Strict EPL Has Had Some Negative Impacts on the Labor Market Despite varying enforcement, available analysis indicates that there remain costs from strict EPL in the Region. First, strict EPL may pro- mote informality, self-employment, and small firms (which often fall outside EPL). The positive correlation between strict EPL and the share of informal sector employment can be seen in figure 6.7. A sim- ilar positive association holds between the strictness of EPL and the share of self-employment.41 Given that informal employment is fre- quently low-productivity, strict EPL can hamper productivity growth. Second, strict EPL gives rise to less favorable employment out- comes in transition economies. The relationship between EPL and employment outcomes is difficult to identify statistically. One reason is that the system is still moving toward equilibrium, and thus the steady state (un)employment rate is not known. For example, cross- country regressions for the late 1990s suggest insignificant correlation between EPL and aggregate unemployment in countries from OECD, CEE, and SEE (Cazes and Nesporova 2003; Micevska 2004), although they find that stricter EPL increases the risk of long-term unemploy- ment. However, these static studies do not capture the effects of the significant institutional changes that transition countries carried out over the 1990s, nor do they use a steady state unemployment rate as a dependent variable. In contrast, more complex, cross-country, pooled regression analy- sis (which adds the time dimension to the relationship) finds negative 216 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 6.7 Informality Tends to Be Higher in Countries with Strict EPL 60 population 50 age 40 working of 30 % as 20 force 10 labor 0 Informal 30 35 40 45 50 55 60 65 70 75 80 Least regulated Most regulated Employment law index One country in the Region Sources: Doing Business in 2004 for EPL; Schneider (2002) for the size of the informal sector. effects of strict EPL for CEE, Russia, and Ukraine.42 In countries with low coordination among partners in wage bargaining,43 strict EPL is found to weaken labor market outcomes, but in countries with a high degree of coordination, the effect is less marked. Significantly, the effects of EPL on employment rates are weaker in the Region than in OECD economies, while they are stronger for unemployment rates. This is consistent with strict EPL being unable to prevent job destruc- tion, but still contributing to less formal sector job creation. It is also consistent with employer surveys, in which firms that are expanding are more likely to report EPL as a major obstacle in their operations, while nonexpanding firms are less likely to do so. In addition, greater job security for those with jobs appears to come at a cost for the unemployed and new entrants in the labor market, so that (re)entry into formal sector employment is negatively affected. This "insider-outsider" impact on formal sector employment rates would appear to undermine the claim that strict EPL is "always good for workers" in the aggregate. Moreover, evidence from the Region indicates that reforms of EPL have not been associated with higher income inequality and that there are, if anything, higher income dis- parities in countries with stricter--but weakly enforced--EPL.44 A further important finding of regression analysis is that EPL has a significant association with the pace of labor reallocation. For CEE, Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 217 Russia, and Ukraine, regression analysis reveals that a more-stringent EPL is associated with lower job creation, lower job destruction, and thus lower job reallocation.45 This in turn is likely to feed through to impacts on productivity. These findings are consistent with more- detailed country case studies (for example, in Croatia, Bosnia and Herzegovina, and the Slovak Republic). This supports the intuition that strict EPL dampens job flows, even with weak enforcement. Empirical findings on employers' perceptions of the impact of EPL also suggest that firms that are in the process of upgrading or devel- oping new product lines and potentially need additional skilled work- ers are more affected by strict EPL than others (see chapter 5). Finally, regression analysis on pooled data from OECD, CEE, and SEE countries for the second half of the 1990s finds that strict EPL influences the composition of unemployment: the strictness of temporary EPL provisions is positively correlated with higher youth and female unemployment, stricter EPL increases the risks of long-term unemployment, and stricter temporary EPL discour- ages labor supply.46 Taxes on Labor Taxes on labor may impact equilibrium wages; this in turn may affect equilibrium employment. The extent of the employment impact is affected by the level of competition in the labor market. In particular, the impact of high taxes on labor use depends on the extent to which there is "tax shifting" of labor taxation onto the cost of labor (and hence labor demand), as opposed to shifting it onto take-home pay (and hence labor supply). High Taxes on Labor Use, Especially in CEE Countries The tax wedge on labor in the transition countries is high--in many cases, very high. At the same time, like most labor market institu- tions, it exhibits considerable variation across countries, both in aggregate level and composition (figure 6.8). Overall, the tax wedge on labor in CEE and Turkey is very high by any standard, while labor taxation in CIS countries is lower (on average). The tax burden on labor is at an intermediate level in SEE countries, but remains high by international standards.47 On average, the tax wedge on labor in Belarus, Russia, Ukraine, and other CIS countries is more than 14 percentage points lower than that in CEE and Turkey, while for SEE, it averages around 11 percentage points lower than that in CEE and 218 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Turkey (figure 6.9). The differences in payroll and effective personal income tax rates are likely to be amplified in practice by variable tax enforcement capacity across the Region. For CEE, the comparison with OECD countries is of interest (figure 6.9).48 The regional average tax wedge is around 8 percent lower than the OECD average (excluding CEE OECD countries). However, the averages mask considerable variety in both groupings. The tax wedge on labor in CEE countries is noticeably higher than the OECD average.49 Once OECD countries are disaggregated into high, medium, and low labor tax wedge groups, CEE averages are 10 per- FIGURE 6.8 The Region's Tax Wedge on Labor, 2003 Armenia Kazakhstan Tajikistan Slovenia Albania Azerbaijan Russian Fed. Belarus Kyrgyz Rep. Ukraine Uzbekistan Georgia Serbia without Montenegro Macedonia, FYR Latvia Croatia Estonia Lithuania Poland Turkey Czech Rep. Romania Hungary 0 10 20 30 40 50 60 70 % of gross wage Low income CIS CEE Middle income CIS SEE Turkey Sources: OECD (2004b) for Czech Republic, Hungary, Poland, and Turkey; Bank staff estimates for other countries. Note: Tax wedge for nonagricultural worker on average wage without dependents. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 219 FIGURE 6.9 Tax Wedge on Labor, the Region, and OECD, Early 2000s 70 60 50 wages 40 gross 30 of % 20 10 0 OECD OECD OECD OECD CEE SEE CIS Turkey high- middle- low- all income income income countries countries countries countries Sources: OECD (2004b) for OECD and CEE OECD, except Slovenia (based on 2001); Bank staff estimates for others (using 2003). Note: Tax wedge calculated for nonagricultural AW worker without dependents in the Region and production AW worker without dependents in OECD. OECD groupings exclude CEE OECD: (i) low (less than 40% wedge) includes Australia, Ice- land, Ireland, Japan, Korea, Mexico, New Zealand, Portugal, Switzerland, UK and USA; (ii) mid (40-50%) includes Cana- da, Denmark, Greece, Luxembourg, Norway, Spain and Sweden; (iii) high (51+ %) includes Austria, Belgium, Finland, France, Germany, Italy and Netherlands. centage points lower than the average for the OECD group with the highest average tax wedge. Hungary is the only country of the Region that would fit firmly within the high labor tax OECD group. In con- trast, CIS has an average tax wedge on labor equal to the average for the low-labor-tax OECD group, which includes Korea and Mexico, which are the only OECD members with comparable income levels. The SEE average is between the low- and medium-labor-tax OECD groups and somewhat below the overall OECD average. For six CEE countries, there is also information on the evolution of the tax wedge during the 1990s (figure 6.10). Contrary to the pat- terns of employment protection and the decline in the generosity of unemployment benefits (see below), the tax wedge has increased (on average) in these countries, with a reduction in the cross-country dis- persion of tax wedge rates. The surge in other social expenditures, predominantly pensions, has led to further increases in the fiscal bur- den on the use of labor. In this context, it is not surprising that many employers in the Region found the tax burden to be one of the most serious constraints to operating their firms (see chapter 5). Although the high tax wedge in the Region is important, there is also considerable evidence--particularly in the earlier part of the 220 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union FIGURE 6.10 Surge in the Tax Wedge in EU Transition Countries during the 1990s EU transition countries* 70 Upper extreme 60 75% 50 wedge, Median wages 40 25% tax 30 gross of 20 Lower extreme % Effective 10 0 1991 1996 2001 Source: Haltiwanger, Scarpetta, and Vodopivec 2003. Note: See note to figure 6.4 for details on the construction of the graph. The tax wedge on labor use is the ratio of total taxes--income taxes and social security contributions--over wages plus employers' social security contributions. The ef- fective tax wedge is based on tax, social security receipts and the wage bill, as opposed to the notional tax wedge which is based on average tax and social security rates for the average production worker. transition, and before reforms of tax administration in a number of countries--of partial tax compliance, particularly in low-income countries and those affected by conflict. The combination of high taxes on labor and the low capacity of the public sector to enforce compliance points to ambiguous effects on aggregate (un)employ- ment. However, the relatively stronger enforcement capacity of CEE countries, combined with their very high labor taxation, could raise labor costs and affect employment. This is true at least for countries with low-productivity potentials whose wages cannot fully compen- sate for the high taxes because of minimum-wage rules. High Taxes on Labor Use Are Associated with Negative Labor Market Outcomes Evidence from OECD countries suggests that the pass-through of taxes onto wages depends on wage-bargaining systems.50 In coun- tries with sectoral wage bargaining, where there is a lack of coordina- tion between social partners, the pass-through onto wages is more limited because workers have fewer incentives to consider the effect of their bargaining on employment. In such countries, labor taxes are therefore found to have greater negative effects on labor market out- comes. Similar effects may be at work in the transition countries. What are the findings on the impact of labor taxation on labor market outcomes in the Region? Available cross-section and panel regression analyses on the impact of taxes on labor in the Region, although limited, find impacts similar to those in OECD countries.51 Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 221 This is reinforced by information on employer perceptions reported in chapter 5, identifying high taxes as a major constraint on firm operations. The high tax wedge on labor, especially in CEE, is associ- ated with lower employment rates, higher unemployment rates, higher long-term unemployment rates, and higher youth unem- ployment rates. It is clear that to foster job creation in the longer term, a number of countries in the Region need to consider significant reductions in their tax burden on labor. A challenge facing the transition countries is the need for a reduction in payroll taxes from a labor demand view- point amid social expectations of maintaining social insurance sys- tems at current levels. This raises important reform issues in pensions and health insurance. Given demographic trends, further reforms of social insurance programs will be needed simply to avoid substantial increases in payroll taxes across the Region. The Role of Passive and Active Labor Market Programs High and persistent unemployment in many transition economies creates political demand for the use of passive and active labor poli- cies. The case is reinforced by equity arguments, given the signifi- cantly higher-than-average poverty rates among households with unemployed heads in the transition countries (for European transi- tion economies, mean expenditures of unemployed-head households is only 68 percent of the average, but in the CIS around 85 percent) and by political economy arguments at a time when significant reforms are necessary and public acceptance is crucial. However, because of poor incentive effects and moral hazard, passive labor mar- ket policies may create economic inefficiencies and entail large gov- ernment expenditures. Active labor market programs (ALMPs) may also be very costly, and their success in helping the unemployed needs to be evaluated to justify such spending. The developments in both passive and active programs for the unemployed have been rapid in the Region and have included major adjustments in many countries during transition. These interventions are a crucial part of the menu of policies that governments can use to help displaced workers and other disadvantaged groups find new jobs. This section draws from available empirical evidence to outline which types of policies and designs have been more successful. It argues that if there is adequate administrative capacity, ALMPs can be used to help disadvantaged worker groups and improve the matching function of the labor market. But they hardly increase aggregate 222 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union employment. ALMPs are costly, and thus monitoring and evaluation are necessary to ensure that they have the net impact (for example, significantly improving the job-finding chances of program partici- pants) that justifies the cost. The level of spending and the mix of pro- grams will depend on what individual countries can afford in their financial and administrative capacity. All Transition Economies Introduced OECD-Style Unemployment Insurance, and They Have Reduced Their Generosity to Contain Expenditures The need for major reallocations of labor during transition led to the introduction of unemployment insurance systems in all transition countries by the early 1990s. They adopted OECD-type mandatory unemployment insurance (UI) schemes, financed in nearly all cases from payroll taxes, with some CEE countries introducing unem- ployment assistance following expiration of unemployment bene- fits (UB). As in the OECD, unemployment benefits in the Region are typi- cally earnings-related, though subject to floors and ceilings. In recent years, the range in replacement rates for UI has been 35­60 percent of the previous wage. This is equivalent to 60­100 percent of the national minimum wage and 10­45 percent of the national average wage in CEE. The effective replacement rate in some countries was also considerably below the legislated rate because of payment arrears (for example, in Russia, the low income CIS countries, and Serbia and Montenegro) and benefit ceilings (for example, Croatia). The replacement rate has been adjusted downward in a number of CEE, SEE, and CIS countries during the 1990s in the face of fiscal constraints and significant flows into unemployment. It had reached an average of just more than 40 percent of the average wage by the end of the decade in EU transition countries (figure 6.11). The decline has, however, been accompanied by a widening of the replacement rates across these countries, with some (for example, Poland) moving to a flat level and others maintaining the earnings-related system. Although replacement rates are broadly comparable to EU and OECD averages, it is important to note the significant variation in rates across countries and the implications of the typical practice in the Region of designing programs with benefit floors related to national benchmarks. This can result in replacement rates relative to average wages in depressed regions that can be very high, risk- ing employment disincentives. This issue is a concern in CEE, in particular. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 223 FIGURE 6.11 Unemployment Benefit Replacement Rates Have Declined in EU Transition Countries during the 1990s EU transition countries* 90 80 Upper extreme % 70 75% rate, 60 50 Median 40 25% 30 20 Replacement 10 Lower extreme 0 1991 1996 2001 Source: Haltiwanger, Scarpetta, and Vodopivec 2003. Note: See note to figure 6.4 for details on the construction of the graph. The replacement rate is the ratio of unemployment benefits over previous wage. It refers to a worker earning an average wage while at work and with sufficient work experi- ence to claim benefits for the maximum duration. Benefits duration in the transition countries falls within a range comparable to that in OECD countries--between six months and a year at the end of the 1990s. In some SEE countries, unemployment benefits duration reaches 24 months. Across the Region, there was a convergence in benefits duration during the 1990s, with a number of countries reducing high benefits duration in response to fiscal pres- sures and a growing pool of long-term unemployed. Coverage Is Low, Especially in the CIS Coverage rates of UI programs (that is, the share of the registered unemployed receiving unemployment benefits) vary between coun- tries and over time. This is in part related to the varying treatment of noncontributing groups like school leavers and other special groups and to the extension of UI duration for older unemployed. It also reflects the growing pool of unemployed who have exhausted their benefits entitlement. On average, the share of registered unemployed receiving benefits is just under 30 percent in the Region, and it has been falling over time in most countries. In CEE, the average cover- age rate was around 45 percent at the end of the 1990s, while for SEE the coverage was much lower, at around 12 percent, largely reflecting tighter qualification rules. Overall, the combination of coverage and replacement rates and benefits duration means that UI systems in CEE and SEE are at the lower end of OECD generosity and well below such levels in the CIS. The very low generosity of the CIS systems has made UI systems 224 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union increasingly irrelevant, contributing to low registered unemployment rates. The level of generosity is also reflected in spending shares on unemployment benefits, which in the late 1990s ran to an average of around 0.7 percent of GDP in CEE and SEE, but only 0.1­0.3 percent of GDP in the CIS. This compares with OECD and non-Region EU averages of around 1.4 and 1.7 percent, respectively.52 Two other considerations are important concerning UI in the tran- sition countries: · In a number of countries, a range of noncash benefits attached to registered unemployment status may have a strong impact on incentives to register. The most significant is free health insur- ance. There is evidence that in countries of former Yugoslavia, as well as in Poland, workers register as unemployed to ensure fam- ily health insurance coverage when no one in the household is paying social contributions. In Bosnia and Herzegovina, for exam- ple, only 25 percent of registered unemployed were actually unemployed, according to ILO definitions.53 In Croatia, this pro- portion is still higher and was 40 percent in 2001. This contributes to substantially inflated official unemployment figures in some countries. Conversely, the effective absence of additional entitle- ments for registered unemployed in the CIS reinforces low incen- tives to register. Also in Russia and other CIS countries, the services provided by public employment agencies do not provide better alternatives than individual job search, so the payoff to reg- istering is negligible. · As suggested above, the stringency of work test enforcement varies substantially across the Region. Country studies, such as one for the Czech Republic, indicate that the stringency of work test enforcement has been a contributing factor to keeping registered unemployment rates low. In contrast, most countries in SEE and the CIS make minimal efforts to enforce work tests, in part because of low labor demand and in part because of low administrative capacity. However, in Russia, administrative restrictions often lead the long-term unemployed to be struck from the register. More Generous UI Systems Have Been Associated with Poor Employment Outcomes Multivariate analysis for CEE, Russia, and Ukraine suggests that, if too generous, unemployment benefits might have a negative impact on labor market outcomes, with the level of benefits negatively affect- ing employment rates, and duration of benefits being strongly associ- Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 225 ated with a higher unemployment rate, and especially a higher long- term unemployment rate.54 This is consistent with OECD findings, but both the level and duration effects are stronger in transition than in OECD economies. In addition to these aggregate findings, studies of exits from unemployment in CEE confirm the negative effects of potential benefits duration on the probability of exit from unemploy- ment to employment, with notable increases in exits to employment near benefits exhaustion.55 The findings support the point made above that the generosity of CEE unemployment insurance systems in the initial stages of transition acted as a floor on real wages, inducing lower exits from unemploy- ment by raising the effective reservation wage. However, it is also important to note that subsequent tightening of UI through duration and level reductions had only a moderate impact on exits from unem- ployment to employment.56 The explanations include weak labor demand, structural mismatches and informational failures associated with geographic and sectoral segmentation, and inherited attitudes of the unemployed toward the role of the state in providing employment. Social Assistance Programs to Help the Long-Term Unemployed and the Poor Social assistance systems may be important in terms not only of poverty alleviation but also of labor market adjustment. To the extent that they work effectively, they help to protect workers who have lost their jobs and exhausted other entitlements before finding new work. However, there is also a risk that, if not well-designed, they raise the reservation wage above market wages, thereby discouraging job search. In contrast, where social assistance systems provide negligible benefits, workers are more inclined to remain in poorly paid jobs, willing to tolerate extended wage arrears and periods of administra- tive leave. A well-designed and adequately financed social assistance system can therefore have a role to play in labor adjustment--both in ensuring adequate work incentives and in encouraging productivity- enhancing labor turnover. The mix of social assistance­welfare programs within and between transition countries is complex. Apart from means-tested social assis- tance, there are typically child-family benefits; often benefits for util- ities for poor households; in-kind benefits; and benefits for specific groups such as refugees, veterans, and survivor families. Although all countries in the Region introduced an OECD-style mix of social pro- tection programs, the generosity and credibility of the systems have varied dramatically across the Region.57 226 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Over the transition period, the evolution of basic social assistance programs has been varied. Some countries have remained with the initial mix of programs, while others have been more innovative in experimenting with minimum income guarantees (for example, Esto- nia and the Kyrgyz Republic) and benefits structured as negative income taxes (for example, Lithuania).58 Others, such as Armenia and some oblasts of Russia (as well as Turkey), have introduced proxy means-tested programs for the poor. Others still, including Albania and Uzbekistan, have relied on community knowledge to target poor households. Moreover, following the experience of other developing countries, some economies of the Region have introduced programs that provide income support to poor families conditional on either basic needs triggers or behavioral changes (box 6.3). The variations in the type and coverage of social assistance pro- grams are in part a reflection of varying fiscal constraints and adminis- BOX 6.3 Innovative Ways of Targeting the Poor Two innovative ways of targeting the poor have gained momentum in developing countries and have been tested in some countries in the Region in recent years. Conditional cash transfers are an innovative channel for the delivery of social services that is be- ing implemented in an increasing number of countries.a They combine close targeting with cap- ital accumulation by making income support conditional on either basic needs triggers or be- havioral changes. Conditional cash transfers have virtually not been used in the Region so far, but they have been successfully implemented in Turkey, where there is evidence that they are resulting in higher attendance rates and higher continuation rates from primary to secondary ed- ucation. Social funds are investment funds that implement a bottom-up model in which communities and local institutions, rather than central governments, take the lead in identifying and carrying out community-level investments. The Armenia Social Investment Fund, created in 1996, was the first in the Region. A recent review of such funds in six developing countries, including Ar- menia, suggests that they are relatively successful. In Armenia, investments in education (school buildings, furniture, and basic equipment) are found to have increased primary enroll- ment rates, and investments in water and sanitation have decreased the frequency of illness in children.b a. See Rawlings 2004. b. See Rawlings, Sherburne-Benz, and Van Domelen 2004. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 227 trative capacity, both between and within countries. Overall, there is a major gap between spending as a share of GDP on nonemployment programs in CEE, SEE, and the CIS. For example, during 1991­95, nonemployment benefits ranged from 1.6 percent of GDP in Bulgaria to 5 percent of GDP in Poland in CEE countries, while they repre- sented only 0.3 percent in Ukraine and 0.6 percent in Russia.59 Social Assistance Benefits Are Linked to Minimum Wage or Other Social Minimums An important factor in a number of countries is the connection between the minimum wages and social benefits. This may in some cases create work disincentives, as well as overburdening the mini- mum-wage-setting process. As shown above, the Czech Republic dur- ing the 1990s provides an example of both these effects. A further factor about social welfare benefits and labor market out- comes relates to program incentives to (re)enter the workforce for those able to work. There has typically been limited effort to encour- age this in transition economies, either through workfare or incen- tives in the benefits themselves. Bulgaria is an exception, with its "From Social Assistance Toward Employment" program, under which social assistance recipients (with a priority for long-term unem- ployed) must accept temporary work or risk a reduction or removal of benefits. The participants also receive pension and health insur- ance. Participation can be no more than three years. This has led to significant reductions in social assistance rolls (see box 6.4);60 how- ever, the longer-term impact of such programs is less clear. For exam- ple, no positive impact on employment was found in a follow-up study of a similar regime introduced in Hungary in 2000 (Galasi and Nagy 2003). The Generosity of Social Assistance Is Associated with Wage and Labor Turnover Outcomes Overall, the labor market impacts of social assistance programs are not well researched in the Region. However, some evidence from CEE indicates that social welfare benefits spending facilitated restruc- turing in the face of falling real wages.61 By 1997, in CEE countries, there were more unemployed people on means-tested social assis- tance than on unemployment benefits.62 By contrast, in the CIS and most of SEE, lower spending on social welfare benefits, combined with the less progressive incidence of most social transfer spending, meant that workers were willing to tolerate much greater falls in real 228 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union BOX 6.4 Public Works and Workfare: An Alternative to the Unemployment Benefit? Temporary relief work or public works have been used extensively in many transition countries. Normally, they may be a short-term bridge between more-permanent jobs, especially in envi- ronments where jobs are few. They are designed to require high effort and low pay to ensure that only the neediest take part. The length of these jobs can vary, but typically it is between three months to one year. The long-term employment impact of these programs is often in- significant, and the unit costs are high. The organization of public works is often complicated be- cause of a mismatch between the skills needed by employers and the qualifications possessed by the unemployed. Many transition countries have used an offer to take part in public works programs as a work test or a disciplinary measure, to disqualify individuals who may have had hidden employment or income from temporary income support programs. Experience showed that programs operated by private sector employers resulted in more positive impacts than when they were implemented by public agencies. Public works were initially invented to address demand deficiency unemployment; however, they have increasingly turned into income-support programs largely targeted at low-skilled, long- term unemployed and other disadvantaged groups (for example, older workers and those in dis- tressed regions). This role has developed particularly in those countries where work incentives faced by the unemployed are weak because of benefits being close to the expected market wage. The underlying concept, known as workfare, is that able-bodied individuals should receive income support conditional upon, or in return for, performing some publicly useful work, to self- target the poor. For example, workfare was introduced in Albania, Bulgaria, and FYR Macedonia. However, in Al- bania and FYR Macedonia, wages were not set low enough to attract only the poorest workers. wages without exiting their (generally public sector) jobs. The rela- tively weaker role of social welfare benefits is also found to have con- tributed to the higher income inequality that emerged in Belarus, Russia, Ukraine, and other countries of the CIS over the transition period relative to CEE.63 Active Programs Have Been Introduced to Strengthen the Matching Process in the Labor Market All transition countries developed active labor market programs (ALMPs) to support workers affected by labor mobility. The programs include job counseling or job search assistance or both, wage subsi- dies, training and retraining programs, public works, and self- Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 229 Because of considerable income from the informal sector, people in FYR Macedonia did not wish to join public works programs even though the benefit was set at the average wage. In Al- bania, a community job creation program to undertake public services and infrastructure proj- ects within the community (for example, village roads, water drains, and so forth) was piloted in the late 1990s through the Bank Labor Market Development Project and then implemented on a broader scale by the National Employment Service through government funds. Recipients of social assistance had to "delegate" one household member to take part in the program. Be- cause wages earned were significantly higher than social assistance received, there was a sig- nificant demand for those jobs. A similar program in Bulgaria, "From Social Assistance Towards Employment," in which the un- employed individuals--recipients of social assistance--are given an option to either accept tem- porary jobs or else risk their social assistance benefit being cut or reduced, has been more suc- cessful. Wages are set at the minimum wage level, but those employed have to give up their social assistance benefits. Among the reasons for the attractiveness of the program might be that the participants are also covered with pension and health insurance (including maternity benefits). Also, for some of the participants from ethnically mixed regions, employment is com- bined with literacy enhancement and vocational training. In summary, if well designed and carefully targeted at disadvantaged worker groups, public works and workfare programs can provide income support to the unemployed without causing adverse labor supply incentives. Public works can also be effective when the intention is well defined (for example, to improve local infrastructure) and are arranged in close collaboration with local authorities (for example, with cofinancing arrangements). Design features that contribute to the success of workfare schemes also include (a) organization by private employers and (b) wages set below the market level to ensure self-targeting by the neediest among the unem- ployed and to prevent participation in public works becoming an attractive alternative to perma- nent employment. employment support. The relative emphasis of different types of ALMPs has varied considerably between countries. Public expenditures on ALMPs in transition countries have been considerably lower than OECD and EU averages as a share of GDP, with CEE average spending in the late 1990s around 0.4 percent of GDP and SEE only around 0.1 percent of GDP. This compares with an EU-15 average of 1.2 percent and an OECD average of 0.9 percent of GDP. However, there is also considerable variation between CEE countries in the share of spending on ALMPs, with the CEE average concealing a number of countries with spending shares on ALMPs well below the lowest OECD countries. In the CIS countries, spend- ing on ALMPs has been a negligible share of GDP, and they cannot be considered a significant policy in most of the subgroup. 230 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Although Not Increasing Aggregate Employment, ALMPs Have Had Positive Impacts When Supportive Preconditions Prevail The available evaluations in the Region suggest that ALMPs can improve the labor market outcomes of target participants,64 but only when some facilitating preconditions are met. It is important to stress that ALMPs achieve far less when labor demand is depressed and when the investment climate is poor. They are not a substitute for job creation by private firms. Even in conducive circumstances, they are only part of a menu of policies to support employment growth. Although ALMPs in the Region have in several cases improved the matching function of the labor market, they are only a supplemental tool for improving labor market outcomes, the effectiveness of which is largely dependent on employment-enhancing policies outside the labor market itself. Nonetheless, while ALMPs may not raise aggre- gate employment, redistributing job opportunities toward the disad- vantaged worker groups may have more than typical value in the Region because of the larger stagnant pool of long-term unemployed. Some ALMPs Have Proven More Effective Than Others Evaluations of ALMPs in the Region are from CEE and some SEE countries only and are relatively few to date. In some cases, the employment impact of programs in the transition countries has been more positive than OECD comparators. This is most notable with training. Overall, however, they indicate that the net impact of ALMPs in the Region seldom exceeds 10 percent and is often lower.65 Certain ALMPs work better and are more cost-effective than others, with the ranking of interventions varying according to specific labor market conditions and dependent on the effectiveness of local insti- tutions. Questions remain on their relevance for the poorer and more informal economies in the Region, where informal networks may play a stronger role in finding employment. Although the effective- ness of specific programs in specific countries cannot be determined beforehand, some findings emerge from evaluations in the Region:66 · Multivariate regression analysis from CEE in the 1990s found sig- nificant positive effects of ALMPs on outflows to jobs among the unemployed in Bulgaria, the Czech Republic, Poland, and the Slo- vak Republic, while in Hungary the impact was never significant.67 · As in other regions, job search assistance in this Region--generally provided by public employment services--has proved the most Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 231 cost-effective intervention to address the structural component of unemployment. The positive effects in reducing unemployment duration appear to be clearer in transition countries than in OECD countries.68 · The employment impact of training seems positive and notably better than that for the OECD, though the impact on earnings has been mixed. Training programs in the Region also appear to have been more cost-effective. Targeted small-scale programs focused on specific groups have tended to have a positive net impact, par- ticularly for the low-skilled. For example, in Hungary and Poland, the net impact of training on workers with primary education was found to be 15­17 percent, while in Bulgaria it exceeded 20 per- cent. The available evaluations confirm a significant employment effect for the poorer and more informal economies in SEE. As noted above, net impact evaluations have not been carried out in the CIS. · The evidence on the impact of wage subsidies is mixed. In some countries (for example, the Czech Republic and Hungary), they have proven effective in helping disadvantaged unemployed, with a net postprogram impact ranging from 10 to 15 percent. The effects have been particularly positive in high-unemployment regions in countries such as Bulgaria. The results are more positive than evaluations of similar programs in OECD countries. However, some evaluations also show negative employment effects of wage subsidies. · Public works can act as effective safety net interventions, but the impact on the employment prospects of participants has been mixed. In general, the transition countries have relied on public works far less in their mix of ALMPs than both a number of OECD countries and other regions such as South Asia have. Outcomes range from significant negative impacts on future employment prospects and wages in Hungary and Poland, to no impact in Romania, to positive--though perhaps short-term--employment impacts in FYR Macedonia, the Slovak Republic, Slovenia, and Ukraine. However, it is important to view public works more as an income-support program than as a typical ALMP, so the temporary income impact on participants should probably be given higher weight than for other types of programs (see box 6.4). · Programs that promote self-employment--typically credit or grant programs, often combined with business advisory services--have been less evaluated in the Region. The available evaluations indi- 232 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union cate positive outcomes from such programs, but the results are limited because they target a small subgroup of the unemployed. An important consideration in interpreting the above results is the institutional capacity of public employment agencies. There has tradi- tionally been a tendency in the Region for employment agencies that operate both UI and ALMPs to run their own programs. In a number of cases, private employment agencies have been prohibited under the relevant legislation (although this is generally being remedied now). The public provision of ALMPs has in many countries resulted in programs that take limited account of local labor demand in their design. A further concern in some countries has been the governance of unemployment funds; for example, such funds may be used for political or other purposes, or there may be a lack of transparency, making it difficult to track the use of the funds. Finally, the design and evaluation of such programs are important factors in determining their efficiency.69 Further Reforms of Pension and Health Insurance Systems Will Also Be Required In addition to the above, reforms of pension systems and (to a lesser extent) health insurance systems in the Region will continue to have significant effects on the labor market. These are direct (for example, with regard to the longer-term incentives for older workers to increase labor force participation) but also indirect (for example, in the large tax burden on labor imposed by social insurance programs in the Region). Early in the transition period, countries of the Region have recog- nized the necessity of revising the ongoing pay-as-you-go monopillar pension systems that were under pressure because of a shrinking base of contributions and a growing number of beneficiaries. All countries revised the way benefits were adjusted toward rates that were feasi- ble under available resources. Some also started to increase the retire- ment age, which was initially low and had even declined in some countries in the early 1990s. Others revised benefits formulae (for example, increasing the reference period). Beyond these initial changes, two groups of countries emerged. One group of countries (including Albania, Armenia, Azerbaijan, Georgia, and Tajikistan) maintained minimally adequate income floors, which smoothed out any benefit differences resulting from years of contribution or previous earnings. The countries of the sec- ond group looked to introduce multipillar systems through shifting a Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 233 portion of the mandatory contribution from the public pension to pri- vate institutions that established individual defined-contribution accounts for individual workers. As of 2004, nine countries70 were already operating second pillars, while FYR Macedonia, Romania, the Slovak Republic, and Ukraine had some legislation about a second pillar. Despite problems, reversals, and some transitional costs, the systems are overall better than the old ones. Outstanding issues remain, and reforms need to continue to bring full benefits (Holzmann and Hinz 2005). Specifically, three issues arise in the context of this report. First, the evolution of the labor market, combined with the desire to better link benefits with contri- butions, has left some individuals with inadequate pensions and rely- ing on social safety nets. In a context of increasing real wages, there has been political pressure to increase the indexation of existing pen- sions, keeping contribution rates high. Although this has so far not threatened the competitiveness of the transition countries (because of their comparatively low wage rates), this trade-off may become binding as wages continue to increase. Second, more fiscal pressures may arise because of rapidly aging populations and low fertility rates; most countries in the Region will need to continue reforms of the pension system to adjust the system dependency ratio to changing demographics.71 Further increases in the retirement age may be necessary as life expectancy increases. Third, countries of the Region also need to undertake other welfare reforms that will stimulate formal sector labor supply to improve the contributions base for a given working-age population. In addition, CEE and SEE countries will need to improve--and in many cases tighten--their eligibility rules and assessment mechanisms within their disability pension programs to avoid an unaffordable dependence on disability benefits by the working-age population. Furthermore, the need for health system reforms that will allow for gradual reduc- tion in health insurance payroll taxes (particularly in SEE) is clear. Summing Up: The Challenge of Labor Policy Reforms in Transition Economies This chapter indicates that--like many other areas of the economy and society in transition countries--labor market institutions have been subject to major changes in what is a very short period in inter- national comparative terms. It is likely that the full effects of these reforms have not yet been felt. Overall, CEE and, to a large extent, SEE countries have developed labor market institutions similar to 234 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union those in mature market economies. Labor market reforms in Russia, Ukraine, and other CIS countries in contrast have been less thorough, with greater residual institutional rigidities diluted by weak enforce- ment and high informality. Across the Region, however, the remain- ing rigidities in labor market institutions continue to have negative impacts on employment outcomes, albeit probably not as acute as sometimes portrayed. Although labor market institutions cannot be considered as the primary cause of unemployment in the Region, there remains a sig- nificant institutional reform agenda across the Region, more so as neighboring EU countries continue labor market liberalization. The specific elements of the agenda and the relative priority between them vary between subgroups, and often within subgroups. General conclusions that are appropriate for each country on all elements of reform are therefore overly ambitious. It is also very clear that labor market institutions and their impact on labor market outcomes con- stitute an area where substantially more research is needed. With this major caveat in mind, broad patterns of an agenda for the reform of labor market institutions in the Region and priorities for each subgroup are apparent and will need to be assessed in the light of further research. The key elements of a reform agenda by subgroup are summarized below. The Labor Market Institutional Reform Agenda in CEE Countries There is a need for greater adaptability of real wages to productivity. This could be achieved by further decentralization of wage bargaining or by strengthening a social compact that links wage and productivity growth but retains the flexibility for firm bargaining within the aggre- gate wage adjustment parameter. This should avoid an increased role for sectoral bargaining, which appears to be associated with worse labor market outcomes. In public sector wage setting, there is proba- bly a need for an increased flexibility in wage setting along the model developed (for example) in Estonia. Moreover, there is a need for reconsidering minimum-wage setting, which, albeit not binding on average, may be too high for the low-skilled and for workers in back- ward areas of many CEE economies. In this context, the possible impact of the rigid link between minimum wage and social benefits should be reassessed. In the area of employment protection legislation, there is a need to review regulations for permanent employment contracts, which remain overly rigid also in light of the liberalization of temporary Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 235 contracts. This should help avoid a significant segmentation of the labor market. There is also a need to further improve labor inspection methods to balance effective enforcement with avoidance of unnec- essary impositions on firms, including an increased role for firm self- reporting. There is also a need to reduce taxes on labor, which are currently very high in CEE, and appear to be a major impediment to increasing employment. However, this will need to be done gradually and in parallel with further reform of pensions and disability and health insurance to balance fiscal requirements, social protection of social insurance beneficiaries, and the need to improve labor market out- comes. Unemployment benefits have now been reformed in many CEE countries, and their generosity is currently relatively low, limiting possible work disincentives. The main challenge is to strengthen the (re)activation of the long-term unemployed, both through incentives in the benefits and tax systems (for example, back-to-work allowances and tax credits for low-income working households) and through targeting active programs at those groups for whom their net impact is the highest. There is also a need to improve monitoring and evaluation of the effectiveness of different ALMPs to improve their design and targeting efficiency. Moreover, social assistance benefits in depressed regions need to be examined to ensure that eligibility thresholds and benefits levels do not contribute to localized work dis- incentives. The Labor Market Institutional Reform Agenda in SEE Countries Regarding wage determination in SEE countries, there is a need to (a) lessen the role of the government in collective bargaining and strengthen the voice of employers in an attempt to improve the wage and productivity linkage; (b) reduce the scope of general and sectoral collective agreements by promoting greater decentralization of bar- gaining to the firm level; (c) increase flexibility in minimum-wage setting by allowing for youth minimum wages; and (d) ensure that the minimum-wage-to-average-wage ratio is controlled and, in some countries (for example, BiH), reduced. Further reform is also needed in the area of employment protection, where regulations on temporary employment contracts remain fairly strict. SEE countries should also ensure that more liberal EPL provi- sions are not undermined by continued rigidity of employment condi- tions in collective agreements. This should be accompanied by a 236 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union significant improvement in labor inspection practices to balance effec- tive enforcement with the avoidance of unnecessary impositions on firms, including an increased role for firm self-reporting of compliance. SEE countries should aim for a gradual reduction in payroll taxes, with a particular focus on health insurance taxes. This should be accompanied by efforts to improve compliance on tax payments by firms, including SOEs. Moreover, many countries in this subgroup should improve on the (re)activation of the long-term unemployed, both through incentives in the benefit system (for example, back-to- work allowances) and through targeting ALMPs on those with low skills and in depressed regions. This could also imply some realloca- tion of funds across activities. This should be accompanied by encour- aging private employment agencies to play a role in areas where local labor markets permit. The Labor Market Institutional Reform Agenda in the Middle and Low Income CIS Countries This group of countries is even more heterogeneous than the others, and it is thus more difficult to identify common policy objectives. Even so, some common elements could be identified. One concerns the wage-bargaining process, in which most countries in the sub- group should promote better representation of social partners and lower direct intervention from the government. This implies encour- aging the emergence of independent unions and employers' associa- tions that will assume a more active role in bargaining and moving away from wage schedules in which all wages are rigidly set as mul- tiples of the minimum wage. In the area of employment protection, most countries in this sub- group have had more limited reforms, with strict regulations being accompanied by pervasive nonenforcement. Focusing EPL on a core set of labor protections that can be credibly enforced and promoting a reform in labor inspection services could go some way in promoting labor reallocation within the formal sector. Passive and active labor market programs are less developed in this subgroup than in most of the Region. Given the limited administrative capacity, large underreporting of activities, and limited financial resources, especially in the low income CIS countries, an appropriate choice could be to provide a minimum level of protection for some basic risks. Such protection could be financed through general revenues.72 For example, the issue could be raised as to whether a move to flat-rate unemployment assistance is appropriate and whether this might be financed through general revenues, rather than the payroll tax. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 237 More generally, moving toward universal social protection could be considered in some countries. Decoupling minimum pension schemes and health insurance from the labor market status of people (for example, through creating a universal minimum pension or expanding health insurance coverage to informal workers) are options currently under consideration in some emerging economies such as Mexico. Moreover, given widespread informality, the need to reinforce social protection for those without access to formal insur- ance mechanisms should be considered. This could include greater use of public works schemes that are open to all those willing to work at a program wage rate that is set at a level low enough to ensure self- targeting of the most-needy. Countries that have not already done so should also consider using alternative approaches to targeting of social transfers, such as proxy means testing or community identification of beneficiaries within national guidelines. Improving targeting for programs aimed at vul- nerable groups would enable more efficient redistribution of resources. Regarding social insurance, CIS countries should continue reforms of the pension system that will adjust the system dependency ratio to changing demographics, and encourage other reforms that will stimulate formal sector labor supply to improve the contributions base for a given working-age population. Annex 6.1: Construction of Employment Protection Legislation Indexes The indexes are calculated from the raw data provided in the World Bank's Doing Business Database (2005). All the variables range between 0 and 1. Either they are zero-one dummies (no-yes), or they are normalized over the sample. Variables for which higher values (x) represent looser regulations are normalized according to the simple formula (equation 6.1): x - min(xE) xN = 1-((max(xE) - ) (6.1) min(xE) whereas variables for which higher values represent stricter regula- tions are normalized in the following way (equation 6.2): (max(xE)- x) xN = 1- (6.2) (max(xE)- min(xE)) 238 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union where xN is the normalized value, max(xE) is the maximum value that the variable takes over the sample of countries available from the database, and min(xE) is the corresponding minimum value. The permanent employment weighted index is constructed from the raw variables in the following way (weights in parentheses): Procedures (1/2) The employer must notify a third party before dismissing one redundant employee. (1/6) The employer needs the approval of a third party to dismiss one redundant worker. (1/6) The law mandates retraining or replacement before dismissal. (1/6) There are priority rules applying to dismissal or layoffs. (1/6) There are priority rules applying to reemployment. (1/6) Is redundancy considered a "fair" grounds for dismissal? (1/6) Firing costs (1/2) Legally mandated notice period for redundancy dismissal (in weeks) after 20 years of continuous employment. Normalized (1/2) Severance pay for redundancy dismissal as number of months for which full wages are payable after continuous employment of 20 years. Normalized (1/2) The temporary employment regulation index is equal to the simple average of the following variables: Fixed-term contracts are only allowed for fixed-term tasks. What is the maximum duration of fixed-term contracts (in months)? Normalized Finally, the overall employment regulation index is the simple average of the regular and temporary employment indexes. Notes 1. Useful overviews of institutional issues in the Region's labor markets include Haltiwanger, Scarpetta, and Vodopivec (2003); Boeri and Terrell (2002); Riboud, Sanchez-Paramo, and Silva-Jauregui (2002); Garibaldi and Brixiova (1998); and Svejnar (1999 and 2002). 2. See Blanchflower and Oswald (1998); Köllő (1998); Basu, Estrin, and Svejnar (2004); and Iara and Traistaru (2004). 3. See Ilkkaracan and Selim (2003). 4. See Basu, Estrin and Svejnar (2004) and Flanagan (1998). Poland and (to a lesser extent) Hungary were partial exceptions to this general pic- ture by the end of the communist period. 5. "Union density" refers to the proportion of workers who are members of a union. Union coverage refers to the proportion of workers who are covered by a collective bargaining agreement. 6. See ILO (2002, in Russian). Armenia provides an example of the diffi- culty of interpreting union density figures in the CIS, with 75 percent union density by official estimates, but 80 percent of workers claiming to "have nothing in common" with trade unions, and half of those claim- ing to be totally uninformed about unions. Often the inclusion of pen- sioners and segments of the registered unemployed in union membership numbers also inflates density data. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 239 7. See, among others, EIRO (2004). 8. See Micevska (2004). 9. BiH includes two entities: Republika Srpska and the Federation of Bosnia and Herzegovina. 10. In many countries, this has been in part a political reaction against the "state unions" that monopolized the union movement during the social- ist period. 11. See ILO (2002, in Russian) for case studies of CIS country social dia- logue. 12. The Russian coefficient system tends to set lower bounds that are very low so that additional payments make up a significant amount of total earnings of public sector employees. 13. See Schiavo-Campo, de Tommaso, and Mukherjee (1997). 14. In many countries that have not reformed public sector pay determina- tion, there has been a significant incidence of payments to employees that are not subject to the coefficient schedule. Such payments have been both formal, through allowances of various forms, and informal, through undisclosed payments to workers. 15. See Schiavo-Campo, de Tommaso, and Mukherjee (1997). By the mid- 1990s, the Region's public-to-private-sector-wage ratio was the lowest of any region, though it was somewhat higher in CEE and SEE. 16. Although there are issues of recording and interpreting data on coverage and density, data generally refer to the formal economy (that is, they include both the state and private sectors). 17. Slovenia has general collective agreements for both the whole public sector and the whole private sector, with sectoral- and company-level agreements (EIRO 2004). The tripartite private sector agreement for 2004­5 includes provisions on principles for pay increases, aggregate increases, minimum increases for all workers, and the minimum wage, among others. These follow the general principle of restraining real- wage growth below productivity growth. 18. See EIRO (various years). Martin and Cristescu-Martin (2002 and 2003) also provide annual updates of developments in CEE industrial relations. 19. This is in contrast to the bulk of EU-15 countries, where wage bargain- ing takes place primarily at the sectoral level in most countries, though with a trend toward greater decentralization during the 1990s. CEE countries will be subject to social partnership arrangements as part of EU accession. 20. It is estimated that some 30 percent of blue-collar workers and 15 per- cent of white-collar workers in the private sector in Russia are paid according to public sector wage grids (Kapelyushnikov 2003). 21. See Haltiwanger, Scarpetta, and Vodopivec (2003). 22. See Micevska (2004). 23. See Luke and Schaffer (2000) for Russia, Dobbelaere (2004) for Bulgaria, and Grosfeld and Nivet (1999) and Mickiewicz and Bishop (2003) for Poland. Mickiewicz and Bishop also find that relative rent sharing by workers divides more strongly along old and new firm lines than state and private ones. 24. See Haltiwanger, Scarpetta, and Vodopivec (2003). This association does not necessarily imply causality. The result is at least in part driven by the 240 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union coincidence of centralized or coordinated bargaining in Russia and Ukraine with seemingly favorable labor market outcomes (see chapter 2). However, low open unemployment in Russia and Ukraine is mainly the effect of the nature of enterprise restructuring, rather than of the bargaining system. It is also worth noting that although Russia's wage- bargaining system appears centralized on paper, in practice managers have great flexibility at the level of the firm. 25. See OECD (2004a). 26. See Haltiwanger and Vodopivec (2003). 27. See EIRO (2004) for Bulgaria; ONASA (2004) for Bosnia. 28. See Eriksson and Pytlikova (2004). 29. Social benefits were set as a percentage of the "subsistence wage." 30. A minimum social security threshold that exceeds the actual wage implies a higher rate than the nominal tax rate on that wage. In conse- quence, this may discourage the demand for less-skilled labor. 31. See OECD (2003) for discussion of the Latvia case. 32. See Eriksson and Pytlikova (2004). 33. In some countries, the minimum wage is differentiated according to individuals' education and qualifications. For example, in Montenegro, while the "standard" minimum wage is around 17 percent of the aver- age gross wage, the minimum wage for individuals with a university or higher education degree is at a high 54 percent of the average wage. 34. Kertesi and Köllő (2003) provide evidence on the disemployment effects that the substantial hikes in minimum wage in Hungary had in back- ward areas in 2001; see Hinnosaar and Rőőm (2003) for Estonia. 35. Recent literature also points to other functions of EPL, including ensur- ing the income of workers (who are generally risk averse) against labor market uncertainty and providing incentives to employers to internalize the social costs of layoffs in the presence of a higher social than private value of employment. See OECD (2004a); Pissarides (2001). 36. See for example, Nickell (1997). 37. At the same time, the speed of market-oriented reforms, of which labor market reforms are a part, influenced the EU-accession process. 38. Evidence from Argentina, Colombia, France, Peru, and Spain suggests that asymmetric liberalization of contracts has led to significant shifts toward precarious jobs. In Argentina and Spain, these reforms were reversed after a few years, and in the latter, net job creation really picked up only after the government reformed regular contracts in the mid-1990s. In Spain, temporary employment reached almost one-third of the total workforce after the reform of temporary contracts in the mid-1980s (Dolado, García- Serrano, and Jimeno 2002). In Peru, the liberalization of temporary employment in the early 1990s led to an increase in temporary employ- ment from 20 percent in 1990 to 55 percent in 2000. In Colombia, there was a similarly large increase (Saavedra 2003). See also Blanchard and Landier (2002) for France and Hopenhayn (2004) for Argentina. 39. Squire and Suthiwart-Narueput (1995) investigate the impact of labor market regulations in settings where compliance is incomplete. 40. See World Economic Forum (2002). 41. Although neither of these relationships is posited as causal, a link is sup- ported by qualitative assessments. Labor Market Policy and Institutions: Combining Protection with Incentives for Job Creation 241 42. The regressions are done on eight CEE countries: Bulgaria, the Czech Republic, Estonia, Hungary, Poland, Romania, the Slovak Republic, and Slovenia (Haltiwanger, Scarpetta, and Vodopivec 2003). 43. The coordination variable is the simple average of the degree of coordi- nation among employers' associations and trade unions (Haltiwanger, Scarpetta, and Vodopivec 2003). 44. See World Bank (2004c). 45. See Haltiwanger, Scarpetta, and Vodopivec (2003). 46. See Micevska (2004). 47. The calculated tax wedge on labor in reduced form is defined as (labor cost ­ net pay)/labor cost*100. Labor cost comprises gross wage plus employer payroll contributions on wage. Net pay is calculated as gross wage less employee payroll contributions and personal income tax. 48. The comparison of the tax wedge should be treated with caution because of tax benefits and the treatment of social transfers to families with chil- dren, which vary across OECD countries. 49. See Riboud, Sanchez-Paramo, and Silva-Jauregui (2002). 50. See Daveri and Tabellini (1997); Elmeskov, Martin, and Scarpetta (1998). 51. See Haltiwanger, Scarpetta, and Vodopivec (2003); Cazes and Nesporova (2003). 52. Low registered unemployment may lead to artificially high spending per registered unemployed, such as in Russia. 53. See World Bank (2002b). 54. See Haltiwanger, Scarpetta, and Vodopivec (2003). One should bear in mind that reverse causality (endogeneity) may be at work here. Govern- ments often are under pressure to respond to growing unemployment by lengthening the duration of unemployment benefits. This happened, for example, in Poland, where benefit duration is longer in regions of high unemployment. To the extent this is the case, the estimated regres- sion coefficient is biased and overstates the impact of benefit duration on unemployment. 55. See Vodopivec (2004) for a comparative analysis of Estonia and Slove- nia. 56. See, for example, Micklewright and Nagy (1996) for Hungary; Ham, Svejnar, and Terrell (1998 and 1999) for the Czech and Slovak Republics. 57. See Atkinson and Micklewright (1992) for pretransition safety net details. 58. See Staines and Isamiddinova (2004). 59. See table 5 in Boeri and Terrell (2002). 60. See Kuddo (2004). 61. See Boeri and Terrell (2002) and Garibaldi and Brixiova (1998). 62. See Boeri (1997). 63. See Commander and Lee (1998) on Russia, and Garner and Terrell (1998) on the Czech and Slovak Republics. 64. This applies to the employment rates of participants and (in some cases) to their postprogram wages. 65. The net impact is lower than the gross impact. The gross impact is the fraction of program beneficiaries who were placed in jobs. The net impact is the gross impact minus the estimated fraction of program beneficiaries 242 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union who would have found jobs anyway (the deadweight loss). The dead- weight loss can be estimated using experimental or quasi-experimental techniques, as well as econometric methods. 66. See Betcherman, Olivas, and Dar (2004) for a summary of ALMP evalu- ations, including those for selected countries in the Region. See also Boeri (1997) and Bonin (1999). 67. See Boeri (1997). 68. See Fretwell, Benus, and O'Leary (1999) for cost-effectiveness analysis on ALMPs in CEE countries. 69. ALMPs often have several social and economic goals, which need to be both determined before a new program is designed and taken into account when carrying out impact evaluations. 70. Bulgaria, Croatia, Estonia, Hungary, Kazakhstan, Latvia, Lithuania, Poland, and Russia. 71. The system dependency ratio is the ratio of the number of pensioners to the number of contributors to the pension system. 72. There is scope for improving tax collection. For example, De Ferranti and others (2004) propose general principles to improve the efficiency and equitability of taxation in the context of Latin America. Bibliography Acquisti, Alessandro, and Hartmut Lehmann. 2000. "Job Creation and Job Destruction in the Russian Federation." Trinity Economic Paper 20001, Trinity College Dublin, Department of Economics, Dublin. Aghion, Philippe, and Olivier J. Blanchard. 1994. "On the Speed of Transition in Central Europe." In NBER Macroeconomics Annual 1994: 283­320. Cam- bridge, MA: MIT Press. Aiginger, K., M. Boeheim, K. Gugler, M. Pfaffermayr, and Y. Wolfmayr- Schnitzer. 1999. "Specialisation and (Geographic) Concentration of Euro- pean Manufacturing." Working Paper 1, Enterprise DG 3, European Commission, Brussels. Andrienko, Yuri, and Sergei Guriev. 2003. "Determinants of Interregional Mobility in Russia: Evidence from Panel Data." William Davidson Institute Working Paper 551, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Aslund, Anders, Peter Boone, and Simon Johnson. 1996. "How to Stabilize: Lessons from Post-Communist Countries." In Brookings Papers on Economic Activity 1: 217­313. Washington, DC: Brookings Institute. Atkinson, Anthony B., and John Micklewright. 1992. Economic Transformation in Eastern Europe and the Distribution of Income. Cambridge, U.K.: Cambridge University Press. Babetski, Jan, Alexandre Kolev, and Mathilde Maurel. 2003. "The Kyrgyz Labour Market in the Late 1990s: The Challenge of Formal Job Creation." Comparative Economic Studies 45 (4): 493­520. 243 244 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Baily, Martin Neil, Eric J. Bartelsman, and John Haltiwanger. 2001. "Labor Productivity: Structural Change and Cyclical Dynamics." Review of Econom- ics and Statistics, 83 (3): 420­33. Baily, Martin Neil, Charles Hulten, and David Campbell. 1992. "Productivity Dynamics in Manufacturing Plants." Brookings Papers on Economic Activity: Microeconomics, 187­249. Washington, DC: Brookings Institute. Baldwin, John, Timothy Dunne, and John Haltiwanger. 1998. "A Compari- son of Job Creation and Job Destruction in Canada and the United States." Review of Economics and Statistics, 53 (3), 347­56. Barlevy, Gadi. 2002. "The Sullying Effect of Recessions." Review of Economic Studies 69 (1): 65­96. ------. Forthcoming. "Credit Market Frictions and the Allocation of Resources over the Business Cycle." Journal of Monetary Economics. Bartelsman, Eric, John Haltiwanger, and Stefano Scarpetta. 2004. "Microeco- nomic Evidence of Creative Destruction in Industrial and Developing Coun- tries." Policy Research Working Paper 3464, World Bank, Washington, DC. Bartelsman, Eric, Stefano Scarpetta, and Fabiano Schivardi. 2005. "Compara- tive Analysis of Firm Demographics and Survival: Evidence from Micro- Level Sources in OECD Countries." Industrial and Corporate Change 14 (3): 365­91. Basu, Swati, Saul Estrin, and Jan Svejnar. 2004. "Wage Determination under Communism and in Transition: Evidence from Central Europe." William Davidson Institute Working Paper 717, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Beck, Thorsten, Asli Demirguc-Kunt, and Luc A. Laeven. 2005. "Finance, Firm Size, and Growth." Policy Research Working Paper 3485, World Bank, Washington, DC. Betcherman, Gordon, Karina Olivas, and Amit Dar. 2004. "Impacts of Active Labor Market Programs: New Evidence from Evaluations with Particular Attention to Developing and Transition Countries." Social Protection Dis- cussion Paper 0402, World Bank, Washington, DC. Bilsen, Valentijn, and Jozef Konings. 1998. "Job Creation, Job Destruction, and Growth of Newly Established, Privatized, and State-Owned Enter- prises in Transition Economies: Survey Evidence from Bulgaria, Hungary, and Romania." Journal of Comparative Economics 26 (3): 429­45. Blanchard, Olivier, Rudiger Dornbusch, Paul Krugman, Richard Layard, and Lawrence Summers. 1991. Reform in Eastern Europe. Cambridge, MA: MIT Press. Blanchard, Olivier, and Francesco Giavazzi. 2003. "Macroeconomic Effects of Regulation and Deregulation in Goods and Labor Markets." Quarterly Jour- nal of Economics 118 (3, August): 879­907. Blanchard, Olivier, and Michael Kremer. 1997. "Disorganization." Quarterly Journal of Economics 112 (4): 1091­1126. Blanchard, Olivier, and Augustin Landier. 2002. "The Perverse Effects of Par- tial Labour Market Reform: Fixed-Term Contracts in France." The Economic Journal 112 (480): F214­44. London: Royal Economic Society. Bibliography 245 Blanchard, Olivier, and Justin Wolfers. 2000. "The Role of Shocks and Insti- tutions in the Rise of European Unemployment: The Aggregate Evidence." The Economic Journal 110 (462): C1­33. London: Royal Economic Society. Blanchflower, David, and Andrew Oswald. 1998. "Unemployment, Well- Being and Wage Curves in Eastern Europe." Photocopy. http://www2.warwick.ac.uk/fac/soc/economics/staff/faculty/oswald/eur opeeast.pdf. Boeri, Tito. 1995. "Unemployment Dynamics and Labour Market Policies." In Unemployment, Restructuring, and the Labor Market in Eastern Europe and Rus- sia, ed. S. Commander and F. Coricelli, 361­83. Washington, DC: World Bank. ------. 1997. "Learning from Transition Economies: Assessing Labor Market Policies across Central and Eastern Europe." Journal of Comparative Eco- nomics 25 (3): 366­84. Boeri, Tito, Michael C. Burda, and János Köllő. 1998. "Mediating the Transi- tion: Labour Markets in Central and Eastern Europe." In Mediating the Transition: Labour Markets in Central and Eastern Europe, ed. L. Ambrus- Lakatos and M. Schaffer. Forum Report of the Economic Policy Initiative, No. 4. New York: CEPR. Boeri, Tito, and Pietro Garibaldi. 2004. "Labor Market Performance in Tran- sition Economies: A Macroeconomic Perspective." Background Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Boeri, Tito, and Stefano Scarpetta. 1996. "Regional Mismatch and the Transi- tion to a Market Economy." Labour Economics 3 (3): 233­54. Boeri, Tito, and Katherine Terrell. 2002. "Institutional Determinants of Labor Reallocation in Transition." Journal of Economic Perspectives 16 (1, Winter): 51­76. Bonin, John, ed. 1999. "Special Issue on Unemployment and Labor Market Policies in Transition Countries." Journal of Comparative Economics 27 (1). Bornhorst, Fabian, and Simon Commander. 2004. "Regional Unemployment and its Persistence in Transition Countries." Discussion Paper 1074, Insti- tute for the Study of Labor (IZA), Bonn. Brown, J. David, and John S. Earle. 2002a. "Gross Job Flows in Russian Industry Before and After Reforms: Has Destruction Become More Cre- ative?" Journal of Comparative Economics 30 (1): 96­133. ------. 2002b. "Job Reallocation and Productivity Growth under Alternative Economic Systems and Policies: Evidence from the Soviet Transition." Dis- cussion Paper 644, IZA, Bonn. ------. 2002c. "The Reallocation of Workers and Jobs in Russian Industry: New Evidence on Measures and Determinants." Staff Working Paper 02- 83, W. E. Upjohn Institute for Employment Research, Kalamazoo, MI. ------. 2004. "The Microeconomics of Creating Productive Jobs: A Synthesis of Firm-Level Studies in Transition Economies." Background Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. 246 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Bruno, M., and J. Sachs. 1985. Economics of Worldwide Stagflation. Cambridge, MA: Harvard University Press. Caballero, Ricardo J., and Mohammad L. Hammour. 1994. "The Cleansing Effect of Recessions." American Economic Review 84 (5): 1350­68. ------. 1996. "On the Timing and Efficiency of Creative Destruction." Quar- terly Journal of Economics 111 (3): 805­51. ------. 2000. "Institutions, Restructuring, and Macroeconomic Perfor- mance." Working Paper 7720, National Bureau for Economic Research (NBER), Cambridge, MA. Campos, Nauro F., and Fabrizio Coricelli. 2002. "Growth in Transition: What We Know, What We Don't, and What We Should." Journal of Economic Lit- erature 40 (3): 793­836. Castanhiera, M., and G. Roland. 2000. "The Optimal Speed of Transition: A General Equilibrium Analysis." International Economic Review 41 (1): 219­39. Cazes, Sandrine, and Alena Nesporova. 2003. Labour Markets in Transition: Balancing Flexibility and Security in Central and Eastern Europe. Geneva: ILO. Chadda, B., and F. Corricelli. 1994. "Fiscal Constraints and the Speed of Tran- sition." Discussion Paper 993, CEPR, London. Chenery, H. B., and L. Taylor. 1968. "Development Patterns: Among Coun- tries and Over Time." Review of Economics and Statistics, 50: 391­416.Claessens, Stijn, and Kyle Peters. 1997. "State Enterprise Perfor- mance and Soft Budget Constraints." Economics of Transition 5 (2): 305­22. Commander, Simon, and János Köllő. 2004. "The Changing Demand for Skills: Evidence from the Transition." Discussion Paper 1073, Institute for the Study of Labor (IZA), Bonn. Commander, Simon, and Une Lee. 1998. "How Does Public Policy Affect the Income Distribution? Evidence from Russia, 1992­1996." Photocopy, EBRD, London. Coricelli, F., and V. Ercolani. 2003. "Deficits and Fiscal Rules on the Road to an Enlarged European Union." Working Paper 3672, CEPR, London. Daveri, F., and G. Tabellini. 1997. "Unemployment, Growth and Taxation in Industrial Countries." Discussion Paper 9706, Brescia University, Brescia, Italy. Davis, Steven J., and John C. Haltiwanger. 1990. "Gross Job Creation and Destruction: Microeconomic Evidence and Macroeconomic Implications." NBER Macroeconomics Annual 1990: 123­68. Cambridge, MA: MIT Press. ------. 1992. "Gross Job Creation, Gross Job Destruction and Employment Reallocation." Quarterly Journal of Economics 107 (3): 819­63. ------. 1999. "Gross Job Flows." In Handbook of Labor Economics, ed. Orley Ashenfelter and David Card, Vol. 3B: 2712­2803. Amsterdam: Elsevier. Davis, Steven J., John C. Haltiwanger, and Scott Schuh. 1996. Job Creation and Destruction. Cambridge, MA: MIT Press. De Ferranti, David, Guillermo Perry, Francisco Ferreira, and Michael Walton. 2004. Inequality in Latin America and the Caribbean: Breaking with History? Washington DC: World Bank. Bibliography 247 De Nicoló, Gianni, Sami Geadah, and Dmitriy Rozhkov. 2003. "Financial Development in the CIS-7 Countries: Bridging the Great Divide." Working Paper 03/205, IMF, Washington, DC. Decressin, Jörg W. 1994. "Internal Migration in West Germany and Implications for East-West Salary Convergence." Weltwirschaftliches Archiv 130 (2): 231­57. Djankov, Simeon, and Peter Murrell. 2002. "Enterprise Restructuring in Tran- sition: A Quantitative Survey." Journal of Economic Literature 40 (3): 739­92. Dobbelaere, Sabien. 2004. "Ownership, Firm Size and Rent Sharing in Bul- garia." Labour Economics 11 (2): 165­89. Dobrogonov, Anton. 2002. "Social Protection in Low Income CIS Countries." Photocopy, World Bank, Washington, DC. Dolado, Juan, Carlos Garcia-Serrano, and Juan Jimeno. 2002. "Drawing Lessons from the Boom of Temporary Jobs in Spain." The Economic Journal 112 (480): F270­95. London: Royal Economic Society. Dorenbos, Ruud. 1999. Labor Market Adjustments in Hungary and Poland. Groningen: University of Groningen. Duryea, S., G. Marquez, C. Pagés, S. Paternostro, and S. Scarpetta. 2005. "Assessing Informality in Transition and Emerging Economies." Photo- copy, World Bank, Washington, DC. Earle, John S., and Zuzana Sakova. 2000. "Business Start-Ups or Disguised Unemployment? Evidence on the Character of Self-Employment from Transition Economies." Labour Economics 7 (5): 575­601. EBRD. 2003. Transition Report 2003. London: EBRD. ------. 2004. Transition Report 2004. London: EBRD. Ederveen, Sjef, and Nick Bardsley. 2003. "The Influence of Wage and Unem- ployment Differentials on Labour Mobility in the EU." In European Union, AccessLab, Workpackage 3: Analysing and Modelling Inter-regional Migration. http://accesslab.wifo.ac.at/. Elmeskov, J., J. Martin, and S. Scarpetta. 1998. "Key Lessons for Labour Mar- ket Reforms: Evidence from OECD Countries' Experiences." Swedish Eco- nomic Policy Review 5 (2): 205­52. Eriksson, Tor, and Mariola Pytlikova. 2004. "Firm-Level Consequences of Large Minimum-Wage Increases in the Czech and Slovak Republics." Labour 18 (1): 75­103. Eslava, Marcela, John Haltiwanger, Adriana Kugler, and Maurice Kugler. 2004. "The Effects of Structural Reforms on Productivity and Profitability Enhancing Reallocation: Evidence from Colombia." Journal of Development Economics 75 (2): 333­71. Estreicher, Samuel, and Stewart J. Schwab. 2000. Foundations of Labor and Employment Law. New York: Foundation Press. Estrin, Saul. 2002. "Competition and Corporate Governance in Transition." Journal of Economic Perspective 16 (1): 101­24. European Commission. 2002. "Strengthening the Co-ordination of Bud- getary Policies." Communication from the Commission to Council and the European Parliament, November 27, 2002. COM (2002) 668 final. Brus- sels: EC. 248 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union ------. 2004. Industrial Relations in Europe 2004. Luxembourg: Office for Offi- cial Publications of the European Communities. European Industrial Relations Observatory (EIRO). 2004. http://www.eiro .eurofound.eu.int/. Faggio, Giulia, and Jozef Konings. 1999. "Gross Job Flows and Firm Growth in Transition Countries: Evidence Using Firm Level Data on Five Coun- tries." Discussion Paper 2261, CEPR, London. Faini, Riccardo, Giampaolo Galli, Pietro Gennari, and Fulvio Rossi. 1997. "An Empirical Puzzle: Falling Migrations and Growing Unemployment Differ- entials among Italian Regions." European Economic Review 41: 571­79. Fatas, Antonio, Jürgen von Hagen, Andrew Hughes Hallett, Rolf R. Strauch, and Anne Sibert. 2003. Stability and Growth in Europe: Towards a Better Pact. Monitoring European Integration 13. London: CEPR. Fidrmuc, Jan. 2004. "Migration and Regional Adjustment to Asymmetric Shocks in Transition Economies." Journal of Comparative Economics 32 (2): 230­47. Flanagan, Robert. 1998. "Institutional Reformation in Eastern Europe." Industrial Relations 37 (3): 337­57. Fleisher, Belton M., Klara Sabirianova Peter, and Xiaojun Wang. 2004. "Returns to Skills and the Speed of Reforms: Evidence from Central and Eastern Europe, China, and Russia." Discussion Paper 1182, Institute for the Study of Labor (IZA), Bonn. Foster, Lucia, John C. Haltiwanger, and Cornell J. Krizan. 2001. "Aggregate Productivity Growth: Lessons from Microeconomic Evidence." In New Developments in Productivity Analysis, ed. Edward Dean, Michael Harper, and Charles Hulten, 303­72. Chicago: University of Chicago Press. Foster, Lucia, John C. Haltiwanger, and Chad Syverson. 2005. "Reallocation, Firm Turnover, and Efficiency: Selection on Productivity or Profitability?" Working Paper 11555, NBER, Cambridge, MA. Fretwell, David H., Jacob Benus, and Christopher J. O'Leary. 1999. "Evaluat- ing the Impact of Active Labor Programs: Results of Cross-Country Stud- ies in Europe and Central Asia." Social Protection Discussion Paper 9915, World Bank, Washington, DC. Friebel, Guido, and Sergei Guriev. 2000. "Why Russian Workers Do Not Move: Attachment of Workers through In-Kind Payments." Discussion Paper 2368, Centre for Economic Policy Research (CEPR), London. Gács, Vera, and Peter Huber. 2005. "Quantity Adjustments in the Regional Labour Markets of EU Candidate Countries." Working Paper 254/2005, Austrian Institute of Economic Research (WIFO), Wien. http://www.wifo.ac.at. Galasi, Péter, and Gyula Nagy. 2003. "A munkanélküli-ellátás változásainak hatása a munkanélküliek segélyezésére és elhelyezkedésére." (in Hungarian) ["The Effect of Changes in Unemployment Assistance on Benefit Receipt and Job Finding."] Közgazdasági Szemle 50 (7­8): 608­34. Garibaldi, Pietro, and Zuzana Brixiova. 1998. "Labor Market Institutions and Unemployment Dynamics in Transition Economies." IMF Staff Papers 45 (2): 269­308. Washington, DC: IMF. Bibliography 249 Garner, Thesia I., and Katherine Terrell. 1998. "A Gini Decomposition Analy- sis of Inequality in the Czech and Slovak Republics during the Transition." Economics of Transition 6, (1): 23­46. Gimpelson, Vladimir, and Rostislav Kapelyushnikov. 2005. Background Note for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union. World Bank, Washington, DC. Gorodnichenko, Yuriy, and Klara Sabirianova Peter. 2004. "Returns to School- ing in Russia and Ukraine: A Semiparametric Approach to Cross-Country Comparative Analysis." William Davidson Institute Working Paper 719, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Granberg, Alexander, and Ioulia Zaitseva. 2002. "Macroeconomy of the Russ- ian Regions Neighboring with the European Union." Photocopy, Council for the Study of Productive Forces, Moscow. Gray, Cheryl, Joel Hellman, and Randi Ryterman. 2004. Anticorruption in Transition 2: Corruption in Enterprise-State Interactions in Europe and Central Asia, 1999­2002. World Bank, Washington, DC. Griliches, Zvi, and Haim Regev. 1995. "Firm Productivity in Israeli Industry, 1979­1988." Journal of Econometrics 65 (1): 175­203. Grosfeld, Irena, and Jean-François Nivet. 1999. "Insider Power and Wage Setting in Transition: Evidence from a Panel of Large Polish Firms, 1988­1994." European Economic Review 34 (4­6): 1137­47. Grosfeld, Irena, and Gerard Roland. 1996. "Defensive and Strategic Restruc- turing in Central European Enterprises." Discussion Paper 1135, Centre for Economic Policy Research (CEPR), London. GUS (Central Statistical Office). 2003. Demand for Labour in 2002. Information and Statistical Papers series, Warsaw. Gwartney, James, and Robert Lawson. 2003. Economic Freedom of the World: 2003 Annual Report. Vancouver, BC: Fraser Institute. Haltiwanger, John C., Stefano Scarpetta, and Milan Vodopivec. 2003. "How Institutions Affect Labor Market Outcomes: Evidence from Transition Countries." Paper presented at the World Bank Economists' Forum, Wash- ington, DC, April 10, 2003. Photocopy. Haltiwanger, John C., and Milan Vodopivec. 2002. "Gross Worker and Job Flows in a Transition Economy: An Analysis of Estonia." Labour Economics 9 (5): 601­30. ------. 2003. "Worker Flows, Job Flows and Firm Wage Policies: An Analy- sis of Slovenia." Economics of Transition 11 (2): 253­90. Ham, John, Jan Svejnar, and Katherine Terrell. 1998. "Unemployment and the Social Safety Net during Transitions to a Market Economy: Evidence from the Czech and Slovak Republics." American Economic Review 88 (5): 1117­42. ------. 1999. "Women's Unemployment during Transition: Evidence from Czech and Slovak Micro-Data." Economics of Transition 7 (1): 47­78. Hazans, Mihails. 2004. "Determinants of Inter-regional Migration in the Baltic Countries." In European Union, AccessLab, Workpackage 3: Analysing and Modelling Inter-regional Migration. http://accesslab.wifo.ac.at/. 250 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Hinnosaar, Marit, and Tairi Rőőm. 2003. "The Impact of the Minimum Wage on the Labor Market in Estonia: An Empirical Analysis." Research Paper 8, Eesti Pank (Bank of Estonia), Tallinn, Estonia. Holzmann, Robert, and Richard Hinz. 2005. Old-Age Income Support in the Twenty-first Century: An International Perspective on Pension Systems and Reform. Washington, DC: World Bank. Hopenhayn, Hugo A. 2004. "Labor Market Policies and Employment Dura- tion: The Effects of Labor Market Reform in Argentina." In Law and Employment: Lessons from Latin America and the Caribbean, ed. James J. Heck- man and Carmen Pagés, 497­516. Chicago: The University of Chicago Press. Huber, Peter. 2003. "Intra-national Labour Market Adjustment in the Candi- date Countries." Working Paper 218/2004, Austrian Institute for Eco- nomic Research (WIFO), Wien. ------. 2004. "Regional Labor Market Developments in Transition." Back- ground Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Iara, Anna, and Iulia Traistaru. 2004. "How Flexible Are Wages in EU Acces- sion Countries?" Labour Economics 11 (4): 431­50. Ilkkaracan, Ipek, and Raziye Selim. 2003. "The Role of Unemployment in Wage Determination: Further Evidence on the Wage Curve in Turkey." Applied Economics 35: 1589­98. ILO. 2002. Social Dialogue in the CIS Countries: Ten Years of Independence. National reports. Series Publications on Social and Labour Affairs in Eastern Europe and Central Asia, No. 13, in Russian. Moscow: ILO. Jurajda, Stepan, and Katherine Terrell. 2000. "Optimal Speed of Transition: Micro Evidence from the Czech Republic." William Davidson Institute Working Paper 355, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Kallai, Ella. 2003. "Determinants of Regional Mobility in Romania." In Euro- pean Union, AccessLab, Workpackage 3: Analysing and Modelling Inter- regional Migration. http://accesslab.wifo.ac.at/. Kapelyushnikov, Rostislav. 2003. "The Mechanisms of Wage Formation in Russian Industry." (in Russian) Working Paper WP3/2003/07, State Uni- versity Higher School of Economics, Moscow. Kaufmann, Daniel, Art Kraay, and Pablo Zoido-Lobaton. 2004. "Governance Matters III: Updated Indicators for 1996­2002." Policy Research Working Paper 3106, World Bank, Washington DC. Kertesi, Gábor, and János Köllő. 2001. "Economic Transformation and the Revaluation of Human Capital ­ Hungary, 1986­1999." Working Paper on the Labour Market 2001/4, Institute of Economics, Hungarian Academy of Sciences, Budapest. ------. 2003. "Fighting `Low Equilibria' by Doubling the Minimum Wage? Hungary's Experiment." Discussion Paper 970, IZA, Bonn. Keyfitz, Robert. 2004. The Labor Market in Turkey. Photocopy, World Bank, Washington, DC. Bibliography 251 Klapper, Leora, Luc Laeven, and Raghuram Rajan. 2004. "Business Environ- ment and Firm Entry." Policy Research Working Paper 3232, World Bank, Washington, DC. Kolev, Alexandre, and Catherine Saget. 2003. "Towards a Better Understand- ing of the Nature, Causes and Consequences of Youth Labor Market Dis- advantage: Evidence from South-East Europe." Social Protection Discussion Paper 0502, World Bank, Washington, DC. Köllő, János. 1998. "Employment and Wage Setting in Three Stages of Hun- gary's Labor Market Transition." In Enterprise Restructuring and Unemploy- ment in Models of Transition, ed. Simon Commander. Washington, DC: World Bank. Konings, Jozef, Hartmut Lehmann, and Mark Schaffer. 1996. "Job Creation and Job Destruction in a Transition Economy: Ownership, Firm Size and Gross Job Flows in Polish Manufacturing, 1988­91." Labour Economics, Vol. 3 (3), pp. 299­317. Kraft, Evan, and Milan Vodopivec. 1992. "How Soft Is the Budget Constraint for Yugoslav Firms?" Journal of Comparative Economics 16 (3): 432­55. Kuddo, Arvo. 2004. "Public Works in Europe and Central Asia." SPectrum 2004 (Summer): 47­48. Washington, DC: World Bank. Landesmann, Michael. 2000. "Structural Changes in the Transition Economies, 1989­1999." In Economic Survey of Europe, 2000. No. 2/3. Geneva: United Nations Economic Commission for Europe (UN/ECE). Lizal, Lubomir, and Jan Svejnar. 2002. "Privatization Revisited: The Effects of Foreign and Domestic Owners on Corporate Performance." Photocopy. Lopez-Garcia, Paloma. 2002. "Labour Market Performance and Start-up Costs: OECD Evidence." Discussion Paper 565, Centre for Economic Performance, London School of Economics and Political Science (LSE), London. ------. 2004. "Business Environment and Labor Market Outcomes in ECA Countries." Background Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Lucas, Robert E. B. 2005. International Migration and Economic Development: Lessons from Low-Income Countries. Cheltenham, U.K.: Edward Elgar Pub- lishing. Luke, Peter, and Mark Schaffer. 2000. "Wage Determination in Russia: An Econometric Investigation." Discussion Paper 143, IZA, Bonn. Martin, Roderick, and Anamaria Cristescu-Martin. 2002. "Employment Rela- tions in Central and Eastern Europe in 2001: An Emerging Capitalist Periphery." Industrial Relations Journal 33 (5): 523­35. ------. 2003. "Employment Relations in Central and Eastern Europe in 2002: Towards EU Accession." Industrial Relations Journal 34 (5): 498­509. Micevska, Maja. 2004. "Unemployment and Labor Market Rigidities in Southeast Europe." Working Paper, Global Development Network South- east Europe (GDN-SEE) and Vienna Institute for International Economic Studies (wiiw), Vienna. Mickiewicz, Tomasz, and Kate Bishop. 2003. "Wage Determination: Priva- tized, New Private and State-Owned Companies: Empirical Evidence from 252 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Panel Data." William Davidson Institute Working Paper 584, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Micklewright, John, and Gyula Nagy. 1996. "Labour Market Policy and the Unemployed in Hungary." European Economic Review 40 (3­5): 819­28. Munich, Daniel, Jan Svejnar, and Katherine Terrell. 2005. "Returns to Human Capital under the Communist Wage Grid and During the Transition to a Market Economy." Review of Economics and Statistics 87 (1, February): 100­123. Murrell, Peter. 1992. "Evolution in Economics and in the Economic Reform on the Centrally Planned Economies." In The Emergence of Market Economies in Eastern Europe, ed. Christopher C. Clague and Gordon Rausser, 35­53. Cambridge, MA: Blackwell. Newell, Andrew. 2000. "Regional Unemployment and Industrial Restructur- ing in Poland." Discussion Paper 194, IZA, Bonn. ------. 2001. "The Distribution of Wages in Transition Countries." Discus- sion Paper 267, IZA, Bonn. Newell, Andrew, and Barry Reilly. 1996. "The Gender Wage Gap in Russia: Some Empirical Evidence." Labour Economics 3 (3, October): 337­56. ------. 1999. "Rates of Return to Educational Qualifications in the Transi- tional Economies." Education Economics 7 (1, April): 67­84. Nickell, Stephen. 1997. "Unemployment and Labor Market Rigidities: Europe versus North America." Journal of Economic Perspectives 11 (3): 55­74. O'Keefe, Philip. 2004. "Labor Market Institutions and Policies in ECA." Back- ground Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. OECD. 1998. Employment Outlook. OECD: Paris. ------. 1999. Employment Outlook. OECD: Paris. ------. 2000. Employment Outlook. OECD: Paris. ------. 2003. Employment Outlook. OECD: Paris. ------. 2004a. Employment Outlook. OECD: Paris. ------. 2004b. Taxing Wages: 2002­2003. 2003 edition. OECD: Paris. ------. 2005. Trends and Recent Developments in Foreign Direct Investment. Photo- copy, OECD, Paris. http://www.oecd.org/dataoecd/13/62/35032229.pdf. ONASA. 2004. http://www.onasa.com.ba/NewsFlow/web/guest.nsf?Open- Database. Orazem, Peter F., and Milan Vodopivec. 1995. "Winners and Losers in Tran- sition: Returns to Education, Experience and Gender in Slovenia." World Bank Economic Review 9 (2): 201­30. ------. 2003. "Do Market Pressures Induce Economic Efficiency? The Case of Slovenian Manufacturing, 1994­2001." William Davidson Institute Work- ing Paper 621, University of Michigan, Stephen M. Ross Business School, Ann Arbor. Paci, Pierella. 2002. Gender in Transition. Washington, DC: World Bank. Paci, Pierella, and Barry Reilly. 2004. "Does Economic Liberalization Reduce Gender Inequality in the Labor Market? The Experience of the Transition Bibliography 253 Economies of Europe and Central Asia." Photocopy, World Bank, Wash- ington, DC. Patzwaldt, Katja. 2004. "Case Studies on Labor Market Institutions in Esto- nia, Russia, and Poland." Background Note for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Peter, Klara Sabirianova. 2003. "Skill-Biased Transition: The Role of Markets, Institutions, and Technological Change." Discussion Paper 893, IZA, Bonn. Phelps, Edmond. S. 1992. "Consumer Demand and Equilibrium Unemploy- ment in a Working Model of the Customer-Market Incentive-Wage Econ- omy." Quarterly Journal of Economics 107 (3, August):1003­32. ------. 1994. Structural Slumps: A Modern Equilibrium Theory of Unemployment, Interest and Assets. Cambridge, MA: Harvard University Press. Pierre, Gaëlle. 2004. "The Investment Climate in ECA: Views from Employ- ers." Background Note for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Pierre, Gaëlle, and Stefano Scarpetta. 2004a. "Employment Regulations through the Eyes of Employers: Do They Matter and How Do Firms Respond to Them?" Policy Research Working Paper 3463, World Bank, Washington, DC. ------. 2004b. "How Labor Market Policies Can Combine Workers' Protec- tion with Job Creation: A Partial Review of Some Key Issues and Policy Options." Background Paper for World Development Report 2005. Pissarides, Christopher. 2001. "Employment Protection." Labour Economics 8 (2): 131­59. Pissarides, Francesca, Miroslav Singer, and Jan Svejnar. 2003. "Objectives and Constraints of Entrepreneurs: Evidence from Small and Medium Size Enter- prises in Russia and Bulgaria." JournalofComparativeEconomics 31 (3): 503­31. PKPP (Polish Confederation of Private Employers). 2003. "Monitoring Malych i Srednich Przedsiebiorstw Raport z badan (Monitoring of Small and Medium-Size Enterprises Research Report)." Warsaw. Profit, Stefan. 1999. "Twin Peaks in Regional Unemployment and Returns to Scale in Job-Matching in the Czech Republic." Discussion Paper 2135, Centre for Economic Policy Research (CEPR), London. Raiser, Martin, Mark E. Schaffer, and Johannes Schuchhardt. 2003. "Bench- marking Structural Change in Transition." Working Paper 79, EBRD, London. Rashid, Mansoora, and Jan Rutkowski. 2001. "Labor Markets in Transition Economies: Recent Developments and Future Challenges." Social Protec- tion Discussion Paper 0111, World Bank, Washington, DC. Rawlings, Laura. 2004. "A New Approach to Social Assistance: Latin Amer- ica's Experience with Conditional Cash Transfer Programs." Social Protec- tion Discussion Paper 0416, World Bank, Washington, DC. Rawlings, Laura, Lynne Sherburne-Benz, and Julie Van Domelen. 2004. "Evaluating Social Funds: A Cross-Country Analysis of Community Invest- ments." Regional and Sectoral Studies, World Bank, Washington, DC. Riboud, Michelle, Carolina Sanchez-Paramo, and Carlos Silva-Jauregui. 2002. "Does Eurosclerosis Matter? Institutional Reform and Labor Market 254 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Performance in Central and Eastern European Countries in the 1990s." Social Protection Discussion Paper 0202, World Bank, Washington, DC. Römisch, G. 2001. "Regional Disparities within Accession Countries." Paper presented at the East-West Conference of the Oesterreichische National- bank, November 5­6, 2001. Rowthorn, R., and R. Ramaswamy. 1997. "Deindustrialization: Causes and Implications." Working Paper 42, IMF, Washington, DC. Rutkowski, Jan. 1996. "High Skills Pay-Off: The Changing Wage Structure during Economic Transition in Poland." Economics of Transition 4 (1): 89­112. ------. 1999. "Labor Markets and Poverty in Bulgaria." Social Protection Dis- cussion Paper 9918, World Bank, Washington, DC. ------. 2001. "Earnings Inequality in Transition Economies of Central Europe: Trends and Patterns during the 1990s." Social Protection Discus- sion Paper 0117, World Bank, Washington, DC. ------. 2003a. "Does Strict Employment Protection Discourage Job Cre- ation? Evidence from Croatia." Policy Research Working Paper 3104, World Bank, Washington, DC. ------. 2003b. "Rapid Labor Reallocation with a Stagnant Unemployment Pool: The Puzzle of the Labor Market in Lithuania." Policy Research Work- ing Paper 2946, World Bank, Washington, DC. ------. 2003c. "Why Is Unemployment So High in Bulgaria?" Policy Research Working Paper 3017, World Bank, Washington, DC. ------. 2004a. "Firms, Jobs and Employment in Moldova." Policy Research Working Paper 3253, World Bank, Washington, DC. ------. 2004b. "Labor Market Developments during Economic Transition." Background Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Rutkowski, Jan, and Marcin Przybyla. 2002. "Poland: Regional Dimensions of Unemployment." In Labor, Employment, and Social Policies in the EU Enlargement Process, ed. B. Funck and L. Pizzati, 157­75. Washington, DC: World Bank. Saavedra, Jaime. 2003. "Labor Markets during the 1990s." In After the Wash- ington Consensus: Restarting Growth and Reform in Latin America, ed. Pedro- Pablo Kucynski and John Williamson, 213­63. Washington, DC: Institute for International Economics. Sapir, Andre, Philippe Aghion, Giuseppe Bertola, Martin Hellwig, Jean Pisani- Ferry, Dariusz K. Rosati, Jose Vinals, Helen Wallace, Marco Buti, and Mario Nava. 2004. An Agenda for a Growing Europe: The Sapir Report. Oxford: Oxford University Press. Scarpetta, Stefano. 1995. "Spatial Variations in Unemployment in Central and Eastern Europe: Underlying Reasons and Labour Market Policy Options." In The Regional Dimension of Unemployment in Transition Economies, ed. Stefano Scarpetta and Andreas Wörgötter, 27­54. Paris: OECD. Scarpetta, Stefano, Philip Hemmings, Thierry Tressel, and Jaejoon Woo. 2002. "The Role of Policy and Institutions for Productivity and Firm Dynamics: Bibliography 255 Evidence from Micro and Industry Data." Economics Department Work- ing Paper 329, OECD, Paris. Scarpetta, Stefano, and Peter Huber. 1995. "Regional Economic Structures and Unemployment in Central and Eastern Europe: An Attempt to Iden- tify Common Patterns." In The Regional Dimension of Unemployment in Tran- sition Economies, ed. Stefano Scarpetta and Andreas Wörgötter, 206­33. Paris: OECD. Scarpetta, Stefano, and Milan Vodopivec. 2005. "Restructuring, Productivity, and Job Creation in Eastern Europe and the Former Soviet Union." Back- ground Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Schiavo-Campo, Salvatore, Giulio de Tommaso, and Amitabha Mukherjee. 1997. "Government Employment and Pay: A Global and Regional Perspec- tive." Policy Research Working Paper 1771, World Bank, Washington, DC. Schneider, Friedrich. 2002. "The Size and Development of the Shadow Economies of 22 Transition and 21 OECD Countries." Discussion Paper 514, IZA, Bonn. Schneider, Friedrich, and Robert Klinglmair. 2004. "Shadow Economies around the World: What Do We Know?" Discussion Paper 1043, IZA, Bonn. Singer, Slavica, Sanja Pfeifer, Djula Borozan, Natasa Sarlija, and Suncica Oberman. 2003. "What Makes Croatia a(n) (Non) Entrepreneurial Coun- try?" Global Entrepreneurship Monitor (GEM) 2002 Results for Croatia. Zagreb: CEPOR. Solanko, Laura. 2003. "An Empirical Note on Growth and Convergence across Russian Regions." BOFIT Discussion Paper 9, Institute for Economies in Transition, Bank of Finland, Helsinki. Squire, Lyn, and Sethaput Suthiwart-Narueput. 1995. "The Impact of Labor Market Regulations." Policy Research Working Paper 1418, World Bank, Washington, DC. Staines, Verdon, and Dilnara Isamiddinova. 2004. "Social Assistance in Tran- sition Countries of Europe and Central Asia." SPectrum 2004 (Summer): 21­26. Washington, DC: World Bank. Steves, Franklin, Samuel Fankhauser, and Alan Rousso. 2004. "The Business Environment in the CIS-7 Countries." Photocopy, EBRD, London. Svejnar, Jan. 1999. "Labour Market in the Transitional Central and East Euro- pean Economies." In Handbook of Labor Economics, ed. Orley Ashenfelter, and David Card, 2810­57. Vol. 3 of Handbooks in Economics, Elsevier Sci- ence Series. Amsterdam: North-Holland. ------. 2002. "Transition Economies: Performance and Challenges." Journal of Economic Perspectives 16 (1, Winter): 3­28. UNDP. 2004. W Trosce o Pracę: Raport o Rozwoju Spolecznym Polska 2004 (Working Out Employment: Human Development Report Poland 2004). Warsaw. UNICEF. 1999. "Women in Transition." Regional Monitoring Report 6, UNICEF, Florence. ------. 2000. "Young People in Changing Societies." UNICEF, Innocenti Research Centre, Florence. 256 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Vecernik, Jiri. 1995. "Changing Earnings Distribution in the Czech Repub- lic: Survey Evidence from 1988­1994." Economics of Transition 3 (3): 355­71. Verme, Paolo. 2004. "Constraints to Growth and Job Creation in Low Income CIS countries." Background Paper for Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union, World Bank, Washington, DC. Vodopivec, Milan. 1991. "The Labor Market and the Transition of Socialist Economies." Comparative Economic Studies 33 (2): 123­58. ------. 2004. Income Support for the Unemployed: Issues and Options. Regional and Sectoral Studies. Washington, DC: World Bank. World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy. New York: Oxford University Press. ------. 1998. "FYR Macedonia: Country Economic Memorandum: Enhanc- ing Growth." Report 18537-MK, World Bank, Washington, DC. ------. 1999. FYR Macedonia: Focusing on the Poor. 2 vols. Washington, DC: World Bank. ------. 2000. Making Transition Work for Everyone: Poverty and Inequality in Europe and Central Asia. Washington, DC: World Bank. ------. 2001. Poland Labor Market Study: The Challenge of Job Creation. Wash- ington, DC: World Bank. ------. 2002a. Bulgaria: Poverty Assessment. Washington, DC: World Bank. ------. 2002b. "Labor Market in the Postwar Bosnia and Herzegovina: How to Encourage Businesses to Create Jobs and Increase Worker Mobility." Report 24889-BIH, World Bank, Washington, DC. ------. 2002c. Lithuania: Country Economic Memorandum. Converging to Europe: Policies to Support Employment and Productivity Growth. Washington, DC: World Bank. ------. 2002d. Slovak Republic: Living Standards, Employment, and Labor Market Study. Washington, DC: World Bank. ------. 2002e. Transition: The First Ten Years: Analysis and Lessons for Eastern Europe and Former Soviet Union. Washington, DC: World Bank. ------. 2003a. "A Study of Informal Labor Market Activity in the CIS-7." World Bank, Washington, DC. ------. 2003b. Croatia: Country Economic Memorandum. A Strategy for Growth through European Integration. Washington, DC: World Bank. ------. 2003c. FYR Macedonia: Country Economic Memorandum: Tackling Unem- ployment. Washington, DC: World Bank. ------. 2003d. Latvia: Country Economic Memorandum: The Quest for Jobs and Growth. Washington, DC: World Bank. ------. 2003e. Moldova: Public Economic Management Review. Washington, DC: World Bank. ------. 2003f. Russian Labor Market: Moving from Crisis to Recovery. Europe and Central Asia Region, Human Development Sector Unit. Washington, DC: World Bank. Bibliography 257 ------. 2004a. Moldova Poverty Assessment: Recession, Recovery, and Poverty in Moldova. Washington, DC: World Bank. ------. 2004b. Poland: Convergence to Europe: The Challenge of Productivity Growth. Investment Climate Assessment. Warsaw: World Bank. ------. 2004c. World Development Indicators 2004. Washington, DC: World Bank. ------. 2004d. World Development Report 2005: A Better Investment Climate for Everyone. Washington, DC: World Bank. ------. 2005a. "Azerbaijan: Enterprise Restructuring and Labor Redeploy- ment." Photocopy, World Bank, Washington, DC. ------. 2005b. Growth, Poverty, and Inequality: Eastern Europe and the Former Soviet Union. Washington, DC: World Bank. ------. 2005c. "Ukraine Jobs Study. Fostering Productivity and Job Cre- ation." Photocopy, World Bank, Washington, DC. ------. Forthcoming. Enhancing Gains through International Labor Migration in Europe and Central Asia. Washington, DC: World Bank. World Economic Forum. 2002. Global Competitiveness Report 2002­2003. Wyplosz, Charles. 2002. "Fiscal Discipline in EMU: Rules or Institutions?" National Institute Economic Review 191 (1): 64­78. Yoon, Yang-Ro, Barry Reilly, Gorana Krstic, and Sabine Bernabč. 2003. "A Study of Informal Labour Market Activity in the CIS-7." Paper prepared for the Lucerne Conference of the CIS-7 Initiative, January 20­22, 2003. Index administrative barriers, 42, backward areas, 81­82 53­54, 169­172 bargaining. See wage bargaining middle income CIS, 50­52 bargaining power, trade unions, administrative costs, 32­33 197­200 agriculture and agricultural bargaining structures, 204­206 regions, 129­131 Belarus jobs, 13 business start-up costs, 33 low income CIS, 79 obstacles to business opera- unemployment, 83 tion and growth, 32, 164 Albania, start-up costs, 60n.37 blue-collar jobs, 13 ALMP. See labor market pro- Bulgaria grams, active business environment, 158 Armenia obstacles to business opera- obstacles to business opera- tion and growth, 31, 163 tion and growth, 33, 165 unemployment, 12 union density, 238n.6 wages, 122 Azerbaijan, 3 workfare program, 229 business start-up costs, 33 business environment, 4 obstacles to business opera- business operation tion and growth, 33, 165 cost of, 25, 27­35 wage differentials, 92 obstacles, 30, 31, 58n.20 259 260 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union business environment, (contin- Commonwealth of Independent ued) States (CIS) start-up costs, 32­33, 45, employment, 113 60n.37 employment structure, 127 investment climate reform, capital 184­185 formal institutions, 175 labor market institutional formation, gross fixed reforms, 236­237 (GFCF),115, 116, 117 minimum wage, 92 investment, new, 115 wage determination, Central and Eastern Europe 203­204 (CEE),55, 64 wage inequality, 91 employment structure, 127 wages, 114 high-technology industries, commuting, 85 141 competition, 34, 35, 49, 62 inflation, 112 SEE, 48­50 investment climate reform, competitiveness, 56n.5 183­184 conditional cash transfers, 226 labor costs and labor supply, contestability, 153­154n.7 44­48 contracts labor market institutional liberalization, 47 reforms, 234­235 temporary, 210­211, manufacturing and produc- 240nn.37,38 tivity growth, 137 coping strategies, 100 obstacles to business opera- corruption, 49, 159­160, tions, 163­164 181­182, 187, 191n.22 policies to improve labor mar- costs ket outcomes, 45 of doing business, 25, 27­35 taxes on labor use, 217­220 of labor, 44­48 unemployment, 66, 67 of living, 39 wage curve, 196­197 credit and access to credit, wage distribution, 92 29­30, 43, 171, 173, 175, wage inequality, 91 180, 183 wage flexibility, 203 Croatia see also Bulgaria; Croatia; credit market, 29 Czech Republic; Estonia; jobs, 13 Hungary; Latvia; Lithuania; minimum wage, 207 Poland; Romania; Slovak obstacles to business opera- Republic tion, 31, 163 Central Asia, EU vs, 62 unemployment, 60n.38 circular migration, 73 Czech Republic coefficient system, 202, 203, investment rate, 40 239n.12 jobs, 13, 15 Index 261 Czech Republic, (continued) employment protection legisla- labor productivity growth, tion, (continued) 140 enforcement, 59n.31, labor reallocation, 129 211­213, 215­217 minimum wage, 207 EU transition countries, taxes, 28 210­211 unemployment, 66, 67 job creation and destruction, wage differentials, 92 172 wages, 122 relations, 47, 153n.1 revision, 42 decomposition, 153n.4 structure, 215­217 deregulation, 156 subgroup variation, 214 disemployment, 240n.34 typology, 37 distortion index, 153n.3 employment rates, 71­75, dual labor markets, 98 75­76, 100, 241n.64 investment climate and, 169 earnings dispersion. See wage male, 76­77 dispersion service sector, 168 education, employment and, 84 employment structure, 126­127 educational wage premiums, 90 economic development and, employers, 42 128 obstacles to firms' operation employment-to-working-age and growth, 158­167 ratio, 2, 11, 76 employment, 54 enterprise restructuring. See female, 14, 77, 80­81, restructuring 103n.8 entrepreneurs gender gap, 77 Bulgaria, 158 investment climate and, obstacles to firms' operation, 167­169 158­167 macroeconomic policy and regional variation, 162­165 output, 119­120 entry rate, 133 measurement, 103n.11 Estonia Moldova, 69 job creation, 23 output and, 108­118 job flows, 8, 9 prospects, 118, 120­123 jobs, 14 sector distribution, 126 wage bargaining, 200 employment protection legisla- European Union tion (EPL) , 4, 26, 36­38, accession countries, 55 42, 49­50, 51, 53, 59n.30, Central Asia vs, 62 60n.36, 196, 209­221, employment rate, 99 234­236, 240n.35 member countries, 55 construction of indexes, taxes on labor use, 220 237­238 exit costs, 187 262 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union exit rate, 133 gender labor force, 57n.8, 77 finance, access to, 28­29, 42, 43, pay differential, 16 159­160, 169­172, 179, see also women 187, 190n.3 geopolitical regions, 64­65 financial system, 191n.16 differences, 6­7 firm-level database, interna- government effectiveness, 65 tional, 132­133 labor market disparities, firms 17­19 creation and expansion, 157 unemployment rates, 81­82 creation, destruction, and Georgia productivity, 143­145 competition, 34 data, 133 obstacles to business opera- demographics, and labor real- tion and growth, 33, 165 location, 23­24 entering and exiting, number health insurance systems, of, 149­151 232­233 entry, conditions and job cre- high-technology industries, 141, ation, 151­152 146 foreign-owned privatized, housing 149 benefits, 88 job creation and, 131­137, market inefficiencies, 19, 149 87­88 obstacles to activity, 28 human capital, investments, 48 productivity and output and, Hungary 137­147 employment, 103n.7 restructuring, 19­23, 131­147 investment climate, 179 small, investment climate, job creation, 23 165, 166 job flow rates, 136 turnover, 23, 24 labor productivity growth, fiscal deficits, 58n.18 140 fiscal policy, 116­118 taxes, 28 foreign direct investment, unemployment, 2, 66, 67 60n.35 unemployment benefits, foreign privatization, 23 59­60n.33 foreign-owned privatized firms, wages, 122 job creation and, 149 formal sector, 239n.16 industry, 129­130 institutions to support, CIS, downsizing, 128, 130 52­54 high-technology, 141, 146 structural change, 199 low income CIS, 79 productivity and staffing, 139, GDP, 27, 118, 119­121, 153n.1 141, 142 Index 263 inequality, 56n.3 job creation, 3, 4, 45­46, 115, inflation, 110, 112 133 employment, 113 determinants, 172 informal sector employment, EPL and, 36 12­13, 81, 87, 93, 94­95, firm productivity and, 22 102n.2, 103n.11, 105n.29 firm restructuring and, patterns, 95 131­137 informality, strict EPL and, 216 fiscal policy and, 58n.19 infrastructure, 53, 59n.25, 85 foreign-owned privatized instability, 158­159 firms, 149 institutions, 24­35, 52­53, 156 geopolitical location, 8 challenges, 42­43 high costs and, 27 geopolitical location and, 6 investment climate and, job creation and, 193­242 155­191 role, 194, 195­196 job destruction and, 7, 8, 9, interest rates, 114­116, 135 123nn.4,7, 191n.20 labor market policy and insti- fiscal policy and, 26­27 tutions, 193­242 investment climate, 24, 27, 30, low-income CIS, 78­79 45, 60n.34 market regulation and, 171 CEE, 48­49, 183­184 new firms, 23, 25, 134­137 CIS, 184­185 policies, 39­56, 165­167 employment and, regression productivity and restructuring analysis, 186­188, and, 125­154 190­191n.7­11 reforms, 185­186 indicators, 188­189 Turkey, 68­69 information, 190n.1, urban areas, 82­85 191n.14 job destruction, 23­24, 131­134 international comparisons job creation and, 135 and within-region varia- job flow, 134, 230 tion, 173­181 rates, 136 job creation and, 155­191 jobless growth, 5, 26, 110 job creation, regional, jobs 167­172 changing nature of, 7, 12­14, new firms, 152 15, 93­97 reform, 183­186 newly created, 30­31 SEE, 184 opportunities, 100 small firms, 165, 166 productive, 8­12 subgroup variation, 175­177 protection, 36 investments, 4 reallocation, 4, 6, 13, 133 economic growth and job cre- search assistance, 230­231 ation and, 40 security, 93, 216 rate, by country, 41 stability, 34 264 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union jobs, (continued) changing nature of jobs, turnover, 134 12­14, 15 see also employment protec- CIS, 236­237 tion legislation disparities, 8, 17­19, 97­98 dual, 98 Kazakhstan dynamic, 71­74 minimum wage, 207 flexibility, 156 obstacles to business opera- geographical disparities, tion and growth, 32, 164 17­19 knowledge-driven economy, imbalances/disparities, 85, 83 56n.5 institutional reforms, Kyrgyz Republic 234­237 credit market, 29 macroeconomic variables, jobs, 13, 15 111 labor reallocation, 129 outcomes, 97­99 obstacles to business opera- persistent unemployment, tion and growth, 33, 165 8­12 policy, job creation and, labor costs, 117 193­242 labor demand, 58n.17, 70, programs, 222­233, 236, 123n.5 241­242nn.65,69 drivers of, 19­24 programs, active, 50, 231­233 labor force regional disparities, 101­102, female, 14, 57n.8, 80­81, 103­104nn.13,14,16 103n.8 role, 194, 195­196 hiring and firing, 36, 37, 51, SEE, 235­236 213 segmentation, 101 less-skilled, 12­14, 77­81, wage growth, inequity, and 102, 209 low-paid jobs, 14, 16­17 male, 76­77 labor mobility, 85­88 mobilizing, CEE, 44­48 labor productivity growth, 133, participation, 70 140 productivity, 56n.3 labor reallocation, 21, 52, rates, 9 129­130 reallocation, 4­5 firm demographics and, skills, 48 23­24 underutilization, 99­100 productivity growth and, 138 withdrawal, 70­71 labor regulations, 160­162, 163, young workers, 77­81 190nn.5,6 labor markets, 5­19, 54, 56, Latvia 61­105 credit market, 30 adaptable, 4­5 job creation, 23 CEE, 234­235 job flow rates, 136 Index 265 Latvia, (continued) manufacturing labor productivity growth, jobs, 13 140 productivity growth, 137­139 wages, 122 service sector jobs vs, 96­97 law enforcement, SEE, 48­50 market discipline, firms and, 20 legal systems, 42, 187, 199 market economy legislation, 51 economic structure and, labor, 30, 38 126­127 see also employment protec- institutions, 156 tion legislation; regulations markets liquidity, 88 credit and, 171 constraints, 19, 104n.19 disparities, regional, 81­82 Lisbon agenda, 56n.5 regulations, 187, 191n.13 Lithuania services, 129­130, 171 investment climate, 191n.12 see also urban areas labor productivity growth, middle income CIS, 55 140 administrative barriers, 50­52 wages, 122 inflation, 112 low income CIS, 55, 64 obstacles to business opera- inflation, 112 tions and growth, 164 institutions to support the policies to improve labor mar- formal sector, 52­54 ket outcomes, 45 job creation and growth, restructuring, 50­52 78­79 safety nets, 50­52 labor markets, 56, 66 see also Belarus; Kazakhstan; obstacles to business opera- Russia; Ukraine tions, 165 migration policies to improve labor mar- circular migration, 73 ket outcomes, 45 internal, 18­19, 86­87, see also Armenia; Azerbaijan; 104n.16 Georgia; Kyrgyz Republic; international, 71, 72­73 Moldova; Tajikistan; receiving country, 73 Uzbekistan return migration, 73 low-skilled workers, 12­14, sending country, 73 77­81, 102, 209 minimum wage, 17, 35, 49, 92, 195­196, 206­207, Macedonia 240n.33 minimum wage, 207 employment and, 209 unemployment, 66, 67 negotiated, 207 macroeconomic policy, 25, social assistance benefits and, 26­27 227 employment and output and, social benefits and, 207­208 119­120 subminimum wage, 46­47 266 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union minimum wage, (continued) policy, (continued) Ukraine, 207 fiscal, 116­118 wages and, 206 job creation, 39­56, 165­167 Moldova, 3 labor market outcomes, 45 employment, 69 macroeconomic, 107­123 job creation and, 22 priorities, 44, 45 obstacles to business opera- poverty, unemployment and, 3 tion and growth, 33, 165 priorities constraints, 46 new firms, 154nn.9,10 policy, 44, 45 entry barriers, Romania, 176 private sector job creation and, 134­137 employment, 169­170 productivity and, 146­147 spillover effects, 202­204, start-up cost, 177­179 239n.17 privatization, 23, 34­35, 152, OECD countries 154n.8, 9 taxes on labor use, 218­219 domestic, 59n.27, 148 unemployment benefits, existing firms and, 147­148 222­223 foreign, 148 output, employment and, productivity, 57n.9, 108­118 104­105n.22, 23 macroeconomic policy and, firm turnover and, 24, 25 119­120 output per capita growth, 109 restructuring and job creation pensions and pension systems, and, 125­154 38, 232­233, 242n.71 productivity growth, 88, Poland 153­154n.7 competition, 34 between component, 143 investment rate, 40 decomposition of, 143 jobless growth, 5 firm creation and destruction, jobs, 14 143­145 labor demand, 58n.17 labor reallocation and, 138 labor productivity growth, 140 privatization and, 148 unemployment, 2, 12 sources, 145 unemployment benefits, wages and, 139, 141, 142 59­60n.33 within component, 143 wage distribution, 16 profitability, 62 wages, 46, 122 property rights, 156 policy, 24­39, 42­43 public sector challenges, 43­44 pay determination, 239n.14 constraints, 46 wage schedule compression, employment-output link, 202 114­118 public works, 228­229, 231 Index 267 reallocation, labor-productivity Russia, (continued) growth and, 144 job flow, 8, 9, 136 reforms, 44 jobs, 14 payoffs, 2 labor market, 103n.10 within-enterprise, 20­22 labor reallocation, 21 regulations, 169­172, 179­181, obstacles to business opera- 191n.13, 211 tion and growth, 32, 164 labor, 160­162, 163, reallocation, labor-productiv- 190nn.5,6 ity growth and, 144 time spent, 173, 174 wages, 92, 104­105n.23, see also employment protec- 239n.20 tion legislation; legislation rent, 59n.29 safety nets resource reallocation, 19, 20 informal, 19 new firms, 152 middle income CIS, 50­52 restructuring, 30­31, 121­123 sectoral bargaining, 205­206 defensive, 20­21, 57n.14, 139 security costs, 173, 174 existing firms and, 147­149 self-employment, 12­13, 93, 96, middle income CIS, 50­52 102n.2, 103n.11, 231­232 output-jobs link, 112­113 Serbia & Montenegro productivity and job creation start-up costs, 60n.37 and, 125­154 wage differentials, 92 strategic, 22­23, 57n.14, 162 service sector, 13 within-industry, 137­139 employment rate, 168 retirement, 38 investment climate and, 169 return migration, 73 manufacturing jobs vs, risk reduction, 33­34 96­97 Romania services credit market, 30 nonmarket, 129­130 job creation, 23 see also market services job flow, 9, 136 severance pay, 51, 60n.39 labor productivity growth, size threshold, 133 140 skills labor reallocation, 130 deficient, 162, 163 new firm entry barriers, 176, jobs and, 13­14 154nn.9,10 mismatch, 19, 57n.6, 86­87, obstacles to business opera- 105n.30 tion and growth, 31, 163 Slovak Republic taxes, 28 investment rate, 40 unemployment, 66, 67 labor productivity growth, wages, 122 140 Russia unemployment, 2, 12 coefficient system, 239n.12 wages, 122 268 Enhancing Job Opportunities: Eastern Europe and the Former Soviet Union Slovenia Southeastern Europe (SEE), job creation, 23 (continued) job flow, 8, 9, 136 unemployment, 66, 67 labor productivity growth, wage bargaining, 204 140 see also Albania; Bosnia & private sector spillover effects, Herzegovina; Macedonia, 239n.17 FYR; Serbia & Montenegro unemployment, 66, 67 stability, 183 wage differentials, 92 SEE, 48­50 wages, 122 staffing, 139 social assistance programs, start-up costs, 32­33, 45, 225­227 60n.37, 177­179, 187 wage and labor turnover out- state-owned firms, 205 comes and, 227­228 social benefits, 19, 26, 38­39, Tajikistan, obstacles to business 57­58n.16 operation and growth, 33, minimum wages and, 165 207­208 targeting the poor, 226 social dialogue, 195 taxes, 27­28, 29, 45, 49, 51, social funds, 226 58­59nn.21,22, 156, social partnerships, 42 159­160, 169­172, 183, 187 social protection, 54, 237 CEE, 217­220 mechanisms, 87 collection, 242n.72 social safety nets, 4­5, 51­52 labor use, 217­221, 235, social security threshold, 241nn.47,48 240n.30 payroll, 28, 236 social transfers technological advances, alternatives, 236 153­154n.7 targeting, 52 telephone service, unemploy- Southeastern Europe (SEE), 55, ment and, 85 64 temporary employment, 13 competitive markets, 48­50 trade unions, 35, 42, 59n.32, 195 inflation, 112 bargaining power, 197­200 investment climate reform, density, 205, 238nn.5,6 184 state, 239n.10 labor market institutional training, 231, 57n.6 reforms, 235­236 transition economies, 1 law enforcement, 48­50 taxonomy, 56n.1 obstacles to business opera- transition shock, 24 tions, 164 Turkey, 56n.1 policies to improve labor mar- job creation, 68­69 ket outcomes, 45 unemployment, 11­12 stability, 48­50 Turkmenistan, 56n.1 Index 269 Ukraine, 3 wage bargaining, (continued) job flow rates, 136 Estonia, 200 minimum wage, 207 EU countries, 239n.19 obstacles to business opera- SEE, 204 tion, 32, 164 trade unions, 197­200 uncertainties, 158­159 wage determination, 51, 53, underemployment, 63­75 194, 235 job creation and destruction CIS, 203­204 and, 135 public sector, 239n.14 unemployment, 9­12, 60n.38, wage disparities, 3­4, 57n.12, 63­75, 105n.32 92­93, 105n.27 flows into and out of, 74 bargaining and, 205 inflated figures, 224 wages, 49, 57n.12, 59n.28, job creation and destruction 100­101, 104n.20, 21, 117 and, 135 adaptability, 46 long-term, 47, 48, 68, 70 adjustments, 113­114 open, 100 CEE, 203 outflows, 103n.9 differentials, 89­93, 105n.31 persistent, 3, 8­12, 71­75, distribution, 14, 16, 89 97­99, 100 flexibility, 25, 35­36, 92­93, registered, 241n.52 203 unemployment benefits, 38, 47, gender gap, 81 50, 51, 59­60n.33, growth, 7­8, 14 222­223, 235, 241n.54 inequality, 16, 18, 90­91, 98, coverage, 223­224 101, 105n.26 employment and, 224­225 output-jobs link, 112­113 noncash benefits, 224 productivity and, 57n.9, 141, replacement rate, 222 142 unemployment rate, 103n.12 public sector, 122 regional disparities, 81­82 rebounding, 88­93 "unemployment trap," 47 Russia, 239n.20 unions. See trade unions schedules, compression, 202 urban areas, job creation, 82­85 subsidies, 231 Uzbekistan "wage curve," 196, 105n.28 business start-up costs, 33 wage-setting mechanism, 14 obstacles to business opera- white-collar jobs, 13 tion and growth, 33, 165 women, employment, 14, 77, 80­81, 103n.8 wage bargaining, 195, 205, workers; workforce. See labor 239­240n.24 force bargaining structures, 204­206 workfare, 228­229 change within subgroups, 200­201 young workers, 77­81 This report is part of a series undertaken by the Europe and Central Asia Region of the World Bank. The series draws on original data, the World Bank's operational experience, and the extensive literature on the Region. Poverty, jobs, trade, migration, and infrastructure will be among the topics covered. "This is an impressive book that covers a wide range of important policy-related issues. The book is an important contribution to our policy-oriented knowledge of the functioning of labor markets in Eastern Europe and the Former Soviet Union." Jan Svejnar, Everett E. Berg Professor of Business Administration and Professor of Economics, University of Michigan "I am very impressed by this study on enhancing job opportunities in transition economies. It allows better understanding of the various dimensions of unemployment and lack of job creation." Michal Boni, expert for the CASE Foundation, former Minister of Labor of Poland "This is very complex and well-documented research on one of the key issues that all transition countries are facing. The creation of `good' jobs remains a serious problem. The labor market adjustment story that the study is telling us is logical, quite convincing, and quite complete." Vladimir Gimpelson, Director, and Rostislav Kapelyushnikov, Deputy Director, Center for Labor Market Studies, Higher School of Economics, Moscow E nhancing Job Opportunities: Eastern Europe and the Former Soviet Union addresses why labor market outcomes have been disappointing during the transition and suggests policy interventions that can foster job creation and reduce unemployment. The creation of more and better jobs requires firms to engage in strategic restructuring to a greater degree, so as to turn productivity gains into investments and expansion. The authors recommend that countries in the Region pursue a two-pronged strategy: first and foremost, lower the costs of doing business to encourage investment and firm growth; and second, develop an adaptable labor market, where employers have incentives to hire and workers have incentives and skills to take available jobs. ISBN 0-8213-6195-3