The Russian Labor Market: Moving from Crisis to Recovery A copublication of Izdatelstvo Ves Mir and the World Bank © 2003 The International Bank for Reconstruction and Development/ THE WORLD BANK 1818 H Street, NW Washington, D.C. 20433, U.S.A. All rights reserved Printed in the Russian Federation A copublication of the World Bank and Izdatelstvo Ves Mir Izdatelstvo Ves Mir 9a olpachny Per. 101831 Moscow Center Russian Federation The findings, interpretations, and conclusions expressed here are those of the author(s) and do not necessarily reflect the views of the Board of Execu tive Directors of the World Bank or the governments they represent. The World Bank annot 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 on the part of the World Bank any judgement of the legal status of any territory or the endorsement or accept ance of such boundaries. Rights and Permissions The material of this work is copyrighted. N part of this work may be repro duced or transmitted in any form or by any means, electronic or mechani cal, including photocopying, recording, or inclusion in any information storage and retrieval system, without the prior written permission of the World Bank. The World Bank encourages dissemination of its work and will normally grant permission promptly. For permission to photocopy or reprint, please send a request with com plete information to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978 750 8400, fax 978 750 4470, www.copyright.com. ISBN: 5 7777 0274 0 All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818, H Street NW, Washington, DC 20433, fax 202 522 2422, e mail: ubrights@worldbank.org. Library of Congress Cataloging in Publication Data has been applied for. CONTENTS Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .VII Acronyms and Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .VIII Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .IX I. Understanding Employment: Level, Composition, and Flows . . . . . . . . . .1 A. Setting The Stage: Macroeconomic Developments . . . . . . . . . . . . . . . . . . . . . . .2 B. Aggregate Labor Market Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 C. Understanding Aggregate Employment Fluctuations . . . . . . . . . . . . . . . . . . . . .6 D. Adjustment through Hours of Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15 E. Time Allocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18 F. Understanding Unemployment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20 G. Structural Change And Labor Mobility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 H. The New Private Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35 II. Understanding Wages: Structure, Uncertainty, and Inequality . . . . . .45 A. Level and Determinants of Wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .45 B. Nonwage Compensation Practices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .51 C. Wage Inequality And Poverty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .63 III. Labor Market Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 A. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .69 B. Labor Contracts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .72 C. Dismissals and Terminations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .76 D. Wage Determination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .82 E. Trade Unions, Employer Organizations, and Collective Bargaining . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .87 F. Enforcement and Dispute Resolution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .93 IV. Safety Nets for Workers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103 A. Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .103 B. Evaluation of Unemployment Protection Programs in Russia . . . . . . . . .104 C. Policy Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .131 Boxes Box I.I Who Is Unemployed? Some definitional problems . . . . . . . . . . . . . . . . .13 Box IV. 1 The Unemployment Benefit System in Russia . . . . . . . . . . . . . . . . . .106 IV THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Box IV. 2 ALMPs in Russia: A Brief Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .119 Box IV. 3: Profiling to Reduce Long Term Unemployment . . . . . . . . . . . . . . .123 Figures Figure I.1: Real GDP (1990=100), Russia and Select CEE countries . . . . . . . . .3 Figure I.2: Real GDP, Employment, Real Wages, and Labor Productivity . . . .3 Figure I.3: Real Wages in Russia and Select CEE Countries . . . . . . . . . . . . . . . . . . .5 Figure I.4: Alternative Measures of Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . .7 Figure I.5: Employment Trends in Russia and Select CEE countries . . . . . . . .8 Figure I.6 Employment Rates in Russia and Select CEE countries . . . . . . . . . .9 Figure I.7: Unemployment Rates in Russia, Select CEE countries, 1993 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11 Figure I.8: Changes in Labor Force Participation Rates, 1993 2000 . . . . . . .12 Figure I.9: Percentage of Long Term Unemployed, Russia/CEE . . . . . . . . . . .22 Figure I.10: Self Employed as a Share of Employed, Select Countries . . . . .37 Figure I.11: Evolution of Non Agricultural Self Employment . . . . . . . . . . . . . .38 Figure I.12: Foreign Direct Investment in Selected Transition Economies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Figure I.13: World Business Environment Survey (WBES) of Obstacles to Investment in Russia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .39 Figure I.14: Payroll Tax Rates in Russia, CEE Countries, EU, and OECD . . .40 Figure II.1: CPI Deflated Real Wage Rate Due, 1991 2001 (Jan. 1991 = 100) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .46 Figure II.2: Real Wage Arrears, 1990 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .53 Figure II.3: Lorenz Curve for Wages, 1998 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . .65 Figure III.1 Minimum Wage as a Proportion of Average Wage, Transition Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .86 Figure III.2: Percentage of Employees Reporting They Can Be Dismissed Without Any Formal Grounds by Sector, Year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .95 Figure III.3: Number of Person Days Lost in Strikes, 1995 99 . . . . . . . . . . . . . .97 Figure IV.1 Simulation of Unemployment Benefit Expenditures . . . . . . . .137 Tables Table I.1: The Socioeconomic Composition of the Unemployed, 1999 . .21 Table I.2: Regional Unemployment Rates: Standard Deviation and Max/Min Ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23 Table I.3: Labor Market Transitions, 1998 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . .26 Table I.4: Job and Occupational Mobility (Self Reports in the RLMS) . . . . .28 Table I.5: Hiring, Layoff, Quit, and Separation Rates, 1991 98 . . . . . . . . . . . . .29 Table I.6: Composition of Employment by Industry, 1990 99 . . . . . . . . . . . . .30 Table I.7: Job Flows over Time, Russia and select Transition Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .33 Table I.8: Share of Employment in New Private Sector, 1994 2000 . . . . . . .35 Table II.1: Changes in Real Wages by Characteristics of Firms and Workers, 1998 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47 CONTENTS V Table II.2: Results of Simple Earning Functions, RLMS, 1992 2000 for Women . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .48 Table II.3: Provision of Fringe Benefits, by Type, 1990 1998 . . . . . . . . . . . . . .52 Table II.4: Incidence and Magnitude of Forced In Kind Substitutes for Wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .61 Table II.5: Poverty Rates by Socioeconomic Groups . . . . . . . . . . . . . . . . . . . . . . .66 Table III.1: Forms of Labor Contracts by Sector, ISITO Survey, April 1998 . . . . .73 Table III.2: Provisions Stipulated in Contracts (Permanent and Fixed Term) by Property Form, Employer Reports (n=278), 1999 . . . . . . . . . . . . . . .75 Table III.3: Reasons for Leaving Previous Job, by Sector of Previous Employment, Kemerovo Oblast and Komi Republic, October 1997 . . . . .79 Table III.4: Official Minimum Wage and Average Monthly Wage, 1995 2000 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .83 Table III.5: Percentage of Employees by Sector, Reporting Guarantees Stipulated by Legislation or Contract Are Not Fully Provided, 1999 . . . . . . .84 Table III.6: Forms of Wage Payment by Sector, Kemerovo and Komi, 1997 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .85 Table III.7: Who Protects the Employees? Opinions of Employees, Employers, and Trade Union Leaders, 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .89 Table IV.1: Employment Fund Budget (Percent of GDP) and Arrears (Millions of Rubles) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .105 Table IV.2: Share of Benefit Expenditures in Total Unemployment Program Expenditures . . . . . . . . . . . . . . . . . . . . . . . . . .107 Table IV.3: Replacement Rate of Unemployment Benefit . . . . . . . . . . . . . . . .108 Table IV.4: Minimum Unemployment Benefit . . . . . . . . . . . . . . . . . . . . . . . . . . . .109 Table IV.5: Trends in Applicants and Registered Unemployed . . . . . . . . . . .111 Table IV.6: Poverty Impact of Unemployment Programs in Select Transition Economies, Mid 1990s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .113 Table IV.7: International Assessment of Unemployment Benefit Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .115 Table IV.8: Effectiveness of Active Labor Market Programs: International Evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .118 Table IV.9: Employment Fund Expenditures on Active Policies, 1992 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .120 Table IV.10: The Socioeconomic Characteristics of Applicants 1993 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .121 Table IV.11 Jobs Placement of Unemployed Completing Training Programs 1993 1999 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .122 Annex I. Employment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .143 Annex II. Wages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .177 Annex III. Labor Market Regulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .191 Annex IV. Social Safety Nets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .197 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .230 Acknowledgements This report was prepared by a team comprising Mansoora Rashid (Task Team Leader), Emily Andrews, Gordon Betcherman, Harry Broadman, John Earle, Arvo Kuddo, Francesca Rescanitini, Klara Sabirianova, and Elena Zotova. Tatyana Tchetvernina and her team also contributed a background paper for this report. The team benefited from the advice of Russian labor market experts during its preparation, including: Alexander Ananiev, Natalia Chekorina, Lubov Eltsova, Tatyana Gorbacheva, Vladimir Gimpelson, Zoya Hotkina, Veronika Kabalina, Rostislav Kapelushnikov, Alexander Leonov, Valeri Naboishikov, Serguey Panin, Alexander Rasumov, Ludmila Rjanitsyna, Zinaida Ryshikova, Oleg Saenko, Vitali Savin, Igor Shanin, Irina Soboleva, Galina Strela, Vladimir Varov, and Ludmila Xahulina. The team would like to thank the pre senters and participants at the joint Ministry of Labor and Social Development of the Russian Federation (MLSD) / World Bank workshop on this study for their comments and suggestions. We are also grateful to Alexander Pochinok, Minister of Labor and Social Development of the Russian Federation, Mikhail Dmitriev, First Deputy Minis ter, Ministry of Economic Development and Trade of the Russian Federation (MEDT), Eugene Gontmakher, Director, Department of Social Development, Russian Federa tion Government Office, Maxim Topilin, Deputy Minister, MLSD, and Marina Moskv ina, former Deputy Minister, MLSD, for their support of this work and for their com ments to earlier drafts of the report. This work has also benefited from comments from Anders Aslund, Anastassia Alexandrova, Robert J. Anderson, Vladimir Debrentsov, Brian Pinto, Ana Revenga and Ruslan Yemtsov. The report was prepared under the guidance of Michal Rutkowski, Sector Manager, ECSHD, World Bank, and Annette Dixon, Sector Director, ECSHD, World Bank. VIII THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Acronyms and Abbreviations ALMP Active Labor Market Program BLR Balance of Labor Resources BLS Bureau of Labor Statistics CEE Central and Eastern Europe CEET Central and Eastern European Team CIS Commonwealth of Independent States CPI Consumer Price Index EBRD European Bank for Reconstruction and Development FDI Foreign Direct Investment FIAS Foreign Investment Advisory Service FTE Full Time Equivalent FSU Former Soviet Union GDP Gross Domestic Product ICFTU International Confederation of Free Trade Unions ILO International Labour Organization IMF International Monetary Fund ISA Individual Saving Account ISITO Institute for Comparative Labor Relations Research LFPR Labor Force Participation Rate LFS Labor Force Survey MLS Minimum Living Standard MLSD Ministry of Labour and Social Development of the Russian Federation NDC Notional Defined Contribution OECD Organization for Economic Cooperation and Development OLS Ordinary Least Squares REB Russian Economic Barometer RLFS Russian Labor Force Survey RLMS Russian Longitudinal Monitoring Survey SIF Social Insurance Fund UI Unemployment Insurance UISA Unemployment Insurance Savings Accounts VTsIOM All Russian Center for Public Opinion Research WBES World Business Environment Survey WTO World Trade Organization EXECUTIVE SUMMARY The Labor Market in Russia: Moving from Crisis to Recovery I. Overview i. One of the main challenges confronting the Russian Federation today is to increase real wages and productive employment in order to improve the standard of living of its population. This report focuses on labor market policy options that are important for achieving this outcome. Macroeconomic policies that promote competitive prod uct markets, raise aggregate demand for labor, and increase labor productivity may have the most critical impact on employment and wage outcomes. However, labor market policies and institutions also can affect the functioning of the labor market and the level of employment and wages. ii. The report is forward looking, in that it suggests measures to help Russia develop a formal, competitive labor market over the medium term. The study addresses four major questions: (1) How well has Russia been able to redress the mis allocation of labor inherited from its socialist past? (2) Do wages increasingly reflect market forces? (3) Are labor market institutions consistent with those required in a market economy? (4) How well has Russia been able to reduce explicit protection offered by firms and create an effective safety net? This report addresses each ques tion in a separate chapter and also highlights key issues and policy options in each area. The development of a well functioning labor market will contribute to Russia's ability to integrate with the global economy, particularly as it faces the opportunity and challenges that will come with WTO accession. iii. Attempting to evaluate the labor market in Russia, given its vastness, complex ity, and diversity, is a daunting task. This report attempts to remedy these problems in part, by providing a comprehensive picture of labor markets. We rely heavily on work by both Russian and international scholars to inform this report. We also have incor porated salient points from the considerable discussions and debate on labor market policy issues that took place in Russia during the course of this study. We have attempted to overcome data issues by using both nationally representative surveys and smaller regional surveys of enterprises, workers, and the unemployed, and by cross checking their results with each other.1 No data set is perfect; therefore, the strengths and weaknesses of each are discussed in relevant sections of the report. 1 Specifically, the analysis combines official statistics of the Russian Government, detailed findings from the Russian Labor Force Survey (RLFS), and the results from sev eral micro data sets on firms and households, including a large enterprise survey, firm registries, and the Russian Longitudinal Monitoring Survey (RLMS). X THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure 1. Real GDP, Employment, Real Wages, and Labor Productivity 110 100 90 employment 80 and 70 1990=100 60 wages 50 index: 40 real 30 GDP, 20 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Years Employment Real wages Real GDP Labor productivity See Chapter I. iv. Russia experienced a severe recession through most of the 1990s, and the effects were felt in the labor market, with a sharp decline in real wages and a rise in unemployment. Nevertheless, the report finds that Russia made some progress in moving to a market based allocation of labor over this period. The allocation of labor across industries, occupations, and sectors has moved toward that found in a market economy. Unemployment rates have increased over the 1990s as the economy has restructured and enterprises have downsized. Job destruction rates in manufacturing have also substantially increased over the 1990s confirming this trend, and this real location has worked to raise total productivity. The determination of wages also has started to reflect market forces: The returns to education have markedly increased (although returns to vocational education have fluctuated over time), and the returns to work experience in the socialist era have declined. These changes are explained in part by the downsizing of enterprises and the growth of the private sector. v. Recent economic growth has demonstrated the ability of the Russian labor mar ket to respond rapidly to economic growth. Real wages, employment, and labor pro ductivity have increased, with an average annual rate of growth similar to that real ized by Central and Eastern European (CEE) countries in their first two years of economic recovery. Wage arrears and inappropriate fringe benefits (housing, kinder gartens, etc.) also have sharply declined, although they have not completely disap peared. The unemployment rate has declined sharply more so than in CEE countries during a similar growth period. In accordance with the above trend, job creation trend substantially grew and job destruction rate decreased, at least in the production sector that appeared to be in the more favorable position because of the devaluation of the currency in the recent years. EXECUTIVE SUMMARY XI vi. The report cautions that a large outstanding agenda in creating a well func tioning labor market still confronts the Government, employers, trade unions, and workers. Years of slow restructuring, limited economic reforms, and lack of job opportunities have led to a decline in formal labor market activity and a shift of many employed toward subsistence self employment, primarily in agriculture. Despite recent declines, the level and duration of unemployment (ILO definition) are not low by Organization for Economic Co operation and Development (OECD) standards; and the recent slowdown in economic activity has started to gradually reverse previ ous gains in terms of unemployment trends. The employment share of the private sector remains small, and genuine entrepreneurship is limited. While in manufactur ing job creation rates have increased somewhat over the 1990s, they remain lower than in OECD countries and high income transition economies. Job destruction con tinues to dominate job flows, and net employment growth remains negative in this sector.2 Moreover, despite progress in market determination of wages, nonmarket forces (for example, wage arrears, fringe benefits, and in kind substitutes outside the norm in market economies) remain important, and wage arrears have even increased for public sector workers (health and education) in recent months. Wage inequality, already high by regional standards, has increased since 1998, and the incidence of poverty remains very high. vii. Furthermore, labor market regulatory institutions have not evolved signifi cantly since the socialist era and are generally very ineffective. The passage of the new Labor Code was a political achievement and does offer some improvements on the old law. However, the new Code is still quite restrictive relative to many OECD coun tries. Employers are limited in their ability to adjust their workforce in response to economic and technological change; workers and employers do not have adequate opportunity to voice their concerns; contract enforcement is weak; and mechanisms for resolving workplace disputes and addressing health and safety concerns are lim ited. Even though the Government created a modern safety net in the early 1990s, limited financing of this program has made the system largely ineffective, contribut ing to high rates of poverty among the unemployed (relative to national levels). viii. Weak labor market regulation means that the excessively restrictive Labor Code has not greatly constrained labor adjustment in Russia. Poor incentives and rep utational risks for employers in laying off workers have probably played a more impor tant role in constraining Russia's gradual restructuring. Although limited regulation is beneficial for labor market outcomes (e.g., job creation), the virtual absence of labor regulation enforcement "on the ground" has imposed large welfare and productivity tradeoffs: low and uncertain wages, growing wage inequality, poor health and safety standards and other contractual violations. This lack of labor regulation enforcement also has contributed to the informalization of the economy. While wage arrears and other labor violations have declined as a result of economic growth, the absence of effective enforcement and arbitration institutions means that workers remain vulner able to a recurrence of such violations should economic growth subside. 2 The studies on job flows do not include the private service sector, which is likely to have much higher rates of job creation and destruction. XII THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure 2. Real Wage Arrears Rose, and Then Fell Sharply Post 1998 450 400 350 300 250 Percentage200 150 100 50 0 92 94 94 95 95 96 96 97 97 98 98 99 99 00 r92 r93 r r r r r r r a a a a a a a a a M Sep M M Sep M Sep M Sep M Sep M Sep M Sep M Year See Chapter II. ix. The report concludes that increasing productive employment and real wages will require, first and foremost, addressing the remaining restructuring agenda and promoting sustained, private sector led economic growth. An increased emphasis on creating flexible and enforceable labor market regulatory laws and institutions, how ever, also will be critical for promoting labor productivity and improving the welfare of workers. Finally, it will be important to complement the reduced protection offered by firms with an affordable and effective public safety net for workers to allow the lowering of restrictions in the Labor Code, facilitate layoffs in strategic state sec tors, and protect workers in case of job or skill loss. x. The Government recognizes the importance of a labor market policy that pro motes efficiency but protects the basic rights of workers and has made it an impor tant component of its economic reform program. A notable achievement, after pro longed discussion and debate, has been the passage of the new Labor Code, which modernizes labor contracting practices, and the signing of a decree that makes wage arrears a criminal practice. These are steps in the right direction, but more needs to be done. Further reforms along these lines would help improve labor productivity as well as promote worker welfare. The main findings and conclusions of the report are presented in greater detail below. II. Recent Developments xi. The Russian labor market has been gradually restructuring during the past decade, mainly as a result of market liberalization, an ineffectively regulated labor market, and growth of the private sector. (Results from this section are presented in the Chapters I and II of this report). EXECUTIVE SUMMARY XIII Figure 3. The Composition of Employment by Sector of Economic Activity, 1990 99 70 60 50 employment 40 1990 of 30 1999 20 Percentage 10 0 Industry A griculture/forestry S ervice s See Chapter I. · The level and duration of unemployment gradually increased between 1990 98. The unemployment rate rose from 5.2 percent of the labor force in 1992 to nearly 15 percent in 1998. Most of the unemployed have previous work experience, confirming that the exit of workers was the main reason for unemployment growth. Increasing job destruction rates and low rates of job creation over the 1990s in the manufacturing sector are consistent with this increase in unemployment. · Among those remaining officially employed, there has been a significant reallo cation of workers. This movement reflects large shifts of workers across indus tries, occupations and sectors consistent with those found in a market econ omy. Russia now ranks at the median level for reduction in the employment share of industry and growth in the share of services, in the latter case ahead of Bulgaria, Poland, Romania, and Slovenia. This shift is also reflected in the reduced output share of agriculture and industry and the increased share of services in gross domestic product (GDP) in the past decade. These transitions can be explained in part by the decline in state employment and employment growth in mixed, domestic, private, and foreign firms. · Labor mobility across occupations also increased post 1991 and became more complex. The number of people who moved to another industry, firm, or occu pation was considerably higher during the first four years of reforms (1991 95) than during the preceding six years 1985 91). Mobility also became more "complex," more frequently involving simultaneous changes in occupation, XIV THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY firm, and industry (table 1). Another indicator of labor mobility is job tenure, which has declined in Russia (for both men and women) to some of the low est levels found in OECD countries.3 · Wage growth started to reflect a premium to education. The returns to education for women have increased from 3.8 percent (3.2 percent for men) in 1992 to 7.6 percent (6.8 percent for men) in 20004; however, the rate of return to vocational education has fluctuated over this period. As in other transition countries, the rate of return on experience and job tenure is low and negative, signaling the irrelevance of socialist era experience in the emerging labor market. Like other transition countries, the private sector (all else equal) pays more than the state sector, signaling higher worker productivity in that sector. There is a significant gender gap, with women earning less than men with similar characteristics. Moreover, the wage share of worker remuneration has increased, as the initially large fringe benefits (e.g. kindergartens, medical care) wage substitutes in the socialist era have declined over time. The biggest declines have been recorded for housing construction, kindergartens, and recreation and culture. xii. The post 1998 2000 period of 11 percent cumulative growth led to major responses from the labor market. Employment increased cumulatively by 2 percent and nonparticipation fell.5 Unemployment rates also fell sharply to 9 percent in 2001 (but have increased slightly since then). As in advanced CEE reformers, the employment response was much smaller than that of output. Employers reallocated existing labor more productively as opposed to increasing employment, and labor productivity increased (7 percent in 2000). The aver age annual growth rates in employment and labor productivity are similar to those found in CEE countries after the first two years of economic recovery, but the decline in unemployment was far greater in Russia than realized in CEE countries in this period. Labor market transitions between 1998 and 2000 were much higher than in the 1994 96 and 1996 98 period. The transition out of unemployment into employment increased significantly, and the transition to nonparticipation declined in response to 3 High hiring and separation rates in Russia during this period of economic decline remain a puzzle that requires future research. 4 Thus, women obtained a 3.8 percent gain in real wages per year of education (all else equal) in 1992, and this wage gain increased to 7.6 percent per year of schooling by 2000. 5 For assessing changes in employment, the study uses the Balance of Labor Resources (BOLR) employment series. This series represents the statistical agency's attempt to estimate employment on the basis of all available information. In contrast, the RLFS is a pure survey based measure. The report uses the RLFS for evaluating changes in the composition of the labor force aggregates and flows. The discrepancy between the two series has been the subject of some discussion: both show a similar pattern of decline until 1998 and rise thereafter. The RLFS series shows a greater responsiveness to output than the BOLR data. A possible explanation for the difference stems from the fact that until 1999 the RLFS was carried out only in particular months of the year on a somewhat irregular schedule. EXECUTIVE SUMMARY XV Table 1. Job and Occupational Mobility (Self Reports in the RLMS) Years Changed both firm Changed firm but Changed occupa Did not change and occupation not occupation tionbut not firm either firm or occupation 1996 1998 0.170 0.093 0.034 0.704 1998 2000 0.176 0.105 0.035 0.685 Note: The table shows the fractions of employed respondents who reported in 1998 and 2000 that they changed their place of work and occupation as compared with December 1996 and December 1998, respectively. Source: Calculations from RLMS (see Chapter I). positive economic activity. The probability of remaining employed increased, as did new entry to the labor market and re entry into employment from out of the labor force. Economic growth in post crisis Russia was not only strong enough to increase the probability that workers would remain employed, but it also brought the unem ployed back to employment more quickly than before, and it even pulled in labor force nonparticipants. Consistent with the decline in unemployment, job creation rates in the manufacturing sector (which benefited most from the devaluation of the currency) increased, and job destruction rates declined although the latter contin ued to dominate job flows. · Real wages lagged employment and output growth between 1998 and 2000; and wage arrears declined by half. Wages fell between 1998 1999, but increased by 22 percent between 1999 2000. The average annual increase in real wages in the two year period (1998 and 2000) was therefore negative, as wage growth lagged behind employment and output growth. This was also the case in both Poland and Hungary, however, in their first two year growth period. The decrease in wage arrears may be a result of several factors, includ ing a decree by the government criminalizing this practice, positive macroeco nomic trends, devaluation of stock resulting from the burst of inflation at the end of 1998, and the trend toward reduction of barter in the economy in the past few years. · Wage inequality increased, and there were winners and losers post 1998. The Gini coefficient for wages, which measures inequality in the distribution, increased from 0.439 in 1998 to 0.464 in 2000. The increase in wage inequal ity is the result of higher real wage gains among high wage workers relative to low wage workers. Workers who realized real wage gains are younger, highly educated, private sector, urban workers. Older, less skilled workers in rural areas and in the state sector hardly realized any increase in real wages. Thus, economic growth between 1998 and 2000 has benefited some workers more relative to others. XVI THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table 2. The Share of Employment in New Private Sector, 1994 2000* Ownership type 1994 1995 1996 1998 2000 Distribution of the employed by type of ownership State owned 0.754 0.683 0.663 0.647 0.605 Mixed 0.073 0.100 0.116 0.113 0.129 Domestic private 0.134 0.172 0.181 0.196 0.217 Foreign 0.040 0.045 0.039 0.044 0.049 Not available 0.181 0.155 0.148 0.140 0.127 * Using ownership type as definition of new private sector (see Chapter I). Source: Definition of ownership type. Calculations from RLMS. Goskomstat (2000b, p. 112). xiii. The Russian labor market is still far from the formal, private sector based labor markets typical of OECD countries. Despite recent economic recovery, much remains to be done. · Formal employment is low, and a significant share of workers is self employed in subsistence agriculture. Despite recent increases, the decline in employ ment was significant in absolute terms (but less relative to output). The largest declines in the labor force were in the youngest and oldest age groups. What happened to individuals who left employment? Some joined the ranks of the unemployed, but the majority left the labor force, of which a consider able share took up self employment, primarily in subsistence agriculture (table 2). · The state remains an important employer. Government policies that have con strained downsizing of firms, through soft budget constraints and local gov ernment pressure on enterprises to maintain jobs and services, have led to excess jobs in the state sector. While the extent of overstaffing declined during the past decade, and estimates about its size are under debate, its existence is confirmed by the small response of employment to output. Despite recent increases, the private sector share of employment remains lower than that in advanced CEE countries;6 and the share of self employment and genuine entrepreneurship is low as well.7 A recent study by Foreign Investment Advi sory Service (FIAS 2001) finds that there are considerable barriers to entry for 6 The estimates on the private sector share of employment vary by definition of employment and source of data. According to ownership definition, the private share of employment increased from 13 percent in 1994 to 22 percent in 2000; and, using founding date, from 22 to 33 percent between 1995 and 2000. In contrast, the reported private sector share of employment in Poland was 60 percent as far back as 1996. 7 In Russia, the self employment share is approximately 6 percent; relative to more than 10 percent in Japan, the United Kingdom, Germany, the Czech Republic, and Hungary and more than 20 percent in Poland, Italy, and Republic of Korea (in the 1990s). EXECUTIVE SUMMARY XVII Table 3. The Socioeconomic Composition of the Unemployed, 1999 (Percentage) Average < 40 years Education * Previous Longterm age (percent) work his ** (per Basic General Prof. Higher tory (per cent) second seondary cent) ary Total 35.3 64.9 16.9 31.3 38.8 13.0 81.1 47.3 Men 35.5 65.7 20.3 33.4 35.1 11.3 82.8 44.0 Women 35.2 64.1 13.2 29.1 42.9 14.8 79.3 51.0 * Complete and incomplete. ** Period of job search more than 12 months. Source: Goskomstat (1999d). small and medium enterprises including problems with taxes, policy instabil ity, corruption, inflation, and the judiciary. For this reason, private and foreign direct investment is low. The lack of restructuring and continued domination of large enterprises also stymies the ability of small and medium enterprises to emerge, and dampens employment creation. As noted above, manufacturing job creation rates, while they have increased over the 1990s, still remain below OECD and high income transition countries. · Labor productivity is low. Although employment fell sharply, output declined even more, leading to cumulative labor productivity losses of approximately 30 percent in the 1990s much higher rates than those in lead CEE countries. Most of the employment adjustment came in reduction in the work force. Most evi dence indicates that adjustment in hours or secondary employment did occur, but were not as important as adjustments in primary employment suggest.8 Low productivity stems in part from continued over manning (relative to level of output), but also as a result of limited investment in capital noted above; and obsolete skills/experience of some parts of the workforce. · Unemployment is exacerbated by a skills and regional mismatch; and high payroll taxes. Despite recent declines, Russia can no longer be called a low unemploy ment economy. The rate of unemployment and its duration of unemployment are high (relative to OECD and some transition countries), and regional varia tions in unemployment are quite large (relative to transition countries). What factors might constrain the match of demand for workers and supply of unem ployed? (a) A Skills Mismatch. Workers with low levels of education, obsolete skills, and older age have the highest rates of unemployment and the longest 8 Further work is required to ascertain the exact nature of the informal sector in Russia. This report indicates that self employment in subsistence agriculture and infor mal wage payments may constitute important parts of the informal economy. XVIII THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table 4. Incidence of Fringe Benefits by Firm Size, 2000 Fringe benefits Total Firm size <25 26 100 101 500 >500 Paid annual vacation 0,915 0,755 0,944 0,961 0,974 Paid sick leave 0,912 0,742 0,939 0,964 0,980 Health services 0,374 0,195 0,284 0,414 0,622 Vacation subsidies 0,438 0,190 0,375 0,506 0,680 Kindergartens 0,130 0,064 0,097 0,149 0,270 Catering 0,152 0,096 0,133 0,160 0,221 Transportation 0,142 0,083 0,094 0,215 0,194 Training 0,213 0,097 0,194 0,256 0,323 Loans 0,143 0,081 0,111 0,159 0,249 Note: The total sample size ranges from 3746 to 4102 respondents. Source: Calculations from 2000 RLMS. duration of unemployment; (b) A Regional Mismatch. There continues to be a large regional variation in unemployment levels (higher than in Poland and the Slovak Republic, for example). High unemployment regions, concentrated in eastern and western Siberia and the North Caucasus have lower expenditure per capita, high poverty rates, high birth rates, and a high industrial share of output. High unemployment rates in high industrial share regions or particular state sectors (e.g. railways) indicate that unemployment in these regions, and in mono company towns, might be exacerbated (in the short run) by economic restructuring, which will require social policy focus. The evidence on the extent of regional mobility that would act to reduce some regional unemployment dif ferences is mixed. There is some evidence of informal mobility, but other studies suggest that the lack of affordable housing limits worker flows. Finally, it should be noted that high payroll taxes in Russia (higher than OECD, but lower than most CEE countries) may also contribute to higher unemployment than other wise, by raising the cost of labor (as suggested by international evidence). · The incidence of wage arrears has declined but the average amount of wage arrears for those who continue to face them changed relatively little (between 1998 and 2000). Wage arrears remain persistent for particular individuals (less educated, with longer job tenure), regions (rural), occupations (the mili tary), and sectors (agriculture). It is important to note that wage arrears are not caused by contract renegotiations (or wage flexibility) but by contract viola tions. Wage arrears tilt the earnings tenure profile, which together with the lack of contract enforcement, the market power of many employers, and lim ited mobility all serve to moderate workers' quit behavior and to increase the incentives of firms to use wage arrears. EXECUTIVE SUMMARY XIX · Wage remuneration in the form of fringe benefits outside market norms and under reporting of wages still are prevalent, particularly for employees of large firms. It is therefore not surprising that, unlike CEE countries, market forces (such as education) are less important factors explaining differences in wages. Rather, non economic factors or regional differences, or both, are probably the main reasons for wage differences in Russia. There also is con siderable underreporting of wages (which make wage measurements diffi cult), perhaps to avoid high payroll taxes. Recent studies have found that more than one third of private sector employees earn more than their regis tered wage and, in 10 percent of the cases, actual payments are at least six times the official level. · Poverty among the labor force increased over the transition (both according to Goskomstat and RLMS data) and remains high despite recent declines. High rates of poverty reflect the still low level of wages and other income (self employment) in Russia and the very high level of wage and income inequality.9 The highest poverty rates among the labor force are among the unemployed and workers with wage arrears. · Measurement matters. Ensuring that labor market outcomes are measured accurately is essential for better understanding of labor market developments. Counting subsistence Figure 4. Poverty Rates in Russia, 2000, by Socioeconomic Group 70 60 50 poor 40 of 30 20 Percent 10 0 time others part for pensioners pensioners pensioners enterprises recipients at age Unemployment Old Invalid Working benefit Survivor Working Workers Source: Goskomstat (see Chapter II). 9 Wage inequality has not contributed significantly to income inequality because of the low wage share in income. It is the high share of self employment income in total income that explains high income inequality in Russia (World Bank 2001a). XX THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY agriculture workers as employed10 which is not done in the RLFS would increase the employment rate (by 12 percent) and reduce unemployment rates (by 2 percent). The composition of employment also would change because the share of subsistence employment in the total labor force would increase. III. Labor Market Regulation xiv. Labor market regulation is restrictive in law but not in practice. During the transition, labor market regulation in Russia was unrealistically strong and inappropriate for a market economy. Moreover, in practice, for many firms and workers, it was completely bypassed, so that the labor market was virtually unregulated. Recently, after considerable public discussion and debate, the new Labor Code has been adopted. Given the diversity of views about labor law reform, passage of a new Code is a significant political accomplishment. The new Labor Code provides some improvements but more needs to be done, including providing more freedom to employers in deploying their work force. xv. A strict labor code without enforcement leads to violations of labor rights and reduces the welfare of workers below acceptable levels, and impedes labor productivity. A strict code with full enforcement will improve worker welfare but impose high costs on employers and restrict the abilityofthelabormarkettoadjusttoeconomicrealities,alsolimitingeconomicgrowth.Thechal lenge for Russia is to move from a labor regulatory framework that is restrictive and not enforced, to one that is flexible and fully enforced. This solution will both improve labor productivity and worker welfare. The key areas where further reforms are needed are the following (Chapter III presents a detailed discussion of this topic): · Excessive restrictions on flexible forms of contracting. The legal framework in Russia has been geared heavily toward formal, permanent, open ended contracts. There are numer ous restrictions on the use of fixed term contracts. These restrictions induce employers to engage in contracting practices that are in violation of the labor law. According to OECD data,excessiverestrictionsalsocanhurtvulnerablegroups,suchaswomenandyouth.The newLaborCodetakessomepromisingstepstointroducemoreflexiblecontractingprac tices; however, the Code does not make any marked improvements in either the deploy mentoflabororintermsofreducingtheexcessprotectionsofcertaincategoriesofwork ers, including women.11 Future reforms will be necessary to provide employers with the similar scope to deploy workers that their western counterparts have. · Substantial statutory employer obligations toward permanent employees, This appears to have improved with the new Code. Where obligations are large, the international experience indicates that the result is more informalization, 10 This assumes that subsistence agricultural workers were classified as non par ticipants. 11 For example, the Code heavily protects women in case of contract termination, overtime work, business trips, etc., raising the costs of firing female workers, but also making them costly to hire. Maleva et al. (2001) also find that the new Labor Code imposes considerable costs on employers. EXECUTIVE SUMMARY XXI Table 5. Official Minimum Wage and Average Monthly Wage, 1995 2000 Years Official minimum monthly wage (Rbl.) Percent of average monthly wage due 1995 42.6 9.0 1996 72.7 9.2 1997 83.5 8.8 1998 83.5 7.6 1999 83.5 5.3 2000 (August) 132.0 5.7 2001 (Quart. 1 3) 300.0 9.7 Source: Russian Economic Trends, October 2000 (tables 5 and 6) (see Chapter III). lower job creation, and potential reductions in the productive efficiency of enterprises. In some OECD countries, employers often overcome high protec tion accorded permanent employees in the labor law through the use of fixed term and temporary contracts. In Russia, as discussed above, these options have been restricted. As a result, employers have resorted to wage arrears, administrative leave, voluntary quits rather than layoffs, and contracting in the informal sector. These practices, particularly voluntary quits and wage arrears, also reflect the reputational risk of employers in laying off workers. The new law does appear to moderate these excessive termination conditions; how ever, it still imposes costs on employers wishing to adjust their workforces to economic and technological realities. · Ineffective and disortionary wage regulation. For example, Russia has a very low minimum wage that is not currently binding in any sense. Moreover, employers continue to use the tariff scale as a wage setting guidepost, despite its deregula tion. Wage practices such as wage coefficients for hiring northern workers, nonreporting of wages, and the compressed public wage scale continue to make wages an ineffective tool for allocating labor and measuring labor pro ductivity in Russia. The new Labor Code largely continues existing wage regula tions. What is new and worrisome is that the Code now stipulates that the min imum wage for the whole territory of the Russian Federation cannot be lower than the subsistence minimum defined for a working age individual (which may cause both fiscal and incentive problems) and states that wages should be indexed according to a consumer price index. If enforced these changes could be very costly from an efficiency and fiscal perspective; but if not, they would once again create a discrepancy between the law and actual practice. · Ineffective industrial relations. Russia has made some progress in making the transition in industrial relations from a regime designed for the planned XXII THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY economy to one appropriate to a market economy. Russia still has a long way to go, however, particularly in terms of developing the institutions that underpin effective industrial relations. There is an intricate bargaining appa ratus in the law, but there is actually little real collective negotiations deter mining wages and working conditions at the workplace. Unions or the bar gaining structure do not adequately reflect the voices of workers. International research has demonstrated that worker voice, embodied in the true representation of workers and employers in the bargaining process, can improve training and health and safety in the workplace, thereby contribut ing to productivity gains and improvements in worker welfare. In the new Code, provisions remain for collective bargaining at all levels. The Code does change procedures for determining bargaining representatives for employ ees. These new rules specifically pertain to what is considered a "local union" as well as to how a bargaining representative is selected when multiple trade unions exist. While it is still unclear how these rules will function, they may have the effect of limiting the opportunity for small and independent unions to represent workers. · The failure of enforcement and dispute resolution. The virtual absence of these institutions pose major challenges for Russian policymakers, employers, and labor. The consequences of the weak institutional framework for industrial jus tice are exacerbated in a slack labor market, and while disputes and contract violations dissipate when economic activity increases labor demand, workers remain vulnerable to the reemergence of such disputes in times of economic slack. The new Code does not appear to make major changes in this area. A pos itive aspect of the approach is that most conflicts are intended to be resolved at the enterprise level, which should minimize costs and time requirements. How ever, it appears to create a cumbersome practice of reconciliation of differences at the enterprise level. The timetable for hearing and resolution of labor dis putes is very tight. While labor inspectors and inspectorates have significant privileges and rights to monitor the execution of labor legislation, their role as mediators, conciliators and arbitrators of labor disputes is diminished if non existent. · What is the impact of labor market institutions on the labor market in Russia? The weak enforcement of the restrictive labor law has almost certainly allowed more adjustment in wages and employment than if this law had been enforced. Reputational risks of employers and other poor incentives to managers for lay ing off workers more than the labor laws may primarily be at work in reduc ing the pace of layoffs in the past decade. The absence of labor regulation, how ever, has had important tradeoffs in the form of lower worker welfare and worker productivity, including, among other things, low wages, wage arrears, and other contract violations among particular workers and in some regions (perhaps where workers have less bargaining power and job opportunities), poor health and safety standards, and large wage inequality. EXECUTIVE SUMMARY XXIII IV. Safety Nets for Workers xvi. Early in the transition, the government introduced a modern public safety net for workers, including active labor market programs and passive support in the form of unemployment benefits. The safety net has not been effective, however, in protecting workers against loss of income or skills.12 In 1998, the poverty rate of unemployed with benefits was much higher than the average poverty rate in the country. (Chapter IV presents a detailed discussion on this topic) · · The unemployment benefit program is quite generous according to law. The formal target replacement rate of benefits is quite high (75 percent for first three months; 60 percent for the next four months, and so on); and the dura tion of benefits is quite long (12 months; with reentry guarantees) relative to CEE norms. The main eligible groups are laid off workers and voluntary quits.13 However, more so than other countries, many other workers are also eligible, but for a significantly lower benefit. These workers include, for example, indi viduals who have never worked, have been fired for disciplinary reasons, or who have reentered the work force. Some groups receive special (higher) ben efits, for example, Northern workers. · The program is not generous according to practice, however. The coverage of the program is very low. Only 14 percent of the surveyed unemployed were regis tered with employment offices in 2000, out of which about 80 percent receive benefit. This is a much lower coverage rate than found for CEE or OECD coun tries,. The main reason for low coverage is the low and uncertain level of bene fits. Effective, or actual, benefit replacement rates (25 percent of average wage) are similar to those found in CEE countries, but are much lower than rates spec ified by law. Unlike CEE countries, the replacement rate is subject to consider able uncertainty (a result of benefit arrears which also was evident in 2000), and is therefore even lower. The benefit structure is much compressed with about 50 percent of the beneficiaries receive the minimum benefit. · Financing issues. The main reason for low and uncertain benefits is inadequate financing of the program. Funding was 0.16 percent of GDP in 2000 much lower than financing norms for advanced CEE countries (0.68 percent of GDP14) though not inconsistent with Russia's lower level of GDP. However, 12 Recent research (Lokshin and Ravallion 2000; Richter 2000) indicates that the safety net did reduce poverty more than if it did not exist; although more generous financing would have had greater impact. These studies, however, do not review the marginal impact of the unemployment benefit on poverty. 13 The distinction between layoffs and quits is blurred in Russia. Employers some times lay off workers by inducing them to quit for a number of reasons: to avoid pay ment of past wages, satisfy local authorities (who want to see lower layoffs), or to reduce the ability of workers to claim social services from the firm (available to laid off workers, but not to voluntary quits). 14 Data from 1997 99 (most recent years available) for EU accession countries (the Czech Republic, Estonia, Hungary, Poland, the Slovak Republic, Slovenia). XXIV THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table 6. Employment Fund Budget (Percent of GDP) and Arrears (Millions of Rubles) 1994 1995 1996 1997 1998 1999 2000 Incomes 0.51 0.43 0.35 0.36 0.33 0.32 0.31 Expenditures 0.30 0.37 0.34 0.35 0.32 0.29 0.29 Surplus (incomes over 0.21 0.06 0.01 0.01 0.01 0.02 0.02 expenditures) Arrears 1542.5 2843.3 3661.3 1618.6 (as of 01.01 of each year) Source: MLSD (see Chapter IV). funding is uncertain and not sufficient to cover program costs, as evidenced in the accumulation of arrears. Low share of resources spent on unemployment protection programs is not necessarily the result of limited public resources in Russia, but of their misallocation to non targeted programs (such as fringe ben efits, housing allowances or spa related benefits financed under the Social Insurance Fund). Administering an unemployment program is particularly dif ficult when financing is not commensurate to obligations. The informal sector, particularly under reporting of wages complicates the calculation of benefits which are linked to past wages. · The net impact of active labor market programs (ALMPs) in Russia is not well known. There has been no rigorous evaluation of ALMPs in Russia. Existing administrative data on ALMPs raises some concerns. There are four main areas of concern. (1) Russia spends more on programs, such as training and job cre ation, which are generally considered by international experts as cost ineffec tive and spends less on cost effective job counseling and information pro grams. (2) The focus of ALMPs is on younger workers rather than older, experienced, and less educated workers who comprise the majority of the long term unemployed. This is a mixed blessing. The success rate of ALMPs with younger workers may be higher, making programs more cost effective, but the program is not targeting older workers, who have the most difficult time get ting jobs. (3) It is difficult to evaluate program impact based on administrative data. For example, job placement rates of training programs in Russia based on administrative data are quite high, but training recipients are usually those who already have received a guarantee letter on employment from the employer prior to enlisting as trainees, 'clouding' this statistic. (4) There also is worrisome evidence that employment offices are under pressure by local governments to maintain and create jobs and prevent restructuring and layoffs. On a positive note, however, survey evidence (though not based on rigorous evaluations) from the restructuring of the coal sector suggests that the active and passive programming and employment services may be effective in allaying the social and political cost of restructuring in strategic sectors. The Government places EXECUTIVE SUMMARY XXV strong emphasis on the performance evaluation of the Russian ALMPs. The above work is crucial to understand which programs are the most efficient in Russia. Unlike the prevailing idea, the international best practices show that ALMPs have rather modest influence upon the decrease in long term unem ployment. The greatest impact is achieved in the case if the programs are ori ented to the specific groups, but even in this case they are very costly. The most cost effective programs are those of job counseiling and job information. Nev ertheless, ALMPs can play positive role while supporting the restructuring (see below). V. Policy Options xvii. Sustained economic growth that involves increases in employment and labor productivity will be the key for improving the living standard of Russian workers. Achieving sustained growth will involve completing the restructuring process, pro moting private sector development, and investing in education. Creating efficient labor market institutions and an effective safety net also will be important for achiev ing this objective. A. Creating an Enabling Environment for Growth xviii. Growth in labor productivity will require the creation of a strong private sector. Sustaining current economic growth and closing the gap in labor productivity between Russia and CEE and OECD countries will require stronger private sector led growth (and ensuing investment in modern technologies and physical capital). Poli cies to enhance economic growth are extensive and are discussed elsewhere in the Government's reform program and other Bank reports. These policies include achiev ing greater product competitiveness, developing property rights, strengthening financial markets, reducing administrative barriers to the growth of small and medium enterprises, lowering payroll tax rates (while at the same time reducing social expenditures in a consistent fashion), and creating the rule of law. The govern ment reform program is intended to address many of these constraints. The imple mentation of these reforms is essential for ensuring that scarce labor and capital inputs are used by the economy in the most productive way and in the most produc tive sectors. xix. Growth in labor productivity also will require a highly qualified workforce. Con tinuing investment in education will be required to develop a skilled and well edu cated labor force. Education has a large and increasing payoff in Russia. Real wages are higher for more educated than less educated workers, and highly educated individu als have a lower rate and duration of unemployment than less educated workers. The fluctuating rate of return to vocational education indicates that its relevance to the labor market needs to be particularly addressed as part of the education sector reform. A labor force with skills that can adapt to a rapidly changing market for labor XXVI THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY will be critical for Russia as it enters the global marketplace. The reform of the educa tion system to create a highly qualified and adaptable labor force has been given pri ority in the new Government program. The priority areas for reform in this sector are outside the scope of this study, but have been discussed elsewhere in Bank reports. xx. There will be winners and losers. Growth may be more beneficial for some workers than others. Real wages for younger, highly educated workers in the private sector are likely to increase. Older, less educated workers, with obsolete work experi ence, however, may not realize significant wage gains. Workers in regions with high unemployment rates that have high industrial shares of GDP, or in mono company towns, may be particularly vulnerable to the increased pace of restructuring. B. Creating Modern Labor Market Institutions xxi. Moving to market based regulatory practices means reducing excessive protec tions to workers offered by the legislative framework within the firm and, at the same time, beginning to strengthen the role of institutions in allowing workers a voice to ensure that basic rights are protected. These changes need to be complemented by a strong enforcement regime (dispute resolution, labor inspectorates). Social protec tion for workers, beyond the basic rights offered through labor legislation and more effective industrial relations, could be achieved through active and passive labor mar ket programming. Reform strategies in this area must therefore be made in concert with those in the social protection area. xxii. The debate over labor market reform in Russia is a contentious one, but may offer a false choice. The debate divides those who want to see more social protection from those who want to see more labor market flexibility. In a sense this is a false choice: By instituting a more realistic and enforceable, flexible, formal regulatory regime with a modernized safety net, the equity and efficiency concerns of both groups could be alleviated. Achieving these outcomes also will require the develop ment of a broad consensus regarding the need and direction of labor market reforms. xxiii. What should the priorities be? Considering the existing laws, institutions, and actual practices, and in light of international experience, priorities could include the following: · Reducing excessive rigidity in the Labor Code. The new Labor Code appears to make important progress in this area by removing the union veto on dis missals and implementing advance notice and effective appeals procedures. Some progress also has been made in providing for more flexible hiring arrangements, especially with respect to fixed term contracting. More could still be done. Increasing flexibility in hiring and dismissals should bring more employment "out of the shadows," and international experience tells us that it should most help vulnerable segments of the workforce (for example, women and youths). It is true that these amendments will reduce formal job security and, as noted above, it is important that they be coupled with improvements in the social protection system for workers (see Chapter IV). EXECUTIVE SUMMARY XXVII · Continuing to increase minimum wages. The current level plays little role in determining wage floors. Higher minimum wages (given the low base) are unlikely to have negative employment effects, and would reduce poverty among low wage workers. The level of minimum wage should not exceed a low share of average wage (for example, 25 to 30 percent) to ensure that work disincentives are prevented. However, the minimum wage will not be an effec tive policy instrument until the economy formalizes and enforcement improves. The linkage of the minimum wage to the subsistence minimum could lead to fiscal and incentive problems, particularly in low wage regions. Also, policymakers will need to consider how to accommodate the wide regional variations in labor markets and costs of living. · Reducing the influence of tariff in wage setting. The tariff has been uncoupled from non budgetary sector wages; but its continued relevance as a wage set ting guidepost is evidence of poor functioning of the labor market. As such it bears further investigation. The establishment of higher wages for particular areas, such as the North, is a legal requirement that is inconsistent with market practice and should be gradually phased out. · Developing institutions to allow worker voice, improve work conditions, enforce contracts, and resolve disputes, thereby raising worker productivity. Some options are (a) allowing true worker and employer representation in unions and eliminating management representation of workers, which would help improve work conditions; (b) considering decentralized bargaining approaches in collective bargaining, if the centralized approach is not yielding efficient bargaining outcomes; (c) increasing the resources available to the Federal Labor Inspectorate and building its capacity to provide technical assis tance and advisory services to enterprises; and (d) establishing alternative dis pute resolution mechanisms based on professional third party mediation, conciliation, and arbitration services outside the court system. C. Enhancing Public Protection through a Formal Safety Net xxiv. Ensuring that a public safety net exists to protect workers against income loss and job loss is particularly important in Russia, because it would protect workers against poverty, facilitate layoffs, and would help move protection out of firms and into the public domain. The Government has introduced general revenue financing of unem ployment benefit and ALMPs. The benefit design and ALMP strategy, however, have not been fully defined. Moreover, an outstanding restructuring agenda in one company towns and particular sectors (such as railways) will require an adequate safety net to reduce the social costs of layoffs. The report concludes that the following elements might be considered for the design of the safety net for workers in Russia. xxv. Move from firm based support to effective public safety nets. Considerable progress has been made in delinking the safety net from large enterprises. The remaining benefits provided by firms should be divested to municipalities, however, and municipalities should be adequately prepared to take over this responsibility. XXVIII THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY xxvi. Unemployment benefit design should be simple to administer, with incentives, and adequately financed. The report provides several policy options for unemploy ment benefit design. · The report provides three key benefit options: (i) a flat benefit, fixed in nomi nal terms as some percent of average wage, and indexed to prices is one option for policy makers to consider. A flat benefit minimizes administrative require ments, is progressively distributed, and is consistent with general revenue financing. (ii) The Government could also consider simplifying the benefit for mula to one that is some fixed percent of average wage over the entire dura tion of the benefit. (iii) If these options are not politically feasible, and the Government decides to retain the current formula, the report recommends the following changes in the eligibility and duration conditions of benefit. These changes should be considered whatever benefit formula option is cho sen by the Government: · Over the medium term, the level of benefit should be set so as to minimize work disincentives. The benefit level would remain a low share of average wage (e.g. 30 percent) to ensure work incentives. The minimum and maximum benefit levels should be delinked from minimum subsistence and established relative to the average / minimum wage. Over the medium term, the average wage will give more reliable information on the availability of fiscal resources and work disincentives for beneficiaries than the subsistence minimum. Given large regional differentiation in wages, differentiation of regional benefit levels will be important. · The assessment period for benefits should be increased, and benefits established at a fixed proportion of an individual's wages (for example, 30 percent of wages) in order to ease administrative requirements for processing benefit claims. · The duration of benefits could also be reduced to a maximum of six/nine months as in other CEE countries. A long duration of benefits, coupled with more generous level of unemployment benefits in the medium term, might induce longer unemployment spells. · Benefits could be provided to fewer categories of workers, such as laid off work ers and voluntary quits. Over time, as the distinction between voluntary quits and laid off workers is reduced, benefits for voluntary quits should be phased out or the eligibility of voluntary quits should considerably tightened in line with international practice. Special benefits to e.g. northern workers should be phased out as well. Targeting benefits would help save program expenditures, help the truly deserving, and reduce administration costs. xxvii. ALMP strategy. The future thrust of ALMPs in Russia is difficult to determine since programs have not yet been empirically evaluated using best practice evalua tion methods. Implementing such program evaluations should be expedited by the policy makers. On the basis of administrative data and international experience, how ever, the report indicates the following direction for ALMPs: EXECUTIVE SUMMARY XXIX · ALMPs are an important complement to passive programs, such as unemploy ment benefits. They have the potential to help individuals re enter the labor market, and reduce their dependence on public support. Therefore it is impor tant that a basic level of financing for employment services is guaranteed by the budget. · However, in countries where ALMP financing is limited, as in Russia, the focus of ALMPs should be on the most cost effective programs, such as job counsel ing and job information services should be increased. Emphasis on direct job creation programs should be reduced. Efforts to help the most disadvantaged workers (older, experienced workers, with obsolete skills) should increase. The use of employment quotas that state that individuals should have a job before being trained should be discontinued. · Empirical profiling of users, currently being considered for introduction, may be useful for assessing what programs work best for particular groups but the benefits and costs should be evaluated in Russia on a pilot basis prior to intro duction because it is an administratively complex program to implement. · The focus of employment services should be to help individuals find jobs themselves rather than helping preserve or create new jobs. Political pressure on employment agencies to contain unemployment is therefore misplaced. · Private provision could be introduced as the sector develops so that market information can be used to match workers to training programs. Private providers should be regulated, however, so that potential abuse is restricted. xxviii. Financing and administration. The report stresses that adequate financing of the program and its effective administration and monitoring are essential for its success. · The report cautions that the general revenue financing of passive and active pro grams, introduced in 2001, will not necessarily reduce arrears or regional inequity of benefit. The Child Allowance Program, which is now federally financed, contin ues to have these problems. Therefore, adequate and certain financing of the pro gram is required no matter the source of financing. At the same time, it is impor tant that the program is designed to take into account the Government's fiscal constraints and that it uses scarce budgetary resources effectively. It is also impor tant for the Government to provide a transparent allocation mechanism for trans ferring program resources to regions. Finally, the administration of both active and passive programs requires considerable attention to appropriate remuneration and training of staff, and their allocation across regions. · The report finds that an adequately financed safety net for workers is possible in Russia. The simulated cost of the benefit program with a 30 percent replacement rate (30 percent coverage, using 1999 data) would be approxi mately equal to 0.34 percent of GDP. Total costs of the program, including ALMP benefits, would be 0.44 percent of GDP well within the scope of Rus sia's level of income. (These costs would be well below the costs of similar pro grams in advanced CEE countries of 1.1 percent of GDP in 2000). The increase XXX THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY in benefits should be done gradually, as resources are released from improve ments in the targeting / phase out of other social protection programs (privi leges, housing allowances, for example). xxix. Social support restructuring. A combination of unemployment benefits, ALMP and severance benefits has proved important in downsizing the coal sector in Russia and also is widely used internationally to facilitate restructuring. It could there fore be used for downsizing in other sectors and regions in Russia (for example, regions with a high share of the industrial, overstaffed state sectors, or one company towns, or other over manned state sectors). The development of a strategy for iden tifying priority areas for restructuring and social programs for affected workers would be an important first step in this direction. The main elements of this strategy might include the following: (a) identification of the enterprises to be restructured, and the demographic and work skills of their personnel; (b) agreements on parameters of a social support package (determine its scope, costs, source of financing, and adminis tration drawing on existing mechanisms where possible); (c) stakeholder involve ment; (d) a public information campaign; and (e) monitoring and evaluation mech anisms. The latter could help to ensure that workers are not rehired via a "back door" and that workers who have difficulty in re entering the labor market are identified early and given targeted assistance to ensure they do not slip into poverty. If enter prises have social infrastructure (schools, clinics), divestiture of these assets might also be monitored to ensure that it has been successfully transferred to municipalities. D. Monitoring and Evaluation: Bringing Information Closer to Policymakers xxx. The report illustrates the importance of availability of data in order to monitor labor market developments and labor market programs to inform labor market policy. Three main sources of labor market information are important for monitoring the labor market: (1) the Labor Force Survey and Household Budget Surveys, (2) adminis trative data, and (3) enterprise based surveys. All sources are key to monitoring labor market developments. It is important that these surveys be strengthened and modern ized. It is important that policy units within the MLSD and METD of the Russian Federation are strengthened to use administrative and survey data to make basic fore casts of the impact of labor market policies. It is equally important that survey micro databases are available to the public so that labor market researchers, a strong and growing community in Russia, can help the Government evaluate the labor market sit uation and inform policies that would ultimately assist in improving the standard of liv ing of the population. Chapter I Understanding Employment: Level, Composition, and Flows This chapter initiates an analysis of the broad patterns of development in Russia's labor markets in transition through the year 2000.15 The demand for labor is derived from output. Therefore, the chapter begins with a brief overview of macro economic developments that have shaped labor market trends in Russia and pro vides a brief sketch of their impact on wages and employment. The chapter then focuses on understanding employment adjustment in Russia. It focuses on a few key questions: First, what explains the decline in employment? Despite recent increases in GDP, the increase in employment was very modest. The puzzle of aggregate employment decline is addressed through an analysis of unemployment, labor force participation, population, hours of work, and informal sector activities. Second, what is the structure of unemployment in Russia? Does it mirror the com position of unemployment in CEE countries? Third, to what extent have Russian labor markets restructured in the 1990s? Popular opinion suggests that labor mar kets have restructured very little in the past decade. This question is addressed by measuring changes in the composition of employment and by studying labor and job flows: labor market transitions and worker mobility across industries, occupa tions, and firms. Finally, what is the nature of the private sector in Russia? Under standing the private sector is important for assessing the prospects for sustained growth in employment and labor productivity in Russia. A complementary assess ment of labor market functioning, provided in the next chapter (Chapter II) focuses on incentives and returns in the labor market: the wage structure, non wage forms of compensation, and earnings inequality. These two chapters thus pro vide a complete overview of labor market developments in Russia during the past decade until 2000. In doing so, they portray the Russian labor market both at a time of crisis and recovery. 15 This is not the first attempt to provide a broad overview of Russian labor markets. Com mander, McHale, and Yemtsov (1995) provided such an overview using data through 1992 93, and other researchers have focused on a number of special topics (especially unemployment, wage arrears, and earnings differentials), but the behavior of nearly all key variables changed drastically in the mid and late 1990s. Clarke (1999) and Gimpelson and Lippoldt (2001) have provided more recent overviews, but their data cover little of the dramatic developments since the financial crisis of August 1998. 2 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY A. Setting the Stage: Macroeconomic Developments The demand for labor is derived from output. Therefore, labor market developments cannot be viewed outside of overall developments in product markets. Although the topic of this chapter, and indeed the report, is not macroeconomic and other sectoral policies, they are extremely important in understanding much of the labor market behavior. For this reason, this section provides a brief overview of the salient macro economic trends from a labor market perspective. At the end of the 1980s, which begins the period of our analysis, the relevant fea tures of the socialist system included an employment structure skewed toward large, industrial plants and associated engineering, technical, and skilled laborer occupa tions; essentially no self employment (as a main activity) and few small firms; a high labor force participation rate (LFPR), especially among women; little part time work or other flexible arrangements; and many one company and mono industrial towns, widely scattered geographically. The Soviet heritage also included misallocation on an enormous scale and in every economic dimension: across industries, occupations, firms, regions, and within each of these. The tendency to overbuild in the industrial sector, particularly in the military industrial complex, had resulted in excessively large plants producing goods with little civilian demand and engaging in labor hoarding in order to be able to meet plan tar gets in the presence of uncertainty about supplies. It also encompassed a compressed wage structure, but an important role played by fringe benefits and social services pro vided on a large scale by employers; and widespread informal activities, most notably in household plots.(for example, Granick 1987; Malle 1990; Oxenstierna 1989). Into this situation came a gradual decentralization of enterprise decisionmaking beginning in 1988 and an abrupt "big bang" liberalization of prices, entry, foreign trade, and competition on January 1, 1992. Privatization followed rapidly, first in the small firms of the trade and consumer services sectors, and through leasing of larger firms to their employees, and then through the voucher privatization of November 1992 -- June 1994 and subsequent sales of block of shares in companies. Macroeconomic instability and soft credit policy led to near hyperinflation in 1992 93 and periodic crises in the following years. The attempts to achieve macrostabilization between 1995 and 1998 in Russia led to the development of a serious nonpayment prob lem and the growth of a barter economy. At the macro level, the Government sought to stabilize the situation by tightening credit and fixing the exchange rate despite lagging fiscal reform. At the same time the Government did not want to harden budget con straints on enterprises because it would create mass layoffs and unemployment. The shortage of fiscal resources led the Government to borrow heavily. Interest rates increased sharply, causing liquidity problems for enterprises. This in turn con tributed to further nonpayments, but raised the need for further subsidies, tax arrears, cash shortfalls, and government borrowing. When public debt service reached unsupportable levels, the Government defaulted on its debt and caused a massive financial crisis in 1998 (Pinto et al. 2001) The subsequent drastic decline in the value of the ruble and the steep rise in the price of oil, Russia's chief export, were positive shocks to enterprise competitiveness and the state budget. However, they negatively UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 3 Figure I.1. Real GDP (1990 = 100), Russia and Select CEE Countries Source: Annex I Figure I.2. Real GDP, Employment, Real Wages, and Labor Productivity 11 0 10 0 1990=100 90 index: 80 70 employment 60 and 50 wages 40 real 30 GDP, 20 90 91 92 93 19 19 19 19 1994 1995 1996 1997 1998 1999 2000 Years Employment Real GDP Real wages Labor productivity Source: Annex I. 4 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY affected most Russians' real incomes, as well as wiping out any of their ruble denom inated assets, including both bank accounts and back wages owed by their employers. The shocks that have hit the Russian economy are similar to those that hit other CEE countries and the former Soviet Union. In Russia, however, the cumulative fall in real GDP between 1990 and 1998 -- approximately 40 percent -- was larger than that for any CEE country. Real GDP growth also turned around much later in Russia than in CEE countries.16 Significant aggregate growth appeared first in 1999 (a 3.2 percent rate), followed by a strong 7.7 percent growth in 2000 (Goskomstat 2001b) (see figure I.1) Growth has been fueled by rising energy prices, most notably oil, and a downward adjustment in the real exchange rate, and most recently by growth in domestic demand. (Figure AI.117 compares GDP growth rates for all transition countries, includ ing those of the Commonwealth of Independent States (CIS). However, most analysts believe that economic growth remains vulnerable to the collapse of energy prices. B. Aggregate Labor Market Trends Changes in real wages, labor productivity, and employment. The changes in output pre and post 1998 had major repercussions on the Russian labor market. How did the labor market adjust? The following sections look at the changes in aggre gate labor market trends in Russia. According to official statistics for the 1992 98 period, the labor market responded markedly to the decline in GDP (table AI.1) Aggregate employment declined by 12 percent, the unemployment rate nearly tripled, and nonparticipation in the labor force increased by a third. These numbers suggest that a significant share of the labor force left employment. However, the decline in employment was much smaller than the fall in output, and led to a sharp fall (20 percent) in labor productivity. There also were important real price effects. Real wages dropped by nearly 40 percent -- far more than the decline in labor productivity -- and the earnings distribution drasti cally widened, and poverty among workers increased. Labor contract violations in the form of wage arrears also spread to affect two thirds of all workers. Growth in the post 1998 period sharply reversed these labor market patterns. Between 1998 and 2000, GDP grew 11 percent in cumulative terms and employment increased by 2 percent (table AI.2). Thus, much of the increase in output was the result of an increase in labor productivity -- about 9 percent (cumulatively. Real wages fell between 1998 1999, but rose by 22 percent in cumulative terms in 1999 2000. Thus cumulative growth in real wages between 1998 2000 was actually nega tive, as wage growth lagged the growth in output. But real wage arrears declined by more than half over this period. However, wage arrears once again increased by 2.2 percent in May 2001, and remain high among public sector workers. (figure I.2). 16 According to figures provided by the European Bank for Reconstruction and Develop ment (EBRD) (2000). Also see Fischer and Sahay (2000). The EBRD figure of 45 percent cumu lative drop differs slightly from the fall of 42 percent implied by the annual figures from Goskomstat. 17 Table numbers prefixed with A refer to annex tables. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 5 Figure I. 3. Real Wages in Russia and Select CEE Countries Real Wages 1990=100 160 140 120 100 80 60 40 20 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Czech Republic Hungary Estonia Poland Slovakia Russia Source: Annex I. How did labor adjustment to output shocks differ in Russia from high growth CEE countries? Although measurement difficulties preclude precise comparisons, a num ber of observers (for example, Boeri and Terrell 2002) find that pre 1998, employ ment declines in Russia tended to be smaller (and wage and labor productivity declines larger) relative to output declines vis a vis CEE countries (figure I.1). Thus, price adjustment was greater in Russia than quantity (or employment) adjustment in response to output shocks relative to CEE countries. We will explore reasons for this difference in response below. How did aggregate labor market trends in Russia in the post 1998 period com pare with CEE countries? Comparing Russia and high income CEE countries in the first two years of economic growth shows that the average annual growth in employment and labor productivity in Russia was roughly comparable to that real ized by high income CEE countries in their first two years of growth (table AI.2). The wage response is similar to both Poland and Hungary, where wages lagged out put and employment growth. Nevertheless, the much larger decline in wages and labor productivity in Russia in the 1990s, noted above, means that it will require considerable growth in these indicators to close this gap with advanced CEE reformers. The paragraphs below evaluate these labor market trends in greater detail. Measurement issues. In evaluating Russia's labor market, issues of measurement and definition are crucial to bear in mind. Practically all official aggregates in Russia are subject to dispute, as a result of problems of measurement and interpretation. The magnitude of the output decline, for instance, is quite controversial in Russia, as else 6 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY where in transition economies.18 Perhaps more output is produced in the unofficial economy or at least in the more difficult to measure sectors, such as services and home production.19 The consensus appears to be that output has indeed fallen, although the official figures may overstate the magnitude. It also appears that the shock has been quite unevenly distributed across sectors, not only within industry, but also with respect to services. Employment measures may also be suspect, particularly if they are based on the traditional enterprise reporting system inherited from the central planning system, which will fail to take into account or underrepresent self employment, family busi nesses, start ups, and small firms more generally, as well as multiple job holding and a variety of other economic activities. Furthermore, officially reported wages may overstate actually received wages, because of wage arrears and forced in kind substi tutes, but they may also understate wages because firms have become adept at hiding salaries from the tax authorities. Specifically, wage data represent wages due rather than paid, as wage arrears are not taken into account. The real wage variable may not reflect worker welfare for several reasons: It is an average measure that may be unas sociated with any particular worker's welfare; it captures only part of total compen sation (cash payments); and it does not take into account the availability of consumer goods, a situation that was changing radically during the early transition years. There is consensus, however, that the general aggregate trends represented by this data are quite consistent and robust. Finally, it must always be borne in mind that Russia is a huge country, the largest in the world (in area), and extremely diverse. This fact renders generalizations quite dif ficult, and attempts to paint an overview of the Russian situation might frequently be wrong with respect to any particular region of Russia without an exploration of regional variation in the patterns of behavior. At the same time, the regional differ ences provide one source of statistical leverage for sorting out some competing hypotheses purporting to account for observed labor market patterns. C. Understanding Aggregate Employment Fluctuations The declines in employment in Russia are shown using four data series in figure I.4. The topmost series contains the official figures for total employment calculated from the Balance of Labor Resources and reported in yearbooks (Goskomstat 2000b), while the next highest series is derived from the RLFS. The discrepancy between the two series has been the subject of some discussion,20 but while magnitudes differ, 18 See, for example, Gacs, Holzmann, and Winckler (1995) and Fischer and Sahay (2000) for discussions. 19 See Johnson, Kaufmann, and Shleifer (1997) for a discussion of the unofficial sector. 20 See, for example, Clarke (1999) and Gimpelson and Lippoldt (2001. The former series represents the statistical agency's attempt to estimate employment on the basis of all available information, while the latter is a pure survey based measure. Clarke (1999) reweights the RLFS figures to account for alleged age related response bias, which results in a reduced discrepancy. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 7 both show a similar pattern of decline until 1998 and rise thereafter. The Balance series shows a decline of employment of 15.5 percent from 1990 to 1998 and a decline of 11.7 percent from 1992, smaller than the RLFS decline of 18.6 percent from 1992 to 1998. In the next two years, the RLFS shows a larger rise, at 8.6 percent, while the Balance series shows only 1.5 percent.21 The two series on the bottom of the figure pertain to employment in industry (manufacturing and mining) and are derived from reporting by large and medium size enterprises as well as a sample survey of small employers. The topmost of these includes all listed employees, while the bottom series covers only those employees involved in industrial production (excluding those in sales, provision of fringe bene fits, and so on). Again, although there has been some controversy over which series is preferable, they are quite similar, both declining until 1998 and showing a slight increase to 1999. There has been a disproportionate growth in industrial employ ment since 1998.22 Figure I.4. Alternative Measures of Employment 80,000 70,000 60,000 50,000 40,000 30,000 20,000 10,000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 Total employment from the balances of labor resources Total employment, RLFS estimates Total employment in industry Involved in industrial production only Source: Annex I. 21 A possible explanation for the difference stems from the fact that until 1999 the RLFS was carried out only in particular months of the year on a somewhat irregular schedule (for the com plete list of RLFS, see figure footnote). To maintain consistency in the time series and lacking any possibility for seasonal adjustment, figure I.4 reports the results for October 1998, August 1999, and August 2000 (the latest available). Quarterly employment figures from Goskomstat (2001b) show a 4.3 percent rise in employment from 1998:IV to 2000:IV. 22 The industry disaggregated RLFS employment series are discussed in section IV, below. RLFS figures include the industrial classification only since 1997. 8 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure I.5. Employment Trends in Russia and Select CEE Countries 105 R ussi a 100 Cz ech growth 95 R epubl ic 90 Hungary 85 P oland employment 80 of 75 S lovak R epubl ic Index 70 S lovenia 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 E s tonia Year Source: Annex I. As noted above, the decline in employment in Russia was more gradual than that realized in CEE countries, but was significant nonetheless (figure I.5). Taking the Bal ance of Resources employment rate as the metric of comparison, the share of employment to working age population, or employment rate, declined from 67 per cent in 1990 to 58 percent in 1998. The trend is quite similar using the RLFS employ ment figures (line graph in figure I.6), but the magnitudes of decline and recovery are slightly higher in the RLMS series. In both series, the employment rate recovered post 1998 (figure I.6). The labor force drop was largest in the oldest and youngest age groups. Although employment fell between 1990 and 1998, output fell more, and labor productivity declined (figure I.2). Why did employment fall relatively little (although still a great deal) with respect to output declines? The answer lies in the labor hoard ing behavior of firms, both under socialism and under capitalism.23 The reasons for labor hoarding during socialism have already been discussed above. Under capitalism, a key determinant of labor allocation was the mode of privatization followed in Rus sia. The asset transfer based on the vouchers and block sales model of privatizations went to insiders, such as managers, in more successful enterprises. The concentration of wealth and political power in the hands of these well placed business elites led them to block restructuring in order to avoid erosion of their privileges.24 State cap ture occurred on a far greater scale in Russia than in other CEE reformers, perhaps 23 Under socialism, see, for example, Kornai (1992) or Oxenstierna (1989). The classic study in a market economy is Oi (1960). For the Russian transition, see Clarke (1999), Com mander, McHale, and Yemtsov (1995), and Kapeliushnikov (1998). 24 Aslund (1999). UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 9 Figure I.6. Employment Rates in Russia and Select CEE Countries 80.00 70.00 60.00 population 50.00 (percent): age 40.00 rates 30.00 20.00 Employment 10.00 employment/working 0.00 92 93 19 19 1994 1995 6 199 97 98 19 19 1999 2000 vakia land venia HungarySlo Po Slo E s toniaAverage R epublic ech OEOECD Cz Years/Countries Source: Annex I. because Russia possessed fewer strong, market based institutions and a larger natural resources base (World Bank 2000a). Incentives to restructure were further reduced by the pressure imposed by local governments on employers to maintain employment and reduce the political, social, and economic consequences of "open unemployment" (McKinsey Global Institute report 1999). Local pressure is commonly applied through "discussion" with enter prise managers that encourages them to slow down the pace of layoffs, engage in job preservation programs in exchange for tax and social insurance contribution relief, or to phase in labor reduction programs (Pinto et al. 2001) These poor incentives to restructure were compounded by soft budget constraints, or ability of loss making enterprises to obtain noncash settlement for utility payments. All these factors worked together to dampen incentives for enterprises to lay off staff. The accumulated labor surplus is difficult to quantify. According to the REB (Russ ian Economic Barometer) survey data, the share of enterprises retaining surplus labor reached 60 percent in 1996 97, while the share of workers employed in nonproduc tive jobs was a bit higher -- 40 percent. Since then, a small survey by Tchetvernina et al. (2000) shows that the share of surplus labor in enterprises appears to have declined. This is not surprising, given the large decline in formal employment during the past decade. Why was employment growth so limited relative to output, post 1998? The aver age annual growth of employment in the first two years post recovery was very simi lar to that realized in CEE countries in the same period, where the growth in output also far outpaced increases in employment. The growth in labor productivity indi cates that output growth was a result of increases in employment, but also a result of 10 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY improvements in the allocation of labor. The smaller response of employment to out put growth is further evidence of overstaffing in enterprises (figure 1.2). What happened to previously employed workers? According to official fig ures from the RLFS -- which can be used to assess the changes in labor aggregates -- total employment in Russia fell from 71.1 million people in October 1992 to 57.9 mil lion people in October 1998, a drop of 13.1 million (Goskomstat 1999a).25 Where did these workers go? The three factors potentially accounting for the overall employment drop in Russia are growth of population, nonparticipation in the labor force, and unemployment. This section discusses the relative importance of each factor in turn. Population. It should be noted that fertility, mortality, emigration, and immigration are unlikely to be entirely independent from the economy, and indeed it is frequently alleged that the drastic rise in mortality is a direct consequence of economic hardship.26 Thus, the brief discussion here is pertinent not only to understanding employment changes -- from the supply side -- but also some of the social costs of transition. According to Goskomstat (2000b), the total population declined by about 2 mil lion from 1992 to 1999 (after rising since World War II). The age distribution shifted rightward (toward older age groups), and increased the share of the working age population during this period. The net effect of these changes was to increase the working age population (16 to 59 years old for men and 16 to 54 years old for women) by 1.6 million. Trends in mortality are less clear, and have fluctuated over time, but there is a more drastic and unambiguous decline in fertility rates.27 Thus, economic changes have affected mortality rates (although not in a clear fashion), but have had a more substantial impact on fertility rates in Russia. Have net migration flows helped to reduce population over and above natural increases noted above? Immigration and emigration issues are both among the pol icy concerns of Russian Government officials. Concerning the former, the problems include the inflow of Russians from other Republics of the former Soviet Union, the return of soldiers stationed abroad, and the growth of Chinese immigration in the far eastern portion of Russia. Concerning the latter, there are outflows to former Soviet Republics , temporary migration of unskilled workers to Europe, and "brain drain" to Israel, Western Europe, and the United States. In all cases, one must take even such official statistics as are available cautiously, as these are based on admin istrative registration, and they certainly omit many individuals (although the over all bias is difficult to assess). According to Goskomstat (2000a), in 2000, there was a small, positive net inflow from the former Soviet Union to all regions of Russia except for the northern and far 25 Note that these figures are based on the LFS, thus including employment in small firms, illegal or gray activities, and self employment, unlike official reports of enterprises. Also note that the RLFS questionnaire contains standard questions used on LFS throughout the world to define employment, although there are some ambiguities concerning home production, as dis cussed below. 26 See, for example, Field (2000), who also argues that the roots of the population crisis can be traced back to the 1960s. 27 Regional variation is again large, with a 19.5 percent unemployment rate in Dagestan and 6.6 percent in St. Petersburg. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 11 Figure I.7. Unemployment Rates in Russia and Select CEE Countries, 1993 2000 18 16 Czech Republic 14 Estonia rates 12 Hungary 10 Poland 8 Slovakia 6 Slovenia 4 Unemployment Russia 2 0 1993 1994 1995 1996 1997 1998 1999 2000 Years Source: Annex I. eastern regions. The latter regions show net outflows, mostly to Belarus and Ukraine. The only other (nonformer Soviet) countries from which migration flows are in Rus sia's favor are China and Cuba, but the reported flows are marginal. Clearly, these numbers are far below what one would infer from anecdotal reports, which suggests that better data collection will be necessary even to measure the approximate scale of the problem. Concerning flows between Russia and other foreign countries, the fig ures for other regions show net outflows, with the largest to Germany and Israel. Available statistics concerning the educational qualifications of emigrants suggest that many of those leaving the country tend to be the more educated and skilled workers, indicative of a brain drain (Heleniak 2000) However, this may be compen sated in part by a significant inflow of highly educated individuals from FSU countries to Russia. (Garsia Iser et al. 1998). In any case, the overall changes in flows to and from the country are not important in explaining changes in population. Although Russia has seen some dramatic demographic changes -- and the share of working age population has increased -- these changes are not significant enough to account for the substantial drop in employment. The lack of demographic impact on employment changes implies that the major factor accounting for decline in employment must be the increased share of people out of the labor force, or growth in nonparticipation. Unemployment. Unemployment growth was quite substantial in the 1990s, but it also does not fully account for the variation in employment. The six year unem ployment growth rate (October 1992 -- October 1998) of 4.9 million accounts for only 37.8 percent of the drop in employment (13.2 million) in the same period. Sim ilarly, the more recent rise in employment is only partially associated with a decline in unemployment. The answer to the puzzle of employment decline therefore must lie mainly in the trends in nonparticipation in the labor force. Nonparticipation in the labor force. The LFPR in the Russian economy emerging from the Soviet period tended to be high by world standards, although 12 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure I.8. Changes in Labor Force Participation Rates, 1993 2000 90 85 80 75 70 65 60 55 50 45 40 1993 1994 1995 1996 1997 1998 1999 2000 Russia, 15-72 age Russia, working age population United States European Union Source: Annex I. comparability is really possible only since the RLFS started in October 1992. At that time the overall LFPR was recorded at 70.3 percent (77.6 for males and 63.7 for females). The overall rate and the rates for both genders declined until October 1998, when the overall rate was 61.0 percent (68.1 for males and 54.7 for females). By 2000, the LFPR had fallen enormously to slightly lower than those prevailing in OECD countries (figure I.8 / table AI.10).28 In the 1990s, the LFPR declined for every group, but the decline was greatest for teenagers and those in their early twenties. In the prime age range of 25 50, there is relatively little change in the LFPR. The declines in the rates by gender were almost identical. Both rates recovered somewhat in November 1999, before slipping back slightly in August 2000. Retirement rates appear to have increased, as the drop in par ticipation at the official retirement ages became greater over the period. The pub lished RLFS figures for various age ranges reflect the retirement policy, however, as the big drop in the male LFPR occurs in the age ranges 55 59 and 60 72, while the female rate decline is in the ranges 50 54 and 55 59. The year 1999, however, shows evi dence of dramatic re entry by retirees, as the LFPR in the oldest age group of 60 72 rose from 9.1 (October 1998) to 14.6 (November 1999). The decline in the LFPR was associated with an increase in the out of labor force working age population (15 72) of nearly 11 million people from October 1992 to October 1998 (Goskomstat 1999a) (table AI.11). Although the next two 28 Comparison of LFPR across countries is not simple. Most countries define working age as 15, or 16 64 years. The United States uses a working age of 16+, while in Russia the definition of working age is 15 55 for women and 15 60 for men. In the comparison above, we used a def inition of 15 72 years for working age in computing LFPR for Russia. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 13 Box I.I Who Is Unemployed? Some Definitional Problems According to the International Labour Organization (ILO) definition of unemployment (Res olution I of the 13th International Conference of Labour Statisticians, Geneva, October 1982), the "unemployed" comprise all persons above a specified age who, over a specified reference period, are: 1. "without work," that is, are not in paid employment or self employment, "currently available for work," that is, are available for paid employment or self employment dur ing the reference period; and 2. "seeking work," that is, are taking specific steps in a specified recent period to seek paid employment or self employment. The specified steps may include registration at a pub lic or private employment exchange; application to employers; checking at work sites, farms, factory gates, market, or other assembly places; placing or answering newspaper advertisements; seeking assistance of friends or relatives; looking for land, building, machinery, or equipment to establish own enterprise; arranging for financial resources; applying for permits and licenses; and so on. The above definitions have severe limitations when applied to transition countries. First, many workers in developing countries who qualify as employed under the ILO definition are in fact not fully employed or are underemployed (especially in rural areas). These workers may work fewer hours than they would like or work in low productivity jobs and earn low wages. But they are so poor that they cannot afford to be without a job, and so open unemployment is rare. Edwards and Manning (2000) note that "the transition from underemployment to open employment can be partly explained as an income effect: As economies grow and household incomes rise, it becomes possible to go through periods without work while waiting for a job to open." Second, some unemployed may be classified as inactive. Individuals who have a marginal attachment to the labor force, that is, those who are available for and desire work, but are not actively seeking work because they perceive, rightly or wrongly, that no jobs are available, are often considered economically inactive when they should be more appropriately classified as unemployed (sometimes they are called discouraged workers). Moreover, the conventional application of the term "actively seeking work" also falters in light of a fair share of economic activity occurring through informal employment arrangements or where self employment is the norm. Third, some employed workers may be classified as inactive. According to ILO guidelines, an individual who works at least one hour in a week, or who is temporarily absent from work (for example, on vacation or because of illness) is in employment. Those who are out of work but do not meet the criteria of ILO unemployment are classified as economically inactive. However, some forms of informal economic activity may escape this definition of employment (for exam ple, home based work, typically undertaken by women). And because such workers are not available for work, they do not qualify as unemployed either. As a consequence, it is sometimes advisable to complement the unemployment rate with other measures of labor market slack (for example, with measures of underemployment). The ILO acknowledges the possible restrictiveness and "industrialized country" bias of the definition, advising the relaxation of these clauses and the formulation of criteria suitable to the labor mar ket characteristics of the particular developing country. For the purpose at hand, the above dis cussion implies, among other things, that besides those counted as unemployed, unemploy ment support programs may also include the underemployed -- and that the unemployed may not be the most unprivileged group in the labor market. Source: Vodopivec and Raju.(2001). 14 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY years saw the flow reverse by around 2.5 million people (Goskomstat 2000b), the puzzle still remains: Why does it appear that so many people have stopped work ing, dropped out of the labor force, and do not search for another place of work? Particularly given the low level of social security support, these enormous flows out of the labor force also raise the puzzle of how a large fraction of the population even survives. Definition of employment. Information on those engaged in home produc tion provided in the 1999 and 2000 RLFS sheds some light on these questions. In these two years, the survey measured those engaged in home production, divided between those with main jobs and those lacking other work. Three types of home production are included: agricultural production for sale (fully or partially), agricul tural production for own consumption, and industrial and service production for sale. On an average annual basis from November 1999 to August 2000 (four quarterly surveys),29 the total number of individuals engaged in home production for sales and own consumption is about 10 million, of which 8 million received no income from sales of their products. The corresponding figures for the February 1999 -- August 2000 surveys are shown in table AI.12.30 Thus, the question of employment decline in Russia rests on a definitional ques tion, namely. whether those in subsistence agriculture are included as employed. The answer appears to be that they are not. According to Goskomstat (1999a, p. 15), the definition of employment includes individuals engaged in home production only if they sell their products, but not if the production is for own consumption. The rise of home production, most significantly subsistence farming, then, provides a crucial part of the answer to the puzzling drop in employment in Russia. If subsistence agri cultural workers are counted as employed the unemployment rate would decline to 8 percent, and employment would increase by 12 percent. The employment rate would increase to 69 percent -- well above the average level found in most CEE and OECD countries. If this is the case, then employment declines are lower in Russia than in other transition countries, but that there has been a significant shift of employment from formal wage work to self employment in subsistence agriculture (table AI.13). Whether subsistence farmers should be counted in employment is certainly a judgment call, and the International Labour Office standard (ILO 1998) is not explicit 29 Clearly there is enormous seasonality in these activities. The May and August surveys find particularly high rates of participation in agricultural production for own consumption: 22,589,000 in May 1999, of which 11,075,000 of whom did not have any other job. These indi viduals are not exclusively rural; 4,727,000 of them are reported to live in urban areas. An addi tional 2,504,000 nonemployed individuals engaged in the activities where sales of agricultural products were concerned, and 155,000 where the sales included home produced manufactured items and services. 30 In the instructions for the RLFS (Goskomstat 1998b), question 66, which elicits this information, is supposed to exclude household plots of urban dwellers, but the RLFS question naire itself does not make this clear. Goskomstat (1999a) contains separate tables with infor mation on the activities of urban and rural dwellers (tables 2.60 and 2.61, respectively), so it appears the instruction was not followed. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 15 on this point.31 Examining the characteristics of the nonemployed engaged in agri cultural production reveals (see table AI.12) that many are older individuals -- partic ularly those who are engaged in subsistence agriculture as the sole activity, but more than half are in the normal working age range. In addition, a nontrivial number have completed higher education.32 The move to subsistence agriculture is not unique to Russia. In Romania, and also to some extent in Poland (where such workers are clas sified as employed), laid off workers also have shifted to subsistence agriculture as a means of coping with the decline in their income. A final point on the subsistence farmers concerns their hours of work. Goskom stat (2000c) and Table AI.13 report the number engaged in subsistence agriculture for own consumption for 30 or fewer hours in the reference week and those whose engagement was greater than 30 hours. For the year 1999, about 20 percent were engaged full time by this definition, while 80 percent were part time. An evaluation of the appropriateness of including such activities into the category of employment (or of assessing the degree of labor underutilization) may depend on the hours inten sity, particularly for work weeks shorter than 30 hours, but unfortunately there are no other available tabulations of the RLFS data. In summary, these results indicate that the fall in employment can be attributed mainly to a fall in labor force participation (of youth and older age groups) and somewhat to an increase in unemployment. The impact of demographics on employment was minimal. Most workers who left the labor force took up self employment in subsistence agriculture. Therefore, the drop in employment is a measurement issue: If self employed subsistence farmers were considered employed, the fall in employment in Russia would be less dra matic as compared with transition countries, but its composition would change toward one that was more comprised subsistence agricultural activities. (Table AI.14). D. Adjustment through Hours of Work The labor productivity analysis so far has not taken into account changes in hours of work. If hours of work have declined, then employment (totaling all workers) 31 The Russian definition of employment appears to be inconsistent with the ILO (1998, p. 93) statement that "[P]ersons engaged in the production of economic goods and services for own and household consumption should be considered as in self employment if such produc tion comprises an important contribution to the total consumption of the household." Else where, however, ILO (1998, p.3) states that "...in general, the data on economically active popu lation do not include...persons living entirely on their own means...", which seems to imply that subsistence farmers should not be counted in employment. Thus, the definitional ambiguity remains. 32 Unfortunately, no information is available on whether they are job searchers, and there fore on whether they are classified as unemployed or as nonparticipants in the reported labor force statistics. Counting them as employed would in either case raise employment and lower the measured unemployment rate -- by a large amount in the first case and by a smaller amount in the second. However, these individuals are probably not counted as unemployed, as they would have answered all the standard LFS questions addressed to the jobless, and any searchers would already be appropriately categorized, as would discouraged workers. 16 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY will overstate labor usage, and the decline in labor productivity will be overstated as well.33 An important role for hours adjustments would be consistent with a widespread view of Russian enterprise behavior whereby firms have responded to shocks by avoiding layoffs and hoarding labor, while engaging in work sharing through unpaid leaves and short time work and permitting employees to earn their livings through secondary activities outside the firm (for example, Aslund 1997). While the hours dimension in most countries is used only for temporary adjustments associated with the business cycle or periods of uncertainty, in Russia this situation has become a way of life, so goes the argument. Because of the pop ularity of this view and the complications in finding appropriate data for evaluat ing it, this subsection devotes detailed attention to a variety of types of evidence on the issue. To start with, table AI.15 reports results from the RLFS. Average hours worked on the respondent's main job -- defined either as "usual hours" or as "actual hours in the reference week" -- have fluctuated little from 1992 to 2000 (Goskomstat 1999a and 2000c), showing only the slightest of dips in 1994 95 relative to the other years. Unless the hours cuts took place prior to 1992, these data are inconsistent with a large role for this method of adjustment. Even in October 1998, there is no perceptible decline in the aggregate figures. "Actual" hours are always reported to be lower on average than "usual" hours, with a difference of 2.6 weekly hours in October 1992 but of only one weekly hour in November 1999, which may reflect some use of work hours as a temporary adjustment mechanism. It is notable, however, that most indi viduals reporting a discrepancy between actual and usual hours report a larger value for the former than the latter (Goskomstat 1999a); this is true for every RLFS except March 1996 when the two groups are in rough balance. Data by reason show that, among those working fewer hours than usual on aver age in the 1999 RLFS, 49.4 percent report involuntary reasons: 37.2 percent on "short hours by initiative of firm management," 10.0 percent "due to lack of orders," 1.9 per cent on unpaid leave, and 0.3 percent on paid leave.34 Yet, taken together, these invol untary part timers account for less than 1 percent of the total labor force. If the underemployed are added to the unemployed as an alternative measure of labor underutilization [Bureau of Labor Statistics (BLS) 2000], the figure would therefore differ little from the RLFS unemployment rate. Other reasons for working fewer than usual hours include illness (8 percent), vacation (8.2 percent), normal work regime (14.0 percent), seasonal work schedule (3.6 percent), with miscellaneous voluntary reasons accounting for the rest. Perhaps the rather constant length of the average work week is masking differen tial trends at a more disaggregated level. The RLFS figures in table AI.15 do show that, 33 Virtually all enterprise reported employment figures in Russia are defined in a way that partially accounts for hours of work (contractual days paid are counted for full time and invol untary part time employees, contractual hours paid are counted for employees on part time contracts, and actual hours are counted for workers on civil contracts), but the adjustment is far from creating a full time equivalent measure. 34 See Goskomstat 1999a, table 3.20. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 17 after falling initially, the dispersion of hours has increased somewhat, particularly at the top end, where the percentage of workers reporting actual work weeks longer than 40 hours fell from 14.5 percent in October 1992 to 2.0 percent in October 1995, then rose to 10.2 percent in October 1999. There was also a slight increase on the bot tom end, reflecting a rise in part time work. The overall rate of part time employ ment, however, is still very low by international standards, with only 5.6 percent of all employed reporting a usual workweek of 30 hours or less, and only 2.8 percent reporting 20 hours or less.35 One interpretation of the low part time rate, inherited from the Soviet period, is that employers have not become very flexible in terms of hours of work. Probably the old industrial sector, inherited from the socialist period, may have reduced hours downwards as the expanding new private sector has raised them, resulting in little change in the average figures. A first bit of evidence, also included in table AI.15, concerns work hours of the self employed versus those of employees. The reported hours of the self employed in Russia are not systematically higher than those of employees -- a difference from most other countries where the self employed typ ically work longer hours. Rather, self employment in Russia appears to be a primary vehicle for part time work, in the face of inflexible hours offered by employers. Sev enteen percent of the self employed reported usual work hours of 30 or less (as com pared with 4.8 percent for employees) and 9.4 percent reported 20 hours or less (as compared with 2.3 percent for employees) in November 1999.36 Table AI.16 addresses this issue by examining differences in average actual hours by industry, using available RLFS data for 1998 2000. The average work week varies rather little across industries, implying that hours adjustments are relatively unimpor tant across the entire economy. This finding again runs counter to the widespread claims of factories not functioning, workers leaving the job to engage in other activi ties, and so on. An analysis of the hours of work data in the RLMS is presented in table AI.17. As noted above, the RLMS questions pertain to the previous 30 day period, rather than to the reference week, as is standard for an RLFS. Nonetheless, the data are fairly con sistent across the two surveys, with the RLMS similarly showing high levels of hours worked (both on the primary job and on all jobs) and only modest fluctuations over the period. Disaggregated by industry, the RLMS hours data are again consistent with the RLFS in displaying little deviation from a full time work week. Somewhat higher estimates of hours adjustments appear in firm reports and sur veys. Table AI.18 shows the allocation of days worked and not worked in large and medium sized industrial firms, from 1980 to 1996, when the series ends. The number of "not worked days" rose nearly 8 days in 1992 over 1991, but this was associated 35 By contrast, 16 percent of workers in the United States were on part time schedules (defined as less than 35 hours per week, thus a broader definition of part time than in the Russ ian figures) in December 1999 (BLS 2000). 36 A further interesting difference in Russia compared with most other countries concerns gender differences in working time: men report only slightly longer workweeks (about 1 to 2 hours longer) than do women (Goskomstat 1999a). 18 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY with an increase in holidays, vacations, and "absences allowed by the administration" (for personal reasons); the category of work stoppages, resulting from reduced pro duction, rose by only 3 days. In 1993 96, work stoppages continued to increase, how ever, reaching 22.8 days by 1996. This data implies four and half weeks of leave on average for industrial workers, or around a 10 percent reduction in working time. Although much smaller than the 38 percent cumulative fall in industrial employment that was documented above, it is clear that hours of work did show some adjustment in the industrial sector. A well known method of hours adjustments in Russia has been the use of invol untary, unpaid leaves, which function similarly to temporary layoffs in the United States. Table AI.19 contains information from Goskomstat on involuntary leaves and short time employment, again collected from large and medium enterprises, for 1995 2000. The use of involuntary leaves peaked in 1996 in these data, when 15.8 percent of employees in the reporting enterprises had an average leave duration of 318 hours, or about eight weeks. Across all employees in these firms, the average was 50 hours, or about 2.5 percent of annual hours. The data on short time employment (reduced hours) are less complete, but show a similar peak in 1996 but involving only about three days per employee. Compared with the employment drops, the implied hours adjustment from both these methods, at about 9 days, is not very substantial but it is not trivial either. The RLMS also contains information on involuntary leaves reported by the respondents with respect to the previous year. The figures displayed in table AI.20 show a relatively low incidence of such leaves in 1995, unlike the administrative information in the previous table. Because the RLMS is a panel, it is possible to calcu late the persistence of involuntary leaves, measured as the conditional probability of such a leave in a particular year, conditional on having experienced a leave in one or more earlier years. The results, also in table AI.20, demonstrate that involuntary leaves tend to be concentrated in certain segments of the workforce. How the incidence of leaves varies with worker characteristics is discussed below. In summary, this review of the evidence suggests that changes in working time have not been the major method of labor adjustment in Russia. Thus, the fall in labor productivity based on employment measures does not appear to be overstated. Of course, the evidence could be wrong, but it comes from a wide variety of sources and uses a number of alternative measurement methods. The one exception to the over all picture may be in the industrial sector, where firm level evidence shows up to 10 percent of working days cut through work stoppages associated with reduced pro duction. Nonetheless, this figure pales beside the much larger drop in employment in the same period. E. Time Allocation Associated, although not exclusively, with the view that hours adjustments have been considerable in Russia is the contention that workers have increased their participa UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 19 tion in second job holding and other economic activities (outside of subsistence agriculture). According to the surveys conducted by the State University Higher School of Economics (2000), for example, an estimated 7.5 million individuals are employed only in the shadow economy (that is, they do not have any other job and potentially work in subsistence agriculture), with another 18 million having both for mal and shadow jobs. A significant portion of individuals who are officially classified as economically inactive, including students, pensioners, or housewives, or are for mally unemployed, are also shown to be engaged, permanently or temporarily, in the shadow economy activities. Informal work is reported to be especially prevalent in construction, trade, and the services sector. Together, these individuals are estimated to comprise 33 percent of the labor force. These additional activities are purported to be a major coping mechanism for households trying to make ends meet in the face of sharp, real wage declines, and create a mixing of the formal and informal sector activ ities in Russia. Is this result robust to hard evidence? Concerning the allocation of time out side of the main job, Goskomstat (1999a) reports RLFS figures for the number of multiple job holders only since March 1996. At that time, only 1.3 percent of the employed reported this status in the reference week; the number rose to 1.6 per cent in November 1999. These remarkably low figures are hard to reconcile with casual observation and anecdotal reports. Perhaps second jobs tend to be highly irregular in nature, but this would also tend to lessen their importance as an adjustment and survival mechanism. Even if the previous month is used as the ref erence period for counting the fraction of workers with second jobs (Goskomstat practice from 1999), however, the percentage of workers reporting multiple jobs was only 2.2 percent in November 1999, still a very low rate.37 Moreover, average hours of work on second jobs per week were only 14.7 in August 1999 and 11.8 in November 1999 -- the difference probably reflecting seasonality. These estimates of multiple job holding may be unbelievably low, but they would have to be off by an order of magnitude to start being really significant. Moreover, few available sur veys find high rates. The RLMS shows only about 7 percent of workers reporting second jobs. One way to calculate the difference that such secondary activities make to the cal culation of employment is to aggregate the hours of participation in all these types of secondary activities (including home production for own consumption) together with hours on the first and second jobs. Dividing by "usual" hours of work (39.3 in November 1999) yields an estimate of full time equivalent (FTE) employment in the economy. The result of this exercise is that Russia had an average 69.6 million FTEs in 1999, compared with measured employment of only 60.4 million (GKS 1999, table 3.25). These results provide an alternative approach, based on time rather than on numbers of individuals, to address the employment change question. It shows that 37 Goskomstat (1999a, table 2.44) appears to contain figures with inconsistent definitions, because as of 1999 respondents reporting second jobs in the previous month (not just the ref erence week) are included. 20 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY despite the popularity of the view that second job holding is common in Russia, the available evidence indicates that it is a relatively minor part of the picture.38 This section suggests that the employed did not significantly change their hours of work, engage in secondary activity, or significantly take up involuntary leave. Thus, the major quantity adjustment in labor was the result of the fall in the number of workers. F. Understanding Unemployment While unemployment growth does not explain a large part of the drop in employ ment, it is useful to understand the evolution of unemployment in Russia to appreci ate the workings of the labor market and to inform labor market policies. It is well known that in the first several years of reforms, measured unemployment in Russia remained low compared with most other transition economies. In 1993, for instance, the Russian rate (according to the RLFS) was 5.3 percent, compared with 16.4 percent in Poland, 12.1 in Hungary, and 10.4 in Romania. As noted above, the slow rate of restructuring was consistent with the gradual rise in unemployment. This observed difference led many observers to praise the "Russian way" of labor adjustment, in which the flexibility of real wages ameliorated the social costs associated with layoffs and unemployment.39 However, the flexibility of wages did not forestall significant unemployment growth in the late 1990s. The unemployment rate reached as high as 15 percent by February 1999 (figure I.7 and table AI.3). The inflows into unemployment40 were highest prior to 1995 and have gradually declined. The recent significant drop in unemployment rate (to 10 percent in August 2000, and to a further 8 percent in 2001) notwithstanding, it has been incorrect to characterize Russia as a "low unemployment economy" for some time. The administrative unemployment rate (defined by the number registering at local labor offices) is low -- in the 2 to 3 per cent range -- but this discrepancy is certainly the result of the low incentives to reg ister: low unemployment benefits, frequently paid late, and little by way of retrain ing and job placement support. Why did unemployment increase significantly over 38 Even if secondary employment is greater than indicated by these two surveys, it is unlikely that our results on declines in labor productivity would greatly change. If, as is com monly believed, hours worked are sticky because individuals may work full time for the firm, but sell in kind goods (received as wages) during regular work hours, workers can be thought of being sales agents of firms, selling goods and retaining 100 percent of sales revenue as wages. Alternately, it is often believed that wages are unaltered because workers use the facilities of firms in which they work to engage in secondary occupations during regular work hours. In this case, workers might be thought of as producing output for the firm (using firm equipment) but retaining 100 percent of profits as wages. While the output produced would not be accounted for in the second case, it would have to be quite large in order to dampen estimates of decline in labor productivity. 39 See, for example, Layard and Richter (1995) and OECD (1995). 40 Duration of unemployment one month or less. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 21 Table I.1. The Socioeconomic Composition of the Unemployed, 1999 (Percentage) Average < 40 years Education * Previous Long age (percent) work term** history (percent) (percent) Basic General Prof. sec Higher secondary ondary Total 35.3 64.9 16.9 31.3 38.8 13.0 81.1 47.3 Men 35.5 65.7 20.3 33.4 35.1 11.3 82.8 44.0 Women 35.2 64.1 13.2 29.1 42.9 14.8 79.3 51.0 Source: Goskomstat (1999d). * Complete and incomplete. ** Period of job search more than 12 months. this period? The sharp growth in unemployment reflects gradual and increasing restructuring, coupled with continued economic decline, or inability of the econ omy to create new jobs. Post 1998, as a consequence of economic growth, the unemployment rate has declined rapidly. The unemployment rate declined to 10.0 percent by November 2000 (and 9 percent in 2001, and increased slightly in 2002). In most high growth CEE countries, (except Hungary) unemployment rates actually increased in the first two years following growth -- because of rapid restructuring -- and declined there after. However, the decline was at most 2 percent over a two year period, lower than the 3 percent decline realized by Russia between 1998 and 2000. Moreover, after a period of decline, unemployment rates have either stabilized or increased yet again in most high growth CEE countries in recent years, as countries have renewed restruc turing (for example, Poland). In Russia, the sharp decline in unemployment shows that the labor market has been more flexible in being able to respond to economic growth than it has in other CEE countries. One reason may be the lack of enforce ment of restrictive legislation -- such as high minimum wages or restrictive termina tion conditions found in other CEE countries -- an issue that we will take up later in this report. Socio economic composition. There were large differences in unemployment rates across socio economic groups in 2000 (table AI.3). Unemployment rates were higher for younger, less educated, and less skilled workers. There was little gender difference. Women had a slightly lower unemployment rate than that of men.41 (Goskomstat 1999a, p.225). These patterns -- high rates of unemployed among youths and less educated individuals42 -- are largely similar to those found in most 41 The essential patterns of unemployment across age, gender, and schooling groups dis cussed here are also found in RLMS data, and in a regression framework where the other factors as well as regions are controlled. 42 The comparisons are with the 1996 data. 22 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure I.9. Percentage of Long Term Unemployed, Russia/CEE Slovenia Latvia Bulgaria Slovakia Estonia Hungary Russia Lithuania Poland OECD Total Czech R. 0 10 20 30 40 50 60 Long term unemployed (percent) Source: OECD CEET database, Goskomstat. CEE countries, although female unemployment rates tend to be higher than male unemployment rates in many CEE countries. As in most CEE countries, households headed by the unemployed have among the highest incidence of poverty. Which groups comprise the majority of the unemployed? Individuals with high rates of unemployment do not comprise a large share of the unemployed population. The majority of unemployed are about 40 years of age (65 percent of all unem ployed), have completed secondary education (69 percent), have previous work experience (81 percent), and are roughly evenly split between men and women (table I.1). The large share of unemployed with previous work experience and their increasing age suggests that the exit of workers from enterprises coupled with years of low aggregate demand were the main reasons for unemployment growth in Rus sia. The recent decline in unemployment has slightly reduced the share of experi enced workers among the unemployed.43 Duration. How long do the unemployed stay without a job? The average dura tion of unemployment (uncompleted spells) increased sharply, from 4 months in 1992 to 10 months in 1999, and fell to 9 months in August 2000 (table AI.3) In the early 1990s, when the unemployment rate was low, the unemployment pool turned over fairly rapidly, but as the rate grew, so did the proportion in long term unemploy ment. By early 1999, the proportion of the unemployed reporting a duration longer than 12 months (long term unemployed) was about half the total unemployment 43 The share of job losers has declined, and that of job quitters has increased. However, this finding is difficult to interpret because these two categories of workers are difficult to distinguish in Russia. Many employers induce workers to quit rather than laying them off explicitly. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 23 Table I.2. Regional Unemployment Rates: Standard Deviation and Max/Min Ratios 1993 1994 1995 1996 1997 1998 1999 2000 Standard deviation 1.93 2.27 5.66 4.84 7.69 6.55 6.37 5.16 Max 17.50 18.00 43.10 32.20 58.20 51.10 51.80 32.00 Min 3.30 5.50 5.40 5.50 3.40 4.70 5.60 3.21 Maximum/ 5.30 3.27 7.98 5.85 17.12 10.87 9.25 9.96 Minimum rate ratio Source: Goskomstat. pool.44 While Russia's long term share was initially at the very low end of the spec trum, by the late 1990s it had reached the upper range of long term shares found in OECD countries (figure I.9 and table AI.5). The share of long term unemployed is lower in Russia, however, than in many CEE countries (with the exception of Poland and Lithuania). Why might this be the case? Some factors that explain this difference may include the lack of enforcement of restrictive legislation, which allows individu als to find jobs faster than in CEE countries; the less generous unemployment benefit system in Russia; or a better match between the skills of the long term unemployed and those demanded in the labor market. It will require further research, however, to understand the precise factors that explain these cross country differences. Despite comparing favorably to CEE countries, the duration of unemployment has not been very responsive to economic growth. Why might this be the case? As in other countries, structural factors that impede the adjustment of the supply and demand of labor of different skills and across different regions may be important for understanding the long duration of unemployment in Russia. Skills mismatch. The groups with the longest duration of unemployment are older, less educated (52 percent),45 laid off workers (56 percent), with previous work experience (47 percent). Moreover, while unemployment rates have declined for all workers, the greatest decline has been among younger workers and those with some education. The unemployment rate of workers with basic education was the same in 2000 as in 1998 (it increased in 1999 but fell to 1998 levels by 2000). One reason for the growth in long term unemployment may be a skills mismatch, or the lack of mar ket demand, for older, less educated, and laid off unemployed workers (tables AI.6 I.9). skills mismatch also explains socio economic composition of unemployment in CEE countries. (World Bank 2001a, 2001b). 44 Goskomstat (2000c) is unclear about the boundary of the category "9 12 months" and "12+ months," but the qualitative picture is little affected. 45 Of all unemployed workers in this category, those that are long term unemployed. 24 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Regional disparities. Regional unemployment rates have converged over time. The standard deviation of regional unemployment rates has declined, from a peak of 7.7 in 1997 to 5.2 in 2000. The ratio of the minimum to maximum unemployment rate also has decreased by almost half in the same time period. This reflects a decline in unemployment in high unemployment regions and low unemployment regions. In particular, there has been a remarkable decline in unemployment in the highest unemployment region of Inghushetia, from 58 to 32 percent between 1999 and 2000. Despite declining over time, regional differences in unemployment rates remain striking. In 2000, when the average unemployment rate in the Russian Federation was 10.7 percent, unemployment rates ranged from a low 3.2 percent unemployment rate in Yevenkisski to a high of 32 percent in Ingushetia 10 fold difference.46 This ratio is higher than that realized in the Slovak Republic (8 percent) and in Poland (4 percent) Moreover, while rankings of regions with the 10 lowest unemployment rates have varied over time, high rates of unemployment have remained persistently concen trated in Eastern and Western Siberia and the North Caucasus regions (table I.2). What explains the differences in unemployment rates across regions? Regres sion analysis using 1999 data revealed that unemployment rates were higher in regions with higher official poverty rates and lower in regions in which average per capita expenditures were high (in richer regions) and where industrial produc tion accounted for a larger share of GDP.47 After controlling for these variables, unemployment rates tend to be higher in urban areas, in areas with high birth rates, and in areas that have welcomed refugees, although the latter effect is very slight (table AI.9a). How can we interpret these findings? Rural populations will be less likely to exhibit open unemployment as long as farming is a welfare augmenting alternative and jobs are scarce. While it is difficult to believe that heads of households with more children are a priori more likely to be unemployed, birth rates likely proxy other vari ables related to employment, such as differences in employment by ethnicity. In other words, the economic disruptions of transition may weigh more heavily on some nationalities than on others. The reasons for this would need to be investigated. There is one interesting policy dimension: Assuming restructuring will take place in regions with relatively high industrial output, then, all else being equal, these regions may face (even) higher unemployment rates in the future. Once these factors are accounted for, unexplained differences in employment still exist for some regions.48 In particular, the north, the North Caucasus, and the far east regions have higher than average unemployment rates. And these differences are considerably higher than differences in the unadjusted rates, suggesting that other correlates with lower unemployment, such as a larger agricultural population or greater industrial production, may be at work in explaining these differences. 46 It should be noted that Ingushetia had the highest regional poverty rate as well. 47 These latter two variables were lagged a year to reduce multicollinearity. 48 All comparisons are made with respect to the central region, which contains Moscow. The choice was arbitrary. The unemployment rate for the region at 10 percent is somewhat less than average. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 25 Regional mobility. Differences in regional unemployment rates may stem from low regional mobility. However, this stylized fact is difficult to establish. One reason is that much internal migration may be unofficial, thus not captured by the registration system. A second reason is that it is difficult to find an appropriate standard by which to assess whether observed rates are low. Russia's territory is huge and extremely diverse, with large heterogeneity, and peculiarities of industrial location and trans portation infrastructure. With these caveats in mind, official figures on migration, which are based on reg istration data,49 show annual intraregional and interregional arrival and departure rates of about 1 percent each (Goskomstat 2000a). Of course, these omit unofficial migration. Large migrations have occurred in post transition Russia. Half of the regions classified as the extreme north have lost more than a quarter of their popula tions during the post Soviet period. The two regions in the far northeast corner of Russia Magadan, across the Bering Sea from Alaska, and Chukotka have respec tively had 42 percent and 58 percent of their populations leave because of deterio rating economic and social conditions. This rapid depopulation of Russia's northern and far eastern periphery was a response to the dismantling of subsidies and the lib eralization of prices. However, by international standards, a very high share of Russian population still continues to live in the north. Northern Russia is 2.5 times as densely populated as Alaska and 50 times as densely populated as northern Canada and Greenland. Of the 11 cities in the northern regions of the world with populations of 200,000 or more, 10 are located in northern Russia (the 11th is Anchorage, Alaska). Barriers to mobility. One of the most commonly assumed stylized facts about Russian labor markets is that there are large barriers to internal geographic mobility. The barriers include continued use of the permit (propiska) system by Moscow and some other cities, where large bribes have to be paid to register as a resident (neces sary to find an official job); poor functioning of the housing market; and poor com munication and transportation links.50 The extent of these barriers and their links to migration are difficult to establish, and evidence is mixed. Recent work by Gerber (2000) shows that this mobility is a result of economic incentives. Regions with higher levels of privatization (of housing, in particular) and higher levels of per capita trade turnover (as a measure of economic activity) are attracting in migration. Rural to urban flows also appear to be related to economic incentives the access to private farms. Other studies stress the role of housing in constraining migration across and within regions. For example, Heleniak (2000) finds that the out migration of able bodied workers from the north has left a large share of vulnerable individuals, such as the elderly, disabled, and unemployed people without the money to migrate. For these and other individuals who remain in the north, the unavailability of affordable housing is the main reason for not migrating to other regions. A recent study of three 49 Anecdotal evidence suggests that there may be considerable informal migration as well. 50 Friebel and Guriev (2000) analyze the impact of employer provided fringe benefits. 26 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY oblasts (states) found evidence of long commutes to Moscow from surrounding regions. Most workers cited the unavailability or high cost of housing as the main fac tor constraining a move closer to work (Pinto et al. 2001) Housing constraints also have been found to be important in limiting labor mobil ity in CEE countries. In Hungary, long commute times for many workers are explained by the high cost of housing in the country. Differences in regional mobility rates within the United States and across OECD countries have also been shown to stem mainly from the availability of rentals. Countries with higher shares of owner occu pied housing tend to have lower mobility rates. (Jackman 1998) To the extent that Russia has a high share of owner occupied housing, it may account for limiting inter and intraregional mobility in the country. In summary, despite recent declines in the unemployment rate, Russia can no longer be called a low unemployment economy. The long duration and high regional variation of unemployment point to a skills mismatch and potential barriers to regional mobility, such as the availability of housing. G. Structural Change and Labor Mobility The analysis so far has focused on the broad aggregates, in an attempt to better under stand the drastic fall and partial recovery in employment and the puzzle of where workers have gone, how they spend their time, and how they survive. This section takes a more disaggregated look at labor reallocation. The big question to answer is the following: Is there genuine restructuring occurring in Russia? Many observers tend to answer negatively, but rather little systematic analysis has been undertaken. In par ticular, the controversies have focused on the firm level issue of the extent of restruc Table I.3. Labor Market Transitions, 1998 2000 1998 2000 Labor force status in 2000 Total 1998 Employment With job/ Unemploy Out of labor not at work ment force Employment 0.826 0.054 0.030 0.091 0.480 With job / not at 0.636 0.170 0.067 0.127 0.039 work Unemployment 0.482 0.028 0.185 0.305 0.060 Out of labor 0.127 0.007 0.045 0.821 0.421 force Total 2000 0.504 0.037 0.047 0.412 1.000 N=6510 Note: Each cell measures the probability of transition from labor force status i to labor force status j. Source: Calculations from RLMS. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 27 turing in response to privatization, but there has been relatively little analysis of changes in the composition of employment and of the flows of labor across sectors; nor has there been much analysis of the extent and determinants of labor mobility51. (The following chapter considers restructuring from the angle of changes in price sig nals in the labor market: the wage structure, earnings inequality, and other aspects of compensation). Labor market transitions. Table AI.21 starts the analysis of restructuring with gross flows of individuals between labor force states, using the definitions of table AI.3. The data are the 1994, 1996, and 1998 rounds of the RLMS, and two year transi tion matrices are displayed. The 1994 96 and 1996 98 matrices are fairly similar, except for a higher propensity for the employed to enter unemployment in the latter period. 1998 2000 is quite different, however, with a substantially higher rate of tran sition out of unemployment into employment and lower to nonparticipation (table 1.3). The probability of remaining employed increases, as does new and re entry into employment from out of the labor force. Evidently, economic growth in post crisis Russia was not only strong enough to increase the probability that workers would remain employed, but it also brought the unemployed back to employment more quickly than before, and it even pulled in labor force nonparticipants. Specifically, between 1998 and 2000, almost 50 percent of the unemployed became employed, and only 20 percent remained in that status. Labor mobility. Another indicator of labor market changes is labor mobility rates. How do these compare over time? Table AI.29 shows the size of gross worker flows across sectors, industries, firms, and occupations for several subperiods between 1985 and 1998. The data are from the RLMS, using both retrospective information on the 1998 survey and the panel from earlier years. Gross worker flows are measured as the fraction of employed respondents who changed sector, industry, firm, and occu pation, respectively, between the first year and last year of the considered period. As in the earlier analysis of movements between industrial sectors, these mobility measures also show an unambiguous increase in worker mobility after 1991, the year when reforms began. The number of people who moved to another industry, firm, or occupation was already considerably higher during the first four years of reforms (1991 1995) than during the preceding six years (1985 1991). During the seven years of transition, 1991 1998, 42 percent of employed respondents changed their indus try, firm, or occupation, nearly twice as great as the share of movers in the previous six pre reform years. Mobility also became more "complex," more frequently involving simultaneous changes in occupation, firm, and industry (Neal 1999). Table AI.29 also indicates that these labor flows were most intense during the first five years of reforms, and after 1996 the rate of labor mobility begins to fall (table I.4). This decline in mobility could be partially the result of the diminishing rate of structural change and the relative sta bilization of labor force composition. 51 Commander, McHale, and Yemtsov's (1995) study is based on data only through 1993, while Clarke (1999) and Gimpelson and Lippoldt (2001) have limited information on gross flows. 28 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.4. Job and Occupational Mobility (Self reports in the RLMS) Years Changed both firm Changed firm but Changed occupa Did not change and occupation not occupation tionbut not firm either firm or occupation 1996 1998 0.170 0.093 0.034 0.704 1998 2000 0.176 0.105 0.035 0.685 Note: The table shows the fractions of employed respondents who reported in 1998 and 2000 that they changed their place of work and occupation as compared with December 1996 and December 1998, respectively. Source: Calculations from RLMS. Comparable information on industrial, occupational, and interfirm mobility for the 2000 RLMS is not yet available, but table AI.30 contains the results from a variable measuring respondents' self reports on changes of occupation and employer in the previous two years. These data, available in the 1998 and 2000 surveys, indicate a small tendency toward increased mobility in the second two year period (1998 to 2000 compared with 1996 to 1998). Relatively little mobility is intrafirm occupa tional change (an indicator of internal organizational change), but much of the inter firm mobility also involves change of occupation. A final approach to measuring mobility in Russia relies on tenure information. Mean job tenure and tenure distributions, by age and gender, are shown for Russia and some comparator countries in table AI.31. The Russian calculations are based on the RLMS cross sections for the years 1994, 1996, 1998, and 2000. In 1994, Russian women had much longer tenure than did Russian men and, relative to the women in other countries, the Russian female job tenure distribution tended toward the high end of the spectrum; only Italy had longer average tenure in each age group, and only Italy and Japan had lower fractions of employment with tenure of one year or less. The female distribution evolved quite rapidly in the late 1990s, so that by 2000 it resembled that of men, and it implied quite high rates of mobility. The tenure distribu tion for men changed less in this period. Particularly striking are the patterns for the two older age groups, 26 45 and 46 60 years of age, in both of which mean job tenure for Russian men was well below the figures in all the other countries; in the oldest group, this was true for Russian women by 2000 as well. The data also are consistent with a pattern of labor force exit of older workers around 1998 and re entry before 2000, when the fraction with new jobs (tenure of a year or less) jumps substantially. These results show that the Russian labor mobility increased in post transition Russia, and its composition exhibited greater complexity. While labor flows declined post 1996, they increased once again although modestly post 1998 in response to economic growth. Job tenure data also confirm labor flexibility in Russia. Employee turnover. An important puzzle in Russian labor markets has con cerned what are reported to be relatively high rates of worker turnover. The rate of labor turnover both hiring and separations is higher in Russia than in most transi tion countries, and approximates the range found in lower income OECD countries UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 29 Table I.5. Hiring, Layoff, Quit, and Separation Rates, 1991 98 Total Year Hiring Layoffs Quits separations 1991 0.21 0.01 0.12 0.20 1992 0.20 0.02 0.15 0.26 1993 0.20 0.02 0.17 0.26 1994 0.18 0.03 0.19 0.28 1995 0.19 0.03 0.18 0.27 1996 0.17 0.04 0.18 0.27 1997 0.18 0.05 0.17 0.28 1998 0.18 0.04 0.18 0.26 Note: Employment separations resulting from death, entrance to army, and retirement are not counted as quits or layoffs. Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (2002). (Gimpelson and Lippoldt). Layard and Richter (1995), for instance, take this to imply substantial worker reallocation and therefore evidence of restructuring, while others cite a number of possible alternative explanations: measurement error in the Goskomstat statistics, churning (rehiring of former employees), employment changes associated with split ups and mergers, or mover stayer heterogeneity in the Russian labor force, such that the observed turnover is accounted for primarily by a rather small group of workers.52 At the moment, evidence is insufficient to be able to evaluate these alternatives, but table AI.27 documents the empirical regularities using official Goskomstat data on hirings and separations in large and medium size firms for the years 1993 2000. The rates are fairly constant over the period, except for an upturn in hiring (and to a smaller extent in separations) in 1999 and 2000, particularly in industry and con struction, but it is useful to consider some independent reporting in a firm survey. To check the possibility of measurement error and to provide separations disag gregated into layoffs and quits, table I.5 shows employee turnover rates for a sample of Russian manufacturing enterprises. Layoff rates are low throughout, but rising steadily during this period. Quit rates are more substantial, also rising. Most striking, however, are the high hiring rates, with an only slightly declining average value dur ing this period. Why Russian manufacturing firms should have engaged in so much hiring during a period of such drastic decline is a puzzle, certainly one that merits future research.53 52 Discussions with Vladimir Gimpelson and Rostislav Kapeliushnikov were very useful in laying out these alternatives. 53 See also Lippoldt and Grey (1997), Gimpelson and Lippoldt (1997), and Kapeliushnikov (1997). 30 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Sectoral shifts in employment. Does the evidence on labor mobility above contain any evidence of productive restructuring? Were the shifts in aggregate labor demand associated with restructuring of the composition of employment? A first question on changes within employment concerns the sectoral allocation of labor. Table AI.22 shows changes in the industrial composition of the labor force, as reported by Goskomstat (2000b), for the period 1970 99. Changes are small from 1970 85, but thereafter accelerate, with a pronounced shift out of industry, construc tion, transport, and communications into trade, finance, and public administration. (Table I.6 shows shifts between 1990 and 1999). The magnitude of these shifts is large even by comparison with CEE countries experiencing similar transition shocks in the early 1990s. Boeri and Terrell (2002), for instance, report OECD figures for 11 transition economies that show Russia roughly at the median level for reduction in the employment share of industry and growth in the share of services, in the latter case ahead of Bulgaria, Poland, Romania, and Slovenia. These changes are mirrored in the large decrease in the agricultural share of GDP and increase in services share of output. From the point of view of inter industry shifts, the evidence does not support the claim that restructuring has been particu larly sluggish in Russia. It is noteworthy, however, that the pattern of shifts slightly Table I.6. Composition of Employment by Industry, 1990 99 (Percentage) Change 1990 1999 1985 99 Industry 30.3 22.4 30.7 Agriculture/ 13.2 13.7 4.2 Forestry Construction 12 7.9 15.9 Transport/ 7.7 7.6 22.4 Communications Trade 7.8 14.6 75.9 Housing 4.3 5.3 29.3 Health services 5.6 7 40 Education, culture, art, 13.3 13 3.2 and science Finance, credit, and 0.5 1.2 140 insurance Public administration 2.4 4.5 136.8 Other industries 2.9 2.8 55.5 Total 100 100 Source: Goskomstat (2000b). UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 31 reversed itself in 1999, the final year shown in table A1.22, as the industry share increased for the first time in the entire period. This results suggest that the recent recovery may, to some extent, represent a reversal of restructuring in the sense of sec toral reallocation of employment.54 To some extent, this proposition is reinforced by an analysis of RLFS data, pre sented in table AI.23. Results with industry coding are available in the RLFS only since 1997, but data since that year show a slight rise in the proportion in industry (imply ing a significant rise in the absolute numbers employed in industry, given the overall employment increase). Trade and public administration also have increased, while construction, utilities, health, and education have declined. Results from similar computations using the 1998 RLMS in table AI.24 show a very sharp increase in employment in finance and commercial activities during the period 1985 2000. These data are based on retrospective questions concerning the respon dents' employment status in 1985 and 1991. The results demonstrate clearly that the shifts accelerated in the 1990s.55 Similar to the RLFS results, however, it is noteworthy that the changes since 1998 to some extent reverse earlier changes, particularly in the rise in certain industrial sec tors. The post crisis period did show strong growth in trade and finance, however. Moreover, because they concern the same individuals, it is also possible to analyze the nature of employment transitions, shown in table AI.25. Taking first the diagonal ele ments in the matrices, which show the rate at which individuals stay in the same sta tus (sector of employment, or nonemployment), it is remarkable how much higher the rates are for the 1985 91 period compared with the 1991 98 period. Even adjusting for the slightly greater length of the latter, the figures show much greater mobility after the transition really began about 1991. The probability of remaining in the same sector declined quite substantially (except for nonemployment, which reflects retirement). On net, the expanding sectors received workers from the declining sectors, but the flows to and from non employment are the largest. A pro nounced difference between the earlier and later periods concerns the transitions from nonemployment to the various employment sectors, where the rates of move ment into industry, agriculture, transportation, and construction fall drastically. This result implies that new entrants to employment tended to enter the service sectors. Size distribution of employees. Now, a fall in employment in a sector may be the result of either firm exit or firm size reductions (or both). The latter is of interest in light of one of the major distortions produced by central planning, a tendency to concentrate production in a relatively small number of large companies. Thus, an important type of evidence of market oriented changes in employment is downsiz ing of large firms. For a sample of firms comprising most large and medium enterprises in the indus trial sector, table AI.26 clearly shows the downsizing effect of transition. Average firm 54 Further useful evidence on this point could in principle be obtained from more recent information on a more disaggregated set of industries, but neither is available at present . 55 As these results are based on retrospective data, they may suffer from age related bias if age is correlated with sectoral choice. 32 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY employment was relatively constant in the late 1980s and early 1990s. After 1993, however, average firm size decreased rapidly, and by 1999 it was less than half what it was during the Soviet period. The fuel, machine building, light industry, and building materials sectors downsized the most proportionately. Thus, the data do suggest substantial reallocation of employment across sectors, although the preliminary evidence also implies that much of it was accomplished through transitions involving nonemployment, that is, through withdrawals from employment, and by employment of entrants.56 Nevertheless, the data also show quite substantial switching of sectors by workers. Changes in occupational structure. The restructuring process also has changed the structure and directions of occupational mobility. Table AI.28 again draws upon the retrospective questions in the 1998 RLMS to show that the occupa tional composition has shifted toward more market oriented and service providing activities from 1985 to 1998. The share of managers, entrepreneurs, specialists in business and law, customer service clerks, salespersons, and other service providing workers increased. At the same time the recent occupational changes are character ized by a strong decline in a number of engineers and skilled laborers that may reflect a shift of employment from goods producing industries to service providing indus tries. The RLMS also indicates that the share of army specialists dropped by 13.3 per cent, which may have certain consequences for the labor market (retraining and high unemployment among former military specialists).57 Job creation and destruction. Thus far, the results show that there was con siderable labor mobility overall in Russia, and across sectors and occupations, indica tive of economic restructuring. To better understand if restructuring has been taking place it is important to evaluate job flows, or job creation and destruction rates in Russia.58 High job turnover (with high job creation and destruction rates) is associated with productivity gains and higher efficiency because the process allows the destruc tion of less productive jobs and the creation of more productive ones. The difference between job creation or job destruction, or net employment growth, is a measure of the extent to which new jobs were created in the economy. Improving the allocation of labor inherited at the end of the socialist period involves two types of restructuring. The first step embodying initial restructuring involves shedding of excess labor by firms, while the second involves deeper restruc 56 This finding may reflect the long intervals during which these changes are measured using retrospective data. This could be checked using the one or two year transition matrices of 1994 2000. 57 An interesting research project would be to follow these individuals through the panel and see how they have fared. 58 The gross job creation rate is the sum of all employment gains in expanding firms in a given year, divided by total employment at the beginning of the year. The gross job destruction rate is defined as the sum of all employment losses in contracting firms in a given year divided by total employment. The sum of gross job creation and gross job destruction gives a measure of gross job turnover (reallocation), and the difference is the net employment growth rate. The excess job reallocation rate is the difference in the job reallocation rate minus the absolute value of net employment growth. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 33 Table I .7. Job Flows over Time, Russia and Select Transition Countries (Percentage) Bulgaria Czech Hungary Poland Romania Slovak Russia Rep. Rep. Job creation 1989 1992 . 0.2 1.5 2.2 1.2 6.7 1.6 0.8 1992 94 1.5 4.5 1.0 6.1 ­ 3.7 2.5 Job destruction 1989 1992 . 25 10.2 19.1 14.9 11.2 15.2 3.8 1992 1994 . 4.9 5.3 9.3 5 ­ 6.5 8.6 Job reallocation 1989 92 25.2 11.7 21.3 16.1 17.9 16.8 4.6 1992 94 6.4 9.8 10.3 11.1 ­ 7.2 11.1 Source: Jackman (1998). turing in which enterprises start to change the product mix, undertake investment, and create new jobs. The former stage involves a rise in job destruction rates. The lat ter stage is evident as job creation rates increase and job destruction rates subside (Blanchard 1997). CEE countries restructured more rapidly than Russia. The shedding of excess labor was reflected by a rapid initial rise in job destruction rates in the early transition years. The second stage of restructuring is also evident in lead reformers. Job creation rates, initially very low, started to increase over time, and job destruction rates declined. The combination of these trends meant that job flows fell in most countries from 1989 to 1994 (Jackman 1998; table 1.7). In Russian manufacturing, both job destruction and job creation rates were lower than those observed in CEE lead reformers between 1989 and 1994, but they increased over time, indicating that the economy was restructuring (table 1.7). The increase in manufacturing job destruction is consistent with the gradual rise in unemployment rates in Russia during the same period. However, despite this increase, job creation rates remain lower; and job destruction rates higher in Russia relative to OECD and other transition countries. This result is consistent across a large number of studies on job flows in Russia, e.g. Jackman 1998; Faggio and Konings 1999; Kon ings and Walsh 1999; Russian Economic Barometer 1996. (table AI.32/33/34). It shows that Russia is still in the first phase of restructuring its economy.59 59 It should be noted that there have as yet been no studies of the private services sec tor where job creation rates are likely to be much higher than in other sectors. 34 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY These results on job flows are also confirmed in three recent studies evaluating evidence on job flows in Russia. (Brown and Earle 2002a, 2002b, 2002c).60 These studies also evaluate the determinants of job flows. Brown and Earle (2002a) evalu ates the evidence on gross job flows comparing Russian and Ukrainian manufactur ing firms using annual industrial census data for 1985 91; and 1991 99. The study finds that the job creation rate is low in Russia throughout the period and rose slightly during the 1990s. Job destruction, reallocation, excess reallocation, and employment growth dispersion increase markedly. Excess reallocation appears to be little affected by firm size, wage, capital intensity, and market structure, but it is increasingly associ ated with non state ownership. Job flows are unrelated to productivity growth under Russian socialism, but the covariance of employment share growth with relative pro ductivity becomes strongly positive in both countries by the mid 1990s. These pat terns were common across the two countries but were stronger in Russia which adopted more rapid reforms than did gradualist Ukraine. (Brown and Earle 2002a) Gross job and worker flows in Russian industry and their determinants were stud ied by Brown and Earle (2002b) using panel data from a recent survey of 530 firms selected through national probability sampling. The results imply that job destruc tion and worker separation rates in industrial firms rose in the early 1990s, as did job flows as a fraction of worker flows and layoffs as a fraction of separations. By contrast, job creation and worker hiring rates were flat until 1999, the former low and the lat ter surprisingly high. What are the determinants of job flows? The study finds that heterogeneity in individual firm behavior increased throughout the period. New firms and old enter prises that have been reorganized display much larger flows compared with un reor ganized enterprises. Unions appear to reduce worker flows, but the structure of nei ther product nor labor markets shows a significant impact. Private ownership has ambiguous effects: insider ownership, particularly by managers, is associated with higher worker flows and excess job reallocation, while outsider ownership, particu larly by block holders, is associated with lower flow rates. A measure of adjustment costs constructed from the work time necessary to hire and train a new employee is strongly related to variables usually associated with adjustment costs, including worker wage, education, firm size, capital intensity, and labor productivity, but only weakly to job and worker turnover. Little evidence is found that firms' employment adjustments have become more sensitive to adjustment costs during the transition, but worker and manager ownership are associated with more sensitivity than are other types of ownership. (Brown and Earle 2002b) An evaluation of job flows by Brown and Earle (2002c) using annual 1985 1999 census data for old Russian manufacturing firms calculates the magnitude, covariates and productivity consequences of gross job flows before and after reforms. The job creation rate was low throughout the study period but increased slightly after 1991, while job destruction, reallocation, excess reallocation, and employment growth dis 60 See also Broadman and Rescanitini (2001) for an evaluation of job creation and destruction rates in the manufacturing sector. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 35 Table I.8. Share of Employment in New Private Sector, 1994 2000* Ownership type 1994 1995 1996 1998 2000 Distribution of the employed by type of ownership State owned 0,754 0,683 0,663 0,647 0,605 Mixed 0,073 0,1 0,116 0,113 0,129 Domestic private 0,134 0,172 0,181 0,196 0,217 Foreign 0,040 0,045 0,039 0,044 0,049 Ownership is missing 0.181 0.155 0.148 0.14 0.127 * Using ownership type as definition of new private sector. Source: Definition of ownership type. Calculations from RLMS. Goskomstat Annual Yearbook, 2000, p. 112. persion rose markedly. Excess job reallocation increased in all firm size, ownership, capital intensity, and market concentration categories, while the relationship with average firm wages and labor productivity became positive post reform. Job realloca tion was unrelated to labor productivity growth under socialism but recent contribu tions were strongly positive. Privatization and competition did not increase job flows, but they are associated with significantly higher covariance of employment growth with relative productivity, suggesting that they may have helped to focus job destruc tion in firms with the lowest productivity. To summarize, manufacturing job flows have increased over the 1990s, but job flows are explained mainly by high job destruction rates. Job creation rates remain lower than OECD norms. These changes have likely increased productivity and effi ciency because destroyed jobs are presumed to be less productive than newly created ones. Some evidence, noted above, supports this hypothesis. To further understand the job creation potential of the labor market, it is important to evaluate the new pri vate sector and the nature of entrepreneurship in Russia. H. The New Private Sector In most transition countries, the private sector is the main source of employment generation. Is this the case in Russia? Despite all the attention paid by both academic economists and policymakers to the new private sector in transition economies, there have been few attempts to measure it carefully. One of the reasons may be a fun damental ambiguity in what constitutes a genuinely new entity versus one that is spun off from an old firm or otherwise created on the basis of assets and labor for merly employed in an organization inherited from the socialist period, and there are ambiguities in defining ownership as well. Simply asking managers if the firm is "new" 36 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY or for the founding date does little to resolve such ambiguities. In any case, there appear to be no estimates of the aggregate size of this sector in Russia.61 The approach taken here is to use detailed information from the RLMS, examin ing three different dimensions: firm size, ownership, and founding date. Results are displayed in table AI.35. The distribution of reported employer size has changed remarkably little during the period, with the exception of growth in the 26 100 employee category, which may reflect both the increased size of new private firms and the decreased size of old firms. Reported ownership type has evolved more sig nificantly. The definitions here use the three RLMS questions on whether the respon dent's employer has an owner that is, respectively, state, domestic private, or foreign. Following Earle and Sabirianova (forthcoming), "state" is defined to include any employer reported to have state ownership but neither domestic private nor foreign; "mixed" includes all employers with both state and either domestic private or foreign ownership; "domestic private" includes employers with only domestic private owner ship; and "foreign" includes firms without state and with foreign ownership. The esti mates show substantial growth in the domestic private (13 to 22 percent) and mixed categories, but foreign ownership remained low and state ownership still dominated in 2000 (table I.8). Information on the employer firm's founding date, available in the RLMS since 1995, displays large growth in the post 1994 category, but there is a decline in the fraction of employees reporting firms founded in the first half of the 1990s, the period when new private entry was really liberalized. Finally, table AI.35 shows the results from a definition of the new private sector that includes both the primary activity self employed and employees of firms with no state ownership that were founded after 1988. According to this measure, the new private sector was growing substantially during the 1995 2000 period, from about 23 to 33 percent of all employment. Of these, approximately 40 percent worked individually as self employed, while about 60 percent were employees. The state still remains the domi nant employer, although its role has significantly declined. In 2000, the state employed roughly 60 percent of all workers down from 75 percent in 1994. The state share of employment in Russia is higher than for CEE transition countries as far back as 1996. In that year 40 percent of all Polish workers worked in the state sector (table AI.36). Where did the new private sector employees come from? Characteristics of the new private sector are shown in table AI.38. According to the RLMS data, men tend to be overrepresented in this sector, as are younger people and those with secondary and vocational education. In 1995, individuals with higher education were overrep resented; then the pattern shifted to underrepresentation. The new private sector is 61 Clarke and Borisov (1999) and Clarke and Kabalina (1999, 2000) review various sources for measuring the new sector and analyze their own data from case studies, two oblasts, and five cities. Gimpelson and Lippoldt (1999) [provide a rough indication of the new sector based on the difference between the Goskomstat figures for total employment and for large and medium enterprises; their own analysis pertains to only four Russian regions, and their data for these regions do not permit a distinction between privatized and new private companies. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 37 Figure I.10. Self Employed as a Share of Employed, Select Countries 30 25 employed 20 total 15 of 10 5 0 Percentage 2000) ) (1997 7) 7) 7) 7) (199 (1997) 7) (199 (199 (199 (1997) 96) (19 99 ov ny ry blic and rea (1 s s ia (N St ates rma J apan Ge Hunga gdom Italy Kin h R epu Pol Ko Ru United United ec Cz Source: Annex I frequently connected with the informal or unofficial economy. As this is a question that concerns more directly the types of contracts than the nature of the activities, this issue is taken up in Chapter III. The question of compensation differences is addressed in Chapter II. Self employment. In most countries, private self employment is a very impor tant category of employment, both in the sense that displaced workers from old enterprises may find livelihoods working independently and because it may repre sent the beginnings of entrepreneurship.62 It also is a category that is difficult to meas ure except by population surveys, which were first conducted in Russia only in the early 1990s. An early independent survey for Russia, Bulgaria, the Czech Republic, Hungary, Poland, and the Slovak Republic in 1993, reported in Earle and Sakova (2000a), found that the rates of self employment had increased dramatically already by 1993 in every country except Russia. Figure I.11 shows that the nonagricultural self employment rate was still only around 3 percent in 1993, compared with around 10 percent in the other five countries (and a typical share of about 8 to 10 percent in most OECD countries; see Blanchflower 2000). The share of employers was particu larly low, under 1 percent of employment. What has happened to self employment in Russia since 1993? Unfortunately, the first RLFS reports are from 1999, and they rely on an unusual definition. In the survey 62 Indeed, the self employment rate is frequently taken as an indicator of the rate of entre preneurship. See, for example, Blanchflower and Oswald (1998). Earle and Sakova (2000a) examine the business start up and disguised unemployment sides of self employment in six East European countries, including Russia, in the early 1990s. 38 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure I.11. Evolution of Nonagricultural Self Employment 0.14 0.12 0.10 0.08 0.06 0.04 0.02 0.00 1988 1989 1990 1991 1992 1993 Bulgaria Czech R. Hungary Poland Russia Slovakia Note: Nonagricultural self employment rates are computed for each country assuming a constant 1993 age structure. Source: Earle and Sakova (2000a). analysis discussed above, results are computed on the basis of a standard Western question asked of all employed respondents, namely if their primary job involved self employment, and if so whether the respondent had employees or unpaid family helpers. The RLFS (Goskomstat 1999b) provides a broader definition of self employ ment on the main job, including not only employers and own account workers (ILO 1999) but also members of production cooperatives and unpaid family helpers. Because the latter two categories are not included in the standard definition, the first two of them.are focused on in this analysis, although information is provided on the latter two as well, since they are of independent interest As shown in table AI.39, employers accounted for 1.0 percent, self employed work ers for 4.3 percent, production cooperative members for 1.8 percent, and unpaid fam ily helpers for 0.2 percent, or a total 7.3 percent of all employment in August 2000. The overall self employment rate, by international definitions, was thus only 5.3 percent, a remarkably low number given Russia's level of development and severe recession, and lower than the level reached by any of the East European countries in 1993 (figure I.11). Self employment rates also may be calculated using the RLMS, which has a differ ent structure of questions compared with the RLFS: Respondents are first asked to specify their main activity, and then later asked about supplementary activities. Sev eral definitions are examined in table AI.40, using information both about the main activity and about supplementary activities when the main activity does not involve employment. According to a simple main activity definition, the self employment rate is 2.3 percent in 1994, rising to 6.8 percent in 2000. When supplementary activ ity self employment (but excluding pure subsistence activities) is included, however, the numbers are larger: 8.5 percent in 1994, rising steadily to 17.1 percent in 2000. It is noteworthy that most Russian self employed do not classify themselves as working for their primary activity, a pattern that is only slightly less true of urban than rural respondents and that became still more pronounced between 1998 and 2000.63 UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 39 Figure I.12. Foreign Direct Investment in Selected Transition Economies FDI as percent of GDP 10.0 9.0 8.0 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 1992 1993 1994 1995 1996 1997 1998 Czech Poland Romania Hungary Russia Source: FIAS (2001). Thus, while the private sector has emerged in Russia, and has been creating jobs, gen uine entrepreneurship remains limited. In contrast, as noted earlier, informal self employment is substantial and growing. What factors constrain growth of private sector employment? As small pri vate firms are the main drivers of employment and productivity growth, impediments to their start up hurts prospects for job creation. The main constraints to formal, pri Figure I.13. World Business Environment Survey (WBES) of Obstacles to Investment in Russia WBES Survey of obstacles to investment in Russia 4.00 3.50 obstacle 3.00 of 2.50 2.00 1.50 severity 1.00 0.50 0.00 Perceived rate crime crime Financing Judiciary Infrastructure& regulation instability Inflation of Corruption Other competitive Exchange Streeet Tax Policy Anti Functioning Source: FIAS (2001). 63 Distinguishing the nonagricultural self employed in the 2000 RLMS is, as yet, not possible. 40 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure I.14. Payroll Tax Rates in Russia, CEE Countries, EU and OECD 60 50 rates 40 30 tax 20 10 Payroll 0 Estonia Russia Slovenia Hungary Republic Poland Slovakia average average average EU CEEC Czech OECD Source: World Bank (2000b). vate sector growth in Russia are, among other things, low rates of gross domestic pri vate investment as well as foreign direct investment (FDI), slow restructuring, adminis trative barriers to entry; limited input/product market reforms, and high payroll taxes. Investment. According to the FIAS report (FIAS 2001), gross domestic investment in the Russian Federation fell sharply between 1990 and 1998, and has only recently started to increase. The Russian Federation received less than 1 percent of GDP from 1992 to 1998 as FDI, as compared with 3 to 4 percent or more in Poland and Romania, and even higher rates in many other CEE countries. Figure I.12 from the same report shows how Russia compares with other large, emerging market economies and other European transition economies, respectively. The low rate of FDI is a particular con cern as this type of investment has a large impact on labor productivity and overall economic growth, by providing not only new sources of capital, but more importantly, new technology, and the most effective management and marketing methods. The composition of investment is also worrisome. The main activities in which investment has been concentrated are the extractive sectors, rather than high technology export oriented manufacturers, which are so prominent in CEE countries. Restructuring. Russia places well below CEE countries on the EBRD privatization index (World Bank 2001a) which measures the extent to which governments have progressed on privatization and other aspects of private sector development. While restructuring or layoffs have occurred, Russia still has a higher share of large enterprises than found in most Central and Eastern European countries. (FIAS 2000).64 In the lat ter countries, there is an exactly opposite pyramid shape structure of firms: a few large firms, more medium sized firms, and a much larger number of small firms, many of whom are new start ups. Large enterprises tend to have more vertically integrated supply chains that limit opportunities for private sector entry. They also have the 64 This section is drawn from FIAS (2001). UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 41 structural advantages that come from market dominance, favored access to infra structure services, and protection from interregional trade and investment (Broadman and Recanatini forthcoming). As such, they do not promote productivity growth. Limited input market reforms. In addition to lack of competition in product markets, reforms of financial markets and land have been limited. Russia scores low relative to CEE countries on a market reform index that measures the pace of market reforms (World Bank 2000a).65 Weak financial markets often provide directed credit to select enterprises and financial conglomerates on very concessionary terms (East erly and Da Cunha 1994), and property rights, essential for promoting competition, remain ill defined. Administrative barriers. The lack of rule of law and the absence of a level play ing field for firms also constrain investment in the private sector. Institutional and administrative barriers, including arduous licensing, registration and inspection requirements, and corruption further compound this problem. The FIAS study (2001) finds that the five main administrative barriers to investment in Russia are problems with taxes, policy instability, corruption, inflation, and the judiciary.56,57 (figure I.13). Payroll taxes. Payroll taxes can have an adverse impact on equilibrium employment, as suggested by international evidence. Payroll taxes in Russia 36 percent of payroll are lower than in CEE countries, but higher than in most OECD countries68 (figure I.14). Evidence from OECD countries finds a negative impact of payroll taxes reduction on unemployment (Daveri and Tabellini 2000).69 A reduc tion of tax rates by 5 percent reduces unemployment by 13 percent (or from 8 per cent to 7 percent, for example) (Nickell and Layard 1997). Evidence from CEE countries also confirms that payroll taxes probably increase equilibrium unem ployment (World Bank 2001a, 2001b). In Russia, weak enforcement of the law on collection of taxes and wages (see Chapters II and III) means the main impact of 65 De Melo, Denizer, and Gelb (1996) index.. 66 The World Business Environment Survey (WBES) of 80 countries carried out in early 2000 also finds "tax and regulations" at the top of the list of complaints by businesses in Russia (to a degree worse than most other CEE countries), followed by inflation and policy instability. Euro money, in a recent survey of FDI in Russia, also put tax issues at the top of the list of prob lems, followed by "insecure property rights", customs, and "risk of political change." 67 "Transparency International" ranked Russia 82nd (alongside Kenya) out of 90 countries in its corruption perception index for 2000. Finally, according to The Wall Street Journal and "Heritage Foundation"'s Index of Economic Freedom, Russia ranked 127 out of 155 countries, with especially poor ratings for monetary policy, the fiscal burden, trade policy, and regulations. See http://www.transparency.org/documents/cpi/2000/cpi2000.html ("Transparency Interna tional") and http://www.heritage.org/news/2000/nr110100indexoverview.html ("Heritage Foundation") (FIAS 2001). 68 Of the total tax, 28 percent goes for pensions, 3.6 percent is for medical insurance, and 4.4 percent is for social insurance. Firms also pay 13 percent personal income tax on behalf of employees. Thus, if an employee receives 100 rubles in gross wages per month, he/she will take home 56 rubles, slightly over half of his/her salary. There is a deduction of 36 rubles for payroll and a further 8 rubles for personal income tax. 69 Theoretically, the impact of payroll taxes on labor and wages depends on the relative elasticities of demand and supply of labor. The actual impact of the tax in any country therefore requires empirical analysis. 42 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY high payroll taxes may be tax avoidance and the informalization of the economy. According to Johnson et al. (1997), Russia's informal sector as a share of GDP ranks higher than all CEE transition countries. (World Bank 2001b) The reduction of pay roll taxes in Russia over the medium term (in concert with improvements in the efficiency of social insurance programs) may therefore reduce informalization of the economy. As the economy formalizes, (all else equal) lower payroll taxes may also reduce equilibrium unemployment. Summary and Conclusions This chapter has explored the responsiveness of labor market aggregates and flows in Russia to economic changes. The main conclusions are as follows: · The deep and prolonged economic decline in Russia between 1990 and 98 led to a significant decline in employment. Employment fell and non participa tion (particularly self employment in agriculture) and unemployment increased. The drop in the labor force was largest in the youngest and oldest age groups. The recent economic recovery has reversed this trend, and employment has increased pulling both unemployed and nonparticipants back to work. · The recent growth in employment has not been very responsive to output growth (as in CEE countries). Despite recent declines, the level and duration of unemployment cannot be considered low relative to OECD countries, and the regional variation in unemployment rates remains high (higher than Poland and the Slovak Republic, for example). · Labor productivity has increased as a result of recent economic growth. Employers have reallocated the existing work force rather than increase new hires. However, large declines in labor productivity in the past decade mean that there is a considerable gap in labor productivity levels between Russia and CEE countries. The main reason for the decline in labor productivity dur ing the transition was overstaffing in the face of output declines as a result of weak incentives to managers to lay off workers. Most evidence indicates that adjustment in hours or secondary employment did occur, but was not as important as adjustments in primary employment suggest. Although the labor surplus declined during the past decade, its continued existence is evident in the small rise in employment relative to recent output growth. · Russia did restructure in the face of economic declines. Faced with declining output, enterprises did lay off workers to cut costs, as evidenced by the grow ing rates of unemployment noted above. Labor transitions post 1991 increased, and there was significant occupational and sectoral change in employment consistent with a move to a market economy. The growth of the private sector facilitated these transitions. Recent economic growth has accel erated these trends. UNDERSTANDING EMPLOYMENT: LEVEL, COMPOSITION, AND FLOWS 43 · Given that real wages and labor productivity are increasing from a very low base, it will require considerable growth for Russia to close its wage and labor productivity gap with fast reforming CEE countries. The challenge will have to be met by private sector led growth. The private sector share of employment is still not very high, and genuine entrepreneurship is limited. Job creation rates in manufacturing have increased somewhat in the 1990s, but remain well below OECD and high income transition country norms. This result may reflect barriers to entry to new enterprises, such as limited financing for start ups, undefined property rights, licensing and other fees, high payroll tax rates, and lack of rule of law. Addressing these constraints will be critical for realiz ing sustained economic growth. · Addressing structural factors that constrain matching of supply and demand in the labor market also will be important for reducing the level, duration, and regional variation in unemployment rates. These factors include: (a) A Skills Mismatch. Older workers with low levels of education and obsolete skills have the highest rates of unemployment and the longest duration of unemploy ment. (b) A Regional Mismatch. High unemployment regions are concentrated in Eastern and Western Siberia and the North Caucuses. These regions have lower expenditure per capita, high poverty rates, high birth rates, and a high industrial share of output. High unemployment rates in these regions might be exacerbated (in the short run) by further economic restructuring. The evi dence on the extent of regional mobility is mixed (and requires further inves tigation), but recent studies suggest that the lack of affordable housing may limit worker flows across regions. Addressing these structural mismatches will require a focus on passive and active programming and on factors that may impede regional labor mobility. The following chapter looks at the wage structure in Russia in order to complete the picture of the labor market in Russia. Chapter II Understanding Wages: Structure, Uncertainty, and Inequality The previous chapter evaluated employment adjustment in Russia. As a companion piece, this chapter looks closely at incentives and returns in the labor market. It asks the following questions: Are wages increasingly determined by market forces in Rus sia? Do non market factors still influence the level and dispersion of wages? Which workers have benefited the most, and which have lost out, in this wage adjustment process? The answers to these questions also shed some light on the nature of labor market flexibility in Russia and on the question of whether Russia has been restruc turing during the past decade. A. Level and Determinants of Wages The level of real wages. As noted in the previous chapter, wage measurement in Russia is subject to many caveats. To briefly summarize, officially reported wages may overstate actually received wages, because of wage arrears and forced in kind substi tutes, but they may also understate because firms have become adept at hiding salaries from the tax authorities (on the order of 20 percent, as estimated by Goskom stat). In addition wage measurement also does not take into account the changing availability of consumer goods. Keeping this in mind, the trend in average real wages is shown in figure II.1.70 It reflects some important aspects of labor market adjustments at the aggregate level, including some of the large macroeconomic events in Russia's transition: first, a large increase in wages leading up to the "big bang" price liberalization of January 1992, fol lowed immediately by a sharp drop associated with the price jump. Next, there is a relatively stable period for most of 1992 94, followed by another sharp fall associated with the inflation after the financial crisis of late 1994. From early 1995 until July 1998, real wages were on a gradually rising trend, only to lose still more ground dur ing autumn 1998. From January 1999 to January 2001, wages have again risen sub stantially, albeit starting from an all time low. Keeping this in mind, cumulative wage growth in 1999 is higher in Russia than in higher income CEE countries The longer time series presented here helps to put some of the frequently dis cussed dramatic events concerning wages into clearer perspective. While real wages 70 The monthly time series of the CPI deflated real wage, presented in figure II.1, show a spike in December of each year, when bonuses are typically paid. 46 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure II.1. CPI Deflated Real Wage Rate Due, 1991 2001 (Jan. 1991 = 100) 160 140 120 100 80 60 40 20 0 91 19 n 19l 91 92 29 t n 29l 92 93 39 t n 39l 93 94 49 t n 49l 94 95 59 t n 59l 95 96 69 t n 69l 96 97 79 t n 79l 97 98 89 t n 89l 98 99 99 t n 99l 99 00 00 t n 00l 00 01 t n Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Apr Ju Oc Ja Sources: Russian Economic Trends and Goskomstat (2001a). declined by nearly 70 percent in cumulative terms between 1990 and 2000 -- much more than realized by many CEE countries (see table AI.2), this decline was not monotonic. The huge crash in real wages in early 1992 appears much smaller when compared with early 1991 levels than to those closer to the end of the year. The recent rise in real wages following the 1998 crisis, which has sometimes been hailed as a sign that genuine restructuring has started in Russia, also appears in different per spective. As figure II.1 shows, this was not the first period of rising real wages, nor was the level reached by January 2001 higher than that in the three years prior to August 1998. The periods of stable or gradually rising real wages can be interpreted as reflect ing sticky adjustment of nominal wages to the price shocks of January 1992, late 1994, and late 1998. Although nominal wages did rise significantly during the infla tionary bursts, they still lagged the price changes quite significantly. As noted in the previous chapter, the fall in real wages was deeper and more pro longed than in CEE transition countries. The fall in real wages reduced household income, and changed its composition. The wage share of household income fell. By 1998, wages comprised only 40 percent of household income, much lower than in many CEE and OECD countries. Consistent with the growth of subsistence agriculture as a primary and secondary activity in Russia, the share of income from self employ ment increased (World Bank 2000a).Why were wage declines so large? The large drop in output and labor productivity, as well as the high rates of inflation realized over the past decade, explain the significant decline in real wages. However, the lack of enforcement of labor legislation and the absence of effective institutions giving voice CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 47 Table II.1. Changes in Real Wages by Characteristics of Firms and Workers, 1998 2000 Worker Mean Stand. dev. Firm characteristics Mean Stand. dev. characteristics Total [N=2474] 0.172 0.541 Rural 0.123 0.547 Female 0.146 0.526 Urban 0.191 0.538 Male 0.204 0.559 Sectors [N=2458] Age Industry 0.264 0.520 15 24 0.328 0.662 Agriculture 0.013 0.584 25 34 0.201 0.558 Transportation/Construction 35 44 0.173 0.534 Public Services 0.109 0.502 45 54 0.147 0.507 Other Services 0.229 0.591 55 72 0.097 0.539 Employment/ Firm [N=1938] Education <26 0.170 0.613 Elementary 0.068 0.559 26 100 0.118 0.490 Secondary basic 0.186 0.541 101 500 0.186 0.528 Vocational 0.186 0.583 >500 0.240 0.502 Secondary/ 0.184 0.527 Ownership Professional [N=2186] University 0.176 0.520 State owned 0.142 0.499 Job to job mobility Mixed 0.238 0.553 Job stayers 0.148 0.502 Domestic Private 0.212 0.636 Job movers 0.313 0.710 Foreign 0.374 0.470 Notes: Sample is restricted to all employees aged 17 72. Changes in real wages between 1998 and 2000 are computed as a difference in log of usual monthly wages deflated by the national CPI. Characteris tics of firms and workers are taken from 2000. Source: Calculations from RLMS. to worker concerns may also explain this phenomenon.71 The issue of labor market institutions is discussed extensively in Chapter III. Who has gained and lost in terms of wages in the big fluctuations in Russia's econ omy in the 1990s? Table AII.172 provides information on the mean and standard devi 71 This is not to say that enforcement of restrictive legislation may have had other adverse consequences in limiting labor and employment adjustment as well. 72 Tables prefixed by AII refer to tables in Annex II. 48 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.2. Results of Simple Earning Functions, RLMS, 1992 2000 for Women 1992 1994 1996 1998 2000 1998 2000 1 2 3 4 5 6 7 Log of usual weekly 0.381 0.163 0.241 0.355 0.349* 0.527* 0.579* hours of work Schooling (years) 0.038* 0.074* 0.056* 0.077* 0.076* 0.085* 0.090* Experience 0.026* 0.020* 0.020* 0.031* 0.039* 0.030* 0.036* Experience squared 0,052* 0,037* 0,046* 0,064* 0,085* 0,063* 0.072* Constant 3.160* 9.875* 10.858* 3.167* 3.816* 2.260* 2.463* N 3133 1968 1693 1664 1737 1915 1952 R2 0.303 0.267 0.331 0.307 0.294 0.419 0.42 Notes: * significant at the 1 percent level. Sample is restricted to employees aged 15 72. In columns (1) (5) EXP is measured as potential labor market experience (age minus schooling minus 6). In column (1) the dependent variable is log of after tax actual monthly wages received in the previous month. Sixteen regional dummies are included. In columns (2) (5) the dependent variable is log of imputed contractual monthly wage. Contractual monthly wage is computed following methodology of Earle and Sabirianova (2000). In columns (6) (7) EXP is measured as actual labor market experience (data on actual labor market experience became available since 1998) and the dependent variable is log of usual monthly wage. Thirty eight regional dummies are included but not shown here. Source: Calculations from RLMS. ation of wages calculated from the RLMS on the CPI deflated average real wage for different population groups, and for the actually received wage and the imputed con tractual wage, respectively. All groups show sharp and usually monotonic declines from 1994 to 1998 and rise thereafter. Because of the reduction in wage arrears (dis cussed below), the actually received monthly wage recovers much more than the contractual wage. Table II.1 also shows that, while the recovery since 1998 has benefited all worker groups, the relative gains across socio economic groups differ. Real wages increased more in absolute terms for highly educated workers, in urban areas, in the private sec tor. Older, less educated workers in agriculture living in rural areas have realized hardly any absolute wage gains at all. However, the 1998 base was so low that every group still remains worse off relative to any earlier year. Wage differences. In most developed market economies, differences in wages are attributable to differences in skills, experience, industry and gender. What are the determinants of wages in Russia? Education. Have wages increasingly started to reflect returns to education? The empirical evidence exploiting data from the All Russian Center for Public Opinion Research (VTsIOM) suggests relatively modest rates of return to education before economic reforms started: 3.1 percent for males and 5.4 for females in 1991 (accord CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 49 ing to Brainerd 1998). These results are generally consistent with the estimates from the earliest rounds of the RLMS (1992). Table AII.3 shows that in the first year of reforms, returns to additional years of schooling were 3.8 percent for women and 3.4 for men, while returns to experience were around 2.4 percent. (See also table II.2) Since the breakup of the Soviet Union and market liberalization in the former centrally planned countries, there have been a number of studies of human capital during the transition to a market economy. Two contrary hypotheses regarding the returns to human capital have been tested: the first states that market liberalization and unconstrained wage setting should shift returns in favor of more educated peo ple, while the second proposes that skills and experience acquired in the previous sys tem have become obsolete in the new market conditions. This human capital deval uation can cause a decline in returns to human capital. A review of numerous studies suggests that returns to education increased during the transition period (Svejnar 1999). The estimates of the extended returns to human capital in transitional Russia, shown in table AII.4, are also consistent with the overall picture in other former centrally planned economies. Empirical findings suggest that the average rate of return to schooling in Russia rose to 8 to 9 percent of an increase in real earnings for each additional year of schooling holding hours of work, experi ence, and regional residence constant. (See Table II.2 for this increase for women) These results support the hypothesis on the positive impact of market liberaliza tion on returns to schooling, especially for university graduates. These results, together with low rates of unemployment among university educated workers, also demonstrate the importance of investing in education in Russia. There are some caveats. We observe fluctuations in the return to schooling for some time periods and for some types of education such as vocational training. The reasons for such changes, including changes in the labor force composition, the devaluation of some skills, and the decreased demand for narrow specialists educated in the previous system, are dis cussed in Nesterova and Sabirianova (1998). An additional issue is the poor measure ment of wages in the context of large arrears, as discussed in Earle and Sabirianova (forthcoming). Gender. Earnings functions estimations not shown in these tables indicate that the gender wage gap has increased over the transition period: 44.9 percent in 1998 versus 32.5 in 1992. On the other hand, after controlling for occupations and industries, the gender wage gap has declined. Estimation results from a more extended specification, including job tenure, type of ownership, and founding date are shown in table II.4. In these specifications, the gender gap is roughly constant at about 45 percent, while the schooling and experience effects are similar to those from table II.3. Occupation/Industry. The estimated differentials across occupations and indus tries reported in table II.4 also are quite substantial. Officials and managers are esti mated to have a higher level of hourly wages while clerks, service, and unskilled work ers comprise the low paid group of employees. Among industries, the fuel sector has the highest level of wages while workers in agriculture, health, and social services receive the lowest wages. Firm ownership. Turning to the results concerning differences across forms of ownership, the wages of workers in foreign owned firms relative to those in state 50 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY owned enterprises rose strongly to a premium of more than 50 percent. Both domes tic private firms and those of mixed ownership pay a premium of below half that: around 20 percent. The premium for working in a new firm (defined here as founded after 1987) fluctuates across the years and definitions; in 2000 it was over 20 percent when the measure is the imputed contractual wage, while it was only 12.6 percent when measured by the reported "usual" monthly wage. Clarke and Kabalina (2000) also find that wage levels are higher in the private sec tor than in state or privatized enterprises. On average, employees in the de novo pri vate sector report earnings that are 40 percent higher than those working in tradi tional enterprises.73 Even after controlling for individual characteristics, they find that the wage gap is still 20 to25 percent. Their analysis shows that workers with high lev els of human capital (that is, prime age, high educational attainment, managerial and administrative occupations) benefit most from private sector employment. Since employees with these characteristics tend to be above average earners in all sectors, wage differentials are largest in de novo private enterprises. Tenure. The tenure effect appears to increase from 1995 to 1998 and then retreat slightly in 2000. The very low tenure effect estimated in 1995 is consistent with the findings of other studies (Flanagan 1995; Chase 1998) that job tenure from the social ist period had little value given the shocks of transition. The increase in the tenure effect to 1998 would suggest that the normal effect was being re established, while the retreat to 2000 implies that a change in the returns to existing jobs may have been one result of the 1998 crisis. Experience. Table II.4 also shows that the experience effect on earnings in Russia, while initially falling and then rising slightly during the 1992 2000 period, is rather small compared with that implied by data from developed market countries. The small experience effect could be the result of the changing nature of the Russian economy that rewards younger, more mobile, more active, and more adaptive people. Returns to experience declined gradually from 1992 to 1996, approaching zero in the case of men. However, the 1998 data suggest that the accumulation of new experi ence and specific human capital from the work in the market economy raises the returns to experience. To summarize results concerning human capital, the positive effect of transition on the returns to education -- particularly in the private sector -- and new market experience, and its negative effect on the returns to past experience can be consid ered important stylized facts of the transitional process in Russia. Labor costs. Goskomstat (1999c) provides some measures of relative labor costs across industries and ownership types reported in tables AII.5. Hourly and monthly wages are shown, as well as the non state/state ratio, by industry. These figures are consistent with the RLMS, showing the highest earnings in electricity, fuels, and non ferrous metals, and the lowest in textiles, food, and restaurants and catering. Employ ees of nonstate firms are reported to earn a significant premium over their state 73 This is based on the 1998 Institute for Comparative Labor Relations Research (ISITO) survey of 4,000 households in Samara, Kemerovo, Lyubertsy, and Syktyvkar. CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 51 owned firm counterparts in most sectors, with the highest premia in fuels, wood and paper, textiles, communications, and other services. The reported premium is nega tive in several sectors, however: chemicals, machinery, construction materials, trade, and finance. Particularly concerning the latter two categories, it is possible that unre ported wages account for (or even exceed) the difference. B. Nonwage Compensation Practices Fringe benefits. An important issue in understanding wage and compensation behavior is the important role played by fringe benefits in both the socialist and tran sition periods.74 The provision by firms of "social benefits," fringes including housing, medical care, childcare, vacation facilities, and so on, has attracted considerable attention (for example, Commander and Jackman, 1993) from Western economists concerned that they pose barriers to restructuring. While privatized firms were legally required to divest housing and medical facilities, analysts say that in practice the divestiture was somewhat incomplete, given the poor capacity of local authorities to take over these responsibilities. The precise magnitudes are difficult to quantify, but table AII.6 shows the compo sition of total costs according to Goskomstat (1999c) estimates from an enterprise survey. The table shows that the share of cash wages increased from 1995 to 1998 (the immediate post privatization years) as housing and recreation costs declined. "Social contributions" denotes the mandatory contributions to social security funds, roughly constant across industries. But the fringe benefits vary quite substantially, with higher rates of housing costs in manufacturing, especially nonferrous metals. The data show that some significant divestiture and restructuring has occurred. More detailed evidence, although consistent to a broad degree with the aggregate figures, comes from another firm survey of manufacturing firms, shown in table II.3. The table shows the changes in the proportion of firms providing each kind of ben efit during the 1990s. Nearly all types of fringe benefits have declined, but none has have disappeared, except for "other goods not produced by the enterprise."75 Medical services and pro fessional training were still, in 1998, provided by more than half the manufacturing firms in the sample. The biggest declines are recorded for housing construction, kindergartens, and entertainment and culture. Again, the data appear to reflect sub stantial restructuring as well as some inertia in behavior. A final piece of evidence relies on the RLMS questions asked for the first time in the 2000 survey for a recent picture of the situation. Most interesting, the results, dis played in table AII.7, show a clear relationship of benefit probability with firm size: 74 See Rein, Friedman, and Woergoetter 1997 for a collection of papers on the topic. 75 Note that fringe benefits are part of a worker's contractual compensation and therefore should not be confused with forced in kind substitutes for wages, a practice discussed in section VI, below. 52 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.3. Provision of Fringe Benefits, by Type, 1990 98 (Percentage of All Firms) Types of fringe benefits 1990 1994 1998 Land plots or cultivation services 37.5 30.5 19.0 Subsidy for housing purchase or repair 35.0 29.5 20.0 Construction of housing for employees 45.0 34.0 18.0 Goods produced by enterprise 24.5 22.0 20.0 Food not produced by enterprise 28.5 25.0 15.5 Other goods not produced by enterprise 11.5 9.5 7.0 Catering during work time or covering costs 54.5 50.0 41.0 Utility subsidies for employees 20.5 18.5 14.0 Medical services or own policlinics 63.5 62.5 55.5 Vacation facilities 62.0 56.0 43.5 Professional training 78.0 70.5 59.0 Kindergartens 66.0 54.5 32.0 Entertainment and cultural facilities 56.0 45.0 27.5 Other fringe benefits 22.4 20.0 19.2 Average number of fringe benefits 6.3 5.5 4.0 Note: Sample is consistent across years (N = 200). Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (1999). employees of larger firms are much more likely to get every type of benefit. The pro portions of RLMS worker respondents reporting that they received each type of ben efit are much lower than the proportion of firms reporting they paid them in the firm survey. This may be explained by the restricted coverage of the firm survey to the manufacturing sector, in which firms are larger and such benefits are likely higher, and by the possibility that not all workers in a firm are recipients. In summary, this section has shown that there has been considerable evolution in the earnings and compensation structures in Russia during the 1990s. The trend toward higher remuneration of human capital, particularly schooling, continued after 1998, but it does not appear as strong for vocational education, and has not drastically accelerated as a result of growth. While experience or job tenure in the socialist era does not yield higher returns, more recent labor market experience appears to be paying off. A restructuring of compensation away from fringe benefits and toward cash wages appears to be under way, although the fact that the former are not monetized makes them hard to value and therefore to calculate their share in total compensation (from the worker's viewpoint). However, fringe benefits and services are still important and are mainly concentrated in larger firms. CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 53 Figure II.2. Real Wage Arrears, 1990 2000 450 400 350 300 250 200 Percentage 150 100 50 0 92 92 93 93 94 94 95 95 96 96 97 97 98 98 99 99 00 march sept. sept. march sept. sept. sept. sept. sept. sept. march march march march march march march Year Source: Goskomstat (2000b). Wage arrears and in kind substitutes. One of the peculiar aspects of the transi tion of the Russian labor market relative to developed countries is the use of wage arrears and forced in kind substitutes. In both cases, workers are denied timely payment of their cash wages, either because of postponement of payment or a choice between no payment and payment in the form of some goods, either those produced by the firm or acquired by it in the course of its own barter transactions. Studies of the Russian labor market, focusing especially on wage arrears, have generally treated these practices as a way for firms to reduce their wage costs. As in other transition economies, Russian firms have faced tremendous shocks to their product and factor markets over the past several years, and have come under pressure to reduce output and costs.76 Delaying wage payments may be a particularly effective cost reduction mecha nism under high inflation. Viewed from the standard paradigm in which some form of wage rigidity is taken as the cause of involuntary unemployment, arrears have even 76 The pressure to cut labor costs has been particularly heavy because of the initial (pretransi tion) situation of overstaffing in the industrial enterprises. See Commander, McHale, and Yemtsov (1995) for a recent analysis. For more information on Soviet labor markets, see Granick (1987). 54 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY attracted some implicit or explicit praise for their contribution to the low levels of layoffs and unemployment in Russia. Layard and Richter (1995), for instance, portray wage arrears as a form of "wage flexibility... explained by the willingness of workers to accept pay cuts in order to preserve jobs." In its 1995 survey of the Russian economy, the OECD praised the "remarkable flexibility...of real wages" and the use of "wage arrears ... to finance this employment surplus."77 This line of thought has provided some answers to the question why Russian employers may have favored wage cuts over layoffs as an adjustment mechanism, but it does not explain why many of them have adopted wage delays and in kind substi tutes as preferred practices. This question is important because, while wage arrears clearly imply a reduction in the effective real wage, they also differ from wage cuts in several important respects, both conceptually and empirically. To begin with, arrears involve uncertainty about the timing and extent of eventual payment; this uncer tainty is perhaps a more important welfare consequence of arrears than the effective real wage reduction. The value of in kind substitutes is also uncertain, as workers must frequently try to sell the goods on street markets. Both practices also imply vio lations of the labor contract, not renegotiations, which may have implications for the popular faith in the rule of law in the transition environment. Furthermore, the theo retical implications of arrears for worker quit behavior also differ from those of a sim ple wage cut, discussed further below. Casual empirical observation also suggests differences between wage cuts and the practices of arrears and in kind substitutes. First, Russian workers perceive wage arrears as different from wage cuts, as evidenced for instance by their tendency in opinion polls to rate arrears as a much larger social problem than low wages (Javeline 1999). Moreover, real wages have hardly been rigid in Russia, certainly not in the aggregate and over a sufficient time span, as high inflation has been associated with large increases in nominal wages and drastic declines in real wages during the 1990s. From September 1994 to 1996, for instance, the average nominal wage rose 235 per cent, while the real wage fell 21 percent. Russian employers were repeatedly agreeing to nominal wage increases and then declining to pay them, or substituting goods for them. Finally, wage arrears are correlated with measures of demand shocks and finan cial distress, but the relationship is only moderately strong. Thus, it is important to search further for additional explanations of wage arrears and in kind substitutes, particularly ones that treat them as distinct practices from wage reductions. Closely related is the notion that financial distress is responsible for wage arrears and barter payments are accounts that focus on liquidity problems in the Russian economy.78 According to one version (usually reported by managers to workers), cus 77 Desai and Idson (2000) Gimpelson (1998), and Lehmann, Wadsworth, and Acquisti (1999) also analyze wage arrears from the perspective of wage adjustment. Brainerd (1998) studies the evolution of the wage structure in Russia from 1991 to 1994, but does not address the problem of wage arrears, although they were quite sizable by 1994. 78 According to Clarke (1999) for instance: "The worst non payment of wages is not found in enterprises which are bankrupt, but in the most prosperous and profitable enterprises in Rus sia. They do not pay wages not because they cannot afford to pay wages, but because they do not have the live money to pay wages." CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 55 tomers have failed to pay on time, thus the firm has no money to pay wages. Another version has it that with little external finance available, firms take advantage of the possibility of interest free loans from their workers. In support of the illiquidity expla nations, it is true that wage arrears have risen in tandem with enterprise and tax arrears (Ivanova and Wyplosz 1998). On the other hand, wage arrears are peculiar in that, unlike the other two types of arrears, they are virtually unheard of in market economies; and while barter among firms (for example, "counter trade") is rather usual in any economy, the same cannot be said for the practice of forcing workers to accept barter payments postcontractu ally. Concerning arrears, Alfandari and Schaffer (1996) and Clarke (1999) show that the levels of overdue interenterprise debt in Russia have not been not particularly high by market economy standards, and tax arrears in OECD countries are of course also far from unknown. Moreover, there are a number of additional reasons to remain dissatisfied with the illiquidity story. An account relying on unexpected liquidity shocks is inadequate to explain why wage arrears and forced in kind substitutes could persist for several years in Russia, as firms have had time to adapt their expectations and to adjust in other ways than by not meeting their contractual obligations to their workers. If the expla nation focuses rather on long run illiquidity in some firms, then the implication is that workers voluntarily agree to make a loan to their employer (as also suggested by Lehmann, Wadsworth, and Acquisti 1999) or to accept a lower implicit wage in the form of less valuable commodities. But again it is necessary to point out that arrears and forced substitutes imply vio lation of the wage contract, not renegotiation. Certainly the outrage, strikes, and other protest behavior (which is discussed in the section on unions and strikes in the next chapter) suggest that workers have not voluntarily agreed to become credi tors.79 A loan also implies some certainty, at least a formal promise, of repayment, but the fact is that receiving back wages in Russia is highly uncertain.80 Finally, empirical analysis shows that wage arrears and forced substitutes are only moderately corre lated with measures of illiquidity. Thus, while it is clear that wage arrears and forced substitutes are related to the broader patterns of economic and financial decline in Russia, they have a somewhat independent dynamic. Before returning to the issue of worker reactions to arrears and substitutes, and the implications for the regional concentration and persistence of arrears, this section discusses some other factors -- in addition to declining per formance and liquidity problems -- that may affect the incentives of firms to adopt 79 One might ignore worker attitudes and argue that arrears are part of an implicit contract, but there is no evidence of any compensating differentials associated with arrears. To some extent, the issue is semantic, as it is still of interest why implicit contracts should take this peculiar form in Russia but in few other economies around the world. 80 Even this could be part of an implicit contract extended to include risk sharing, with repayment of back wages contingent on future firm performance. It is hard to imagine workers voluntarily accepting such an arrangement under any circumstances, much less so in the non transparent environment of Russia, where workers would face insurmountable difficulties in observing performance and enforcing such an agreement. 56 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY these practices: fiscal policies and soft budget constraints, poor corporate governance and managerial self dealing, and worker ownership arising from the Russian privati zation process. Taking each of these in turn, some aspects of Russian fiscal policies may have increased wage arrears and substitutes as firms have sought to reduce tax payments or extract subsidies. In general, high tax rates, on both wages and profits, give firms an incentive to hide cash, and the lack of effective enforcement and accounting trans parency makes it easier for them to do so. Paying wages may attract the tax collector's attention, particularly since enterprises are legally permitted to use only a single bank account for all types of payments; thus wage nonpayment or forced substitution may be useful to signal inability to make tax payments.81 In a similar vein, arrears and forced substitutes may result from attempts by enterprises to extract subsidies from the state (a speculation that appears in a number of articles, for example, Alfandari and Schaffer 1996), especially by firms with close ties to federal or local governments or those with greater bargaining power. An additional aspect of fiscal policies was the frequent sequestration of budgetary funds by the Ministry of Finance in order to reduce the federal budget deficit in the early and mid 1990s. According to the Institute for the Economy in Transition (1994, p. 35), for instance, every expenditure line in the fourth quarter of the 1993 federal budget was sequestered by 20 percent. High inflation and political gridlock led to this unorthodox macroeconomic policy, which resulted in unpaid bills at defense con tractors and late wages of bureaucrats, teachers, and health care providers.82 Seques tration may explain high arrears under state ownership and in particular sectors of the economy, but by itself cannot account for the broader phenomenon. A second aspect of the Russian environment, particularly relevant for understand ing wage arrears, is the poor monitoring of managers, particularly in the large state owned and recently privatized companies. As noted earlier, it is frequently alleged that managers have engaged in massive asset diversions, which would have had the indirect effect of impoverishing their companies (thus making them less capable of paying their wage bill), but such actions may have also involved the direct theft of funds intended for the workers. A further incentive for the diversion of wages may have been the large borrowing of the Russian Government to finance an outsized budget deficit. Short term treasury bills were offered at extremely high interest rates (varying from 30 to 150 percent during the 1994­96 period of rather low inflation and mostly fixed exchange rates). Thus, by postponing some payments, managers stood to earn enormous returns -- on their workers' money. A final set of considerations influencing managerial decisions on arrears and forced substitutes concerns the massive worker ownership that arose from the Russ ian privatization process. One implication of worker ownership could be a greater willingness of workers with equity stakes to help their firms out of a liquidity crisis, by making a voluntary "loan" as discussed above. An alternative possibility is that man agers may have used wage arrears and in kind substitutes to try to force their (even 81 Hendley et al. (1997) make similar points with respect to barter deals between firms. 82 See also Gimpelson (1998). CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 57 more liquidity constrained) employees to sell their shares shortly after the latter became shareholders -- a phenomenon that is frequently alleged to have taken place, and for which there is some anecdotal evidence.83 None of the factors discussed above -- neither the economic depression and illiq uidity, the fiscal policies, the poor monitoring of managers, nor worker ownership -- provides a satisfactory explanation for two particularly puzzling aspects of wage arrears in Russia: persistence over time and variation across regions, regularities that are documented in several studies.84 The key to understanding these regularities con cerns the worker mobility response to arrears: how mobility is attenuated, promoting persistence, and how mobility varies geographically, contributing to regional variation. Researchers have pointed out that worker quits in response to arrears could be reduced by a lack of outside opportunities (Layard and Richter 1995; Lehmann, Wadsworth, and Acquisti 1999). If workers' alternatives are poor -- because of high migration costs and few local options -- then the firms may be able to exploit their low bargaining power and reduce their quasi rents, particularly in the many "one company towns" and "mono industrial cities" remaining from the planning period in Russia.85 Layard and Richter (1995) also argue that sluggish quit behavior in Russia may result from the desire of workers for continued access to fringe benefits, pro duction facilities, and possible opportunities for pilferage at the enterprise. Although these considerations apply equally to wage cuts and wage arrears, there is also an important difference in worker responses to these two actions. While both effectively lower wages, tending to raise quits, arrears also result in an upward tilt of the wage tenure profile. If the worker expects at least some of the back wages to be paid in the future, this deferred compensation effect provides an incentive to remain longer with the employer, and overall the effect of arrears on quits is therefore theo retically ambiguous.86 Furthermore, the incentive not to quit is greatly strengthened by an institutional consideration peculiar to Russia, namely that court enforcement (and any other type of third party enforcement available to workers) is so weak that a worker who quits a job generally loses forever any chance to recover any of the back wages owed. Thus, the tilting of the earnings tenure profile together with the lack of contract enforcement, the market power of many employers, and the nature of local labor markets in Russia serve to moderate workers' quit behavior and to increase the incen 83 This evidence includes press reports and our own case studies of firms. A well developed description is the ISITO (1998) case study of the Novokuibyshevsk Oil and Chemical Plant. 84 See, for example, Earle and Sabirianova (2000). Lehmann, Wadsworth, and Acquisti (1999) also study these regularities, although their analysis of regional concentration is at the oblast level, while this study analyzes more disaggregated rayons (districts). 85 Geographic mobility of labor in Russia is reduced by registration requirements (and large fees in cities such as Moscow), information problems, poorly functioning housing markets, and liquidity problems of workers. Mitchneck and Plane (1995) discuss internal migration in Russia. 86 See, for instance, Salop and Salop (1976) for a discussion of firm use of delayed payment contracts in order to reduce quits. The case of tilted earnings profiles to elicit effort is explored by Lazear (1990) and Akerlof and Katz (1989), among others. Pencavel (1972), Flinn (1986), and Topel and Ward (1992) analyze the role of the level of wages for worker quit behavior. 58 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY tives of firms to use wage arrears. The negative feedback mechanism of worker quit ting that would normally eliminate the practice is reduced, and wage arrears may spread rapidly and persist over longer periods of time than they would otherwise. Moreover, the incentives to use both arrears and forced substitutes are enhanced by the externalities conveyed from the strategies followed by other employers: If one employer increases arrears or forced substitutes, this is likely to reduce the quits from other employers. If workers are unsure they would be paid in cash and on time at a new job, then they are less likely to respond to a late or in kind payment by quitting to search or even to take up a new employment offer. Even firms that have good prospects and that want to expand their operations and hire additional workers may not be able to make credible promises of in cash, on time payment because of the volatility of the environment, the nonverifiability of their prospects, and their incen tives (understood by workers) to reduce costs by delaying payment or substituting lower valued goods once the worker has signed on. Migration to a region where employers typically do pay in cash and on time is both very costly and full of uncer tainties. Nonemployment may become more attractive for some workers, but it is not an option for everyone. Thus, the consequences of paying workers late or forcing them to accept in kind substitutes in order to ease financial problems or to cut labor costs are likely to be quite different when most other firms are doing so than when no others do, particu larly those operating in the same local labor market. This interaction may lead these practices to be self sustaining, so that they persist even if their original cause is removed. Measurement of wage arrears and in kind substitutes is difficult. Official information on wage arrears in Russia is limited to aggregate time series of the reported cumulative overdue wage debts in certain sectors of the economy, while there appears to be no official information on forced substitutes. Until 1996, only three series (for the aggregate industry, construction, and agriculture sectors) for arrears are available, while afterward the set of sectors is expanded. The official time series for the real stock of arrears is shown in figure II.2. The inconsistency of sectoral reporting makes it difficult to draw firm conclusions, but it does seem that the real stock of arrears peaked in August September 1998, and then declined steadily until the spring of 2000. Several plausible explanations exist for the decline. First, the drastic devaluation of wage debts because of the steep inflation in fall 1998 made it much easier for firms to pay off debts. Second, the manufacturing sector, where arrears are to a significant degree concentrated (as shown below), received a big boost from the currency devaluation, which made exports more com petitive and imports less so. Third, there may have been some change in the policy regime, as new legal penalties were put in place in early 1999, and the Government affirmed the reduction of arrears as an important policy priority. Despite the large fall, however, it is notable that wage arrears remain substantial in the Russian economy, and they have even started to increase again since spring 2000, including in the public sector. One interpretation of the increase is that wage arrears continue to follow a political business cycle, because of nominal wage increases and pushes to pay prior to elections and subsequent inability to meet those promises (see CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 59 Treisman and Gimpelson forthcoming). It appears that the Russian economy is still plagued by large scale arrears, even if smaller than before, and that the environment remains vulnerable to a recurrence of a massive outbreak, if the macroeconomic sit uation changes. From the official data, however, one can learn rather little about questions of interest, such as which groups of workers are most affected, whether the incidence tends to remain the same or has changed, and about the extent to which changes in the aggregate involve changes in the number of affected workers or a worsened (or improved) condition for those previously affected. The aggregate data also do not permit, of course, any analysis of the association of wage arrears with other variables. The analysis in this section therefore relies mostly on microdata, from household and firm surveys. Defining a measure of wage arrears also faces several problems. In theory, one would like to measure the worker's present discounted loss from wage delays, taking into account the timing of past payments and the risk premium associated with the uncertainty of the timing (and probability) of future payment. Such a measure would require detailed information on the wage payment history of each worker and on his/her discount rate and expectations concerning future payment. In practice, pay ments of wages and repayments of back wages tend to be highly irregular, creating high volatility in the actually paid monthly wage relative to the promised or contrac tual wage.87 Furthermore, detailed records on the entire histories of wage payments and repayments are hardly kept or reported. The prevalent practice of accounting for arrears -- both in individual firm balance sheets and in official Russian statistics -- is to sum the cumulative debt of the firm to a worker, without regard to the timing of when the debts were incurred. Workers tend to think of their arrears as this stock expressed as the number of overdue monthly salaries that they are owed. Associated with this concept of the level of arrears is the standard practice of paying debts in the order in which they are incurred. For exam ple, consider a worker with three months of arrears in October some year. If he/she is paid one monthly salary at the end of October, this payment is treated as the July wage, and arrears are considered to remain unchanged at three months. If he/she instead receives 2.5 monthly salaries at the end of October, this is considered pay ment for July, August, and half of September, and arrears decline to 1.5 months. If he/she receives nothing, then arrears are recorded as rising to four months.88 Incidence and persistence of wage arrears and in kind benefits. Wage arrears. With this background, table AII.8 shows the incidence, magnitude and per sistence of wage arrears for the years of the second wave of the RLMS: 1994, 1995, 1996, 1998, and 2000. The proportion of workers with arrears and the average num 87 Thus, the RLMS variable corresponding to the reported wage received in the previous month, which has been used by many researchers as though it were a standard wage measure, need bear little relationship to the contractual wage or to the average wage received over some longer period. 88 One reason for this practice is that firms pay no interest or penalties on wage arrears, nor are they indexed. Thus all that matters is the total debt. 60 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ber of overdue salaries increased steadily up to 1998, with a particularly sharp jump in 1996 compared with 1995. This is followed by an equally sharp collapse in 2000. Remarkably, however, the expected value of the magnitude of arrears (measured as the number of overdue monthly salaries) conditional on having some arrears hardly fell from 1998 to 2000. Thus, the data imply that essentially the entire decline in the average arrears is the result of a lowered incidence, while the conditional mean changed little. The data also show strong persistence of arrears across years, with the conditional probability reaching nearly 0.9 for workers reporting arrears in the previous two interviews, and the conditional expectation of the amount of arrears reaching close to 10 monthly salaries in 1998 for those with arrears exceeding 6 months in 1996. The strength of the persistence effect was diminished in 2000, as many workers were repaid back wages. Similar figures, but from a firm survey covering the years 1991­98, are shown in table AII.9. The growth of arrears from a negligible to a substantial level is clearly visi ble. Unfortunately, no data for 1999 or 2000 have yet been collected by either the RLMS or the firm survey, so it is not possible to verify the aggregate trends using microdata. Earle and Sabirianova (forthcoming) report the heterogeneity in the incidence and magnitude of wage arrears for a set of firm and employee characteristics in the RLMS from autumn 1996. The average incidence89 and magnitude90 of arrears were both much higher in rural than urban areas, and there was substantial variation across localities. While the regional variation exists across six major regions of Russia, it is still higher at a more disaggregated level, as some rayons have very low arrears and some have very high, nearly universal arrears. The results for the City of Moscow, where 28.6 percent of employees were owed money and the mean magnitude was 0.6 months in 1996, mostly reflects arrears of the Federal Government. Variation across industries also was reported to be large, with the highest rate in agriculture and in some industrial sectors (shown under "selected industries"), par ticularly machine building and defense ("Military Complex"), as well as in services financed through the state budget (education and health). In a new and rapidly developing sector such as banking, however, arrears were very small at this time. Arrears vary strongly with size, showing a much lower incidence and average magni tude in firms with fewer than 50 employees. Arrears also vary across different forms of ownership. The data show the highest incidence and magnitude of arrears in the agricultural collectives,91 followed by mixed and state owned firms, while they are lowest -- although still not negligible -- in domestic private and foreign firms. Arrears also vary by the employer's founding date, defined on the basis of a question posed to worker 89 (mean of ARRDUM). 90 (mean of ARRMOS). 91 High arrears in agricultural cooperatives may reflect limited opportunities of coopera tive members in rural areas, and limitations to mobility, although as noted above mobility itself maybe impeded by wage arrears. CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 61 Table II.4. Incidence and Magnitude of Forced In Kind Substitutes for Wages (Percentage) Years Percentage Firms with in kind substitutes for wages of firms with in kind substitutes Share of wage bill Share of workers (N = 162) paid in kind paid in kind 1991 3.1 36.2 44.0 1992 3.7 39.3 52.5 1993 4.9 36.3 47.0 1994 9.9 26.4 50.0 1995 16.7 27.1 59.3 1996 22.8 30.2 64.2 1997 25.9 29.8 64.1 1998 27.2 37.0 70.6 Note: Sample is consistent across years (N = 162). Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (1999). respondents in the RLMS. Employees of firms founded before the beginning of per estroika (1988) were much more likely to have arrears in 1996 than those founded subsequently, although the problem was significant even among the latter, some times called de novo firms. In fact, the data show that some of the de novos were themselves state owned (usually by local governments). Even among genuine, pri vately owned start ups, however, it is not surprising to find some arrears, since the start up sector tends to be highly volatile in any economy. The difference in Russia is that it is the old, established sectors and government agencies where wage arrears are the greatest problem. Concerning personal characteristics, men tended to have a slightly higher proba bility and magnitude of arrears than do women. Arrears were lowest in the youngest (under 30) age group, perhaps because of the relatively low mobility costs of this group. Arrears are generally negatively related to the level of schooling and positively related to job tenure. Even new employees, those with tenure less than one year, have a 50 percent rate of arrears, however.92 With respect to ownership by the employee respondent, results are based on RLMS questions on share ownership in the employer and on the percentage of com pany shares owned. Because of the different nature of ownership in the agricultural 92 The implied arrears tenure relationship (also obtained in Lehmann, Wadsworth, and Acquisti 1999) could be spurious if an employer has incurred arrears in the past but more recently has tended to pay wages on time. Unfortunately, the data (particularly on the timing of arrears) are insufficient to permit an assessment of the quantitative importance of this possibility. 62 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY cooperatives and transformed cooperatives, these are distinguished from other own ership types when large stakes are involved.93 As rather few employees report more than 1 percent ownership, however, all responses of 1 percent or greater within these groups have been pooled together. In nonagricultural firms, the arrears ownership relationship appears to be non monotonic, with the highest incidence and magni tude among small shareholders (those owning less than 1 percent) and the lowest among larger shareholders (1 percent or greater), with nonemployee owners in between. With respect to agricultural firms, however, the large shareholders show higher values for both incidence and magnitude of arrears. Concerning variation across occupations, employees of the armed forces experi ence almost universal arrears. The armed forces employees in the sample are not ordinary enlisted soldiers and conscripts but rather service workers and officers resid ing off the military bases, because the RLMS sample did not include bases. Among civilian employees, craft workers and operators and assemblers tend to experience the highest rates, while managers have the lowest, although the rate is high even for this occupation. A final issue concerning arrears is whether the legal system functions to force firms to pay, and if not, why not. As Table A.10 shows (and Chapter III), few firms have had to pay penalties, and those penalties that have been assessed are tiny -- certainly relative to the stock of wage arrears in the firm. Evidently, the Russian legal system functions poorly in enforcing wage contracts, an issue that is taken up in the next chapter. In kind benefits. Concerning forced in kind substitutes for wages, no official esti mates are available. Thus, the analysis draws exclusively on micro data: the RLMS and firm survey. Table AII.11shows the results from analysis of RLMS concerning the inci dence of such forced substitutes. The proportion of workers reporting such forced substitutes during the previous year was 8 to 9 percent in 1994­95, 12 percent in 1996, 15 percent in 1998, and 9 percent in 2000. Similar to wage arrears, in kind sub stitutes show strong persistence: apparently, it tends to be the same group of people who are affected, year after year. Table II.4 uses the firm survey to address the magnitude as well as the incidence of forced substitutes, from the firm's perspective. The fraction of firms reporting the use of the practice rose steadily from 3.1 percent in 1991 to 27.2 percent in 1998. Among firms using the practice, the share of the wage bill paid through this mechanism has fluctuated but does not show a clear trend, but the share of workers paid in kind has steadily increased. Apparently, the practice has become much more regularized over this period. Regional variation in the use of in kind substitutes is similar to that for wage arrears, but the variation across industries shows a much stronger concentration in agriculture. Firms with mixed ownership (including agricultural cooperatives) have 93 A possible explanation might be the nature of the cooperative transformation process in agriculture, which generally resulted in equal division of ownership and closed legal forms, unlike other sectors where managers generally acquired disproportionate stakes and the legal form was usually open (a legal requirement in the State Privatization Program). CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 63 the highest rate, and actually the lowest rate is found in the state sector probably because many state owned organizations, such as schools and hospitals, have little they can offer workers in lieu of their salaries. In contrast to wage arrears, in kind substitutes vary little with tenure, although they do vary strongly across occupations. Like arrears, the highest rate occurs for cooperative owners, while employee owners of joint stock companies are most likely to have in kind substitutes if their ownership is small (less than 1 percent), again belying the sometimes heard claim that these practices represent voluntary recontracting. A possible evaluation of the Russian institutions that lead to practices of wage arrears and in kind benefits is that they demonstrate "flexibility," unhampered by legal and institutional rigidities. Layard and Richter (1995) and OECD (1997), for instance, have praised the use of wage arrears as a way of achieving wage flexibility and low unemployment in Russia. Leaving aside the question of the social desirability of wage flexibility, however, it seems dubious that arrears and contractual failure are socially efficient mechanisms for bringing about a given effective change in the real wage. As a first welfare consideration, it should be noted that the incidence of the con tractual violations is unevenly spread across regions and households, as shown above, and thus their social consequences tend to be concentrated in certain groups. Second, wage arrears and the other practices reduce utility more than equivalent wage cuts, because of the associated uncertainty concerning the timing and probability of eventual payment. Uncertainty in wage payments may also reduce worker effort and reduce investment in training, reducing labor productiv ity. Lack of contract enforcement therefore, compromises both consumption and production efficiency (Rashid and Townsend 1994). Third, as discussed above, arrears may sometimes actually impede mobility, particularly where arrears are widespread in the local labor market; these areas are also likely to be those where mobility -- geographic and industrial -- is most needed. Thus, wage arrears may actually retard the reallocation of labor that is critical to the transition process. Real flexibility may be reduced. A major consideration in a normative evaluation of these practices in Russia, how ever, is the fact that labor contracts are the most important contracts for most indi viduals. When those contracts are not respected and enforced, it reduces confidence in other labor and non labor contracts into which the individual might enter. In short, wage arrears may undermine the development of contract enforcement and rule of law. We take up the issue of contract enforcement in the next chapter. C. Wage Inequality and Poverty Wage inequality. The decline in real wages was accompanied by growing wage inequality (Commander, Tolstopiatenko, and Yemtsov 1999) early in the transition. While increased dispersion of wages is an inevitable process of the transition to mar ket, the level of wage inequality in Russia is very high by CEE standards, and is closer to levels found in Latin America (World Bank 2000a) Most worrisome, the recent 64 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY increase in real wages has not been accompanied by declining wage inequality. Rather, wage inequality has increased between 1998 and 2000. Wage dispersion in Russia and other former Soviet Union (FSU) countries was already much higher than in the centrally planned economies of Europe at the onset of the transition (Atkinson and Micklewright 1992) Wage data based on enterprise surveys show that prior to the transition, the Gini coefficient for earnings for CEE countries ranged from 0.198 in Czechoslovakia to 0.268 in Hungary, while it was 0.273 in Russia and 0.300 in Georgia. For both CEE and FSU countries, the dramatic increase in earnings inequality was concentrated over a very short period, in many cases in the first few years of the tran sition. As such, the increase in inequality has been unprecedented. This increased the wage inequality in Russia and other FSU countries to upwards of 0.500. By contrast, in most European transition economies, the Gini is around 0.3, a value not untypical for developed market economies. While some CEE countries, such as Lithuania, Latvia, and the Czech Republic, have reduced these inequalities, these examples cannot be found among CIS countries (Rashid and Rutkowski 2001). The growing wage inequality in transition economies is evident in the high inci dence of both low paying jobs and top paying jobs, while the number of middle pay ing jobs -- preponderant before the transition -- has decreased (figures AII.1 and AII.2). This polarization is more pronounced in Russia and the other FSU countries than in the CEE countries. In Russia as much as 34 percent of all jobs are low paying jobs (earnings lower than two thirds of the median), and 31 percent are relatively well paying jobs (earnings higher than 1.5 times the median). The incidence of low pay in CEE economies is around 20 percent while it is even lower -- less than 20 per cent -- in OECD countries. The earnings gap between low paid workers and median paid workers has widened substantially during the transition. Before the transition, in most countries, a low paid worker was earning some 60 percent of the median. This share has declined to around 30 percent in Russia, but is much higher, around 50 percent of the median, in CEE transition countries. Thus, low paid workers are among those who suffered most in the wake of market oriented reforms. Why is wage inequality so high in Russia? In CEE countries, wage inequality largely represents the emergence of market factors, such as differences in the level of educa tion. In Russia, education explains only a small part of wage inequality (table AII.12). Most of the factors explaining the large dispersion in wages come from outside mar ket based parameters. Regional differences in wages stemming from wage arrears may be the main determinant of wage inequality (Lehmann, Wadsworth, and Acquisti 1999) Understanding wage determination in Russia therefore requires an evaluation of labor market institutions, a topic taken up in the next chapter. The growing inequality in wages has contributed to increasing income inequality in Russia (Commander, Tolstopiatenko, and Yemtsov 1999). However, the effect of wage inequality on total income inequality was dampened by the decline of the wage share in income noted above. Rather, the main factor contributing to increased income inequality was the rising entrepreneurial household income share. This com ponent of income tends to be more unequally distributed than wages, even in devel CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 65 Figure II.3. Lorenz Curve for Wages, 1998 2000 100 90 ) 80 percent( 70 wages 60 of 50 share 40 30 Cumulative 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 Cumulative share of population (percent) 1998 2000 oped market countries (such as the United States). This distribution is likely to be more unequal in Russia where entrepreneurial access to credit and other inputs is more subject to personal and political connections (Ovtcharova 2000). As noted above, real wage growth between 1998 and 2000 has been accompanied by an increase in wage inequality in Russia. The Gini index for real wages increased from 43.9 percent in 1998 to 46.4 percent in 2000. The growing inequality is illus trated in figure II.3 using a Lorenz curve. The Lorenz curve for 2000 lies further away from the 45 degree line than in 1998, indicating an increased inequality in wages. Growing wage dispersion is the result of a higher proportionate increase in real wages for workers at the higher versus lower end of the wage distribution (Figure AII.4). Poverty in the workforce. The recent increase in economic growth, which has increased real wages and (slightly) increased wage inequality, has contributed to a decline in poverty in Russia. This result is consistent with reductions in income poverty during the same period, as reported by both Goskomstat and RLMS data. The increase in wage inequality derives from a larger increase in real wages of high versus low wage workers (figure AII.4). What are the characteristics of workers realizing real wage gains? As noted above (table II.1), real wage gains from recent economic growth have been skewed to young, well educated, urban, private sector workers. Older workers, with low levels of education, who work in the rural agriculture sector have hardly realized any real wage gains at all. And, again as noted above (table AII.1), 66 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.5. Poverty Rates by Socioeconomic Groups (Percentage) Goskomstat 2000 RLMS, 1998 Individuals Poor* Households headed by: Poor* Old age pensioners 24.0 pensioners 33.9 Invalid pensions 42.3 Employed without wage arrears 34.9 Survivor pensions 50.8 Employed with wage arrears 51.4 Workers at enterprises 24.3 Unemployed and not receiving 60.9 benefits Working part time 53.9 Unemployed and receiving 80.0 benefits Working for others 34.9 Not in the labor force (not pen 41.7 sioner) Receiving unemployment benefits 63.7 Not unemployed 41.3 Unemployed for a year or more 63.9 * At minimum subsistence Source: RLMS (1998) and Ovtcharova (2000). despite wage gains for particular groups between 1998 and 2000, real wages in 2000 remained lower than pre 1998 levels. Which workers receive low wages? In 2000, a higher proportion of rural (vs. urban); younger (less than 35 years) and female (vs. male) workers received low wages (in the first quintile). A higher proportion of agriculture (vs. non agriculture), small (vs. medium and large); and state (vs. mixed or private) workers received low wages. Which groups among the labor force have the highest poverty rates? Some evi dence on consumption poverty is available from the 1998 RLMS data. In that year, the poverty rate was 58 percent among the unemployed and 64 percent among the long term unemployed (compared with 46 percent for the labor force as a whole). Poverty rates among employed workers were lower, at 44 percent.94 These comparisons are bleaker for unemployed heads of household with children. These individuals faced a poverty rate of 67 percent, compared with 52 percent among employed household heads with children (table II.5/Table AII.14)). But poverty was not only greater among the unemployed, it was also higher in households with wage arrears. The poverty rate in households with children whose heads had wage arrears was 60 percent in 1999, compared with a poverty rate of 40 percent for households with children whose heads were receiving wages. High poverty rates among the unemployed (versus wage earners) are also confirmed by Goskomstat 2000 data, which indicate that individu als receiving unemployment benefits are among the poorest of all socioeconomic groups. These findings suggest that adverse developments in the labor market during the 1990s led to high rates of poverty among Russian households, particularly the unemployed and those with wage arrears. 94 Among all individuals in the labor market, 28 percent were heads of families with chil dren. These comparisons are based on analysis conducted with QIII 1998 data from the RLMS. CHAPTER II. UNDERSTANDING WAGES: STRUCTURE, UNCERTAINTY, AND ... 67 In summary, these results indicate that while economic growth has had a modest positive impact on real wages between 1998 and 2000, it has widened already high wage inequality prevailing in Russia. Workers who have realized wage gains are high wage workers, the young, urban, private sector workers. Real wages have hardly increased among low wage workers, with lower levels of education, who work in rural areas, in agriculture, and for the state. Moreover, for all workers, real wages in 2000 remained lower than pre 1998 levels. The chapter also finds that while poverty has declined with recent economic growth, poverty rates likely remain high among workers with wage arrears and the unemployed. Summary and Conclusions The chapter has discussed the level, trends, and determinants of wages and wage inequality in Russia. The main conclusions are as follows: The decline in real wages in absolute terms (and relative to output declines) was greater in Russia than in other CEE countries. Unlike these countries, wages were the main mode of labor market adjustment (versus employment). The decline in wages was also accompanied by the growth of wage arrears and a large increase in wage inequality.95 Real wages have responded to recent economic growth, although their response lagged with respect to output (as in the case of Hungary and Poland). Despite the recent turnaround, the level of real wage in Russia remains very low by CEE standards. Recent real wage gains were concentrated among high versus. low wage workers, increasing already high levels of wage inequality. Wage gains have largely benefited young, highly educated workers working in the urban private sector (versus older, less educated, rural, agricultural sector workers). Recent economic growth has also reduced the incidence of wage arrears and in kind substitutes, but they have not disappeared. The average amount of wage arrears for workers who continue to receive them has not changed. The use of inappropriate fringe benefits has also declined (and was declining in any case over the transition), but remains significant among large firms. The use of wage arrears and in kind substitutes as a form of wage adjustment is quite separate from wage cuts -- and formal wage flexibility. Unlike wage cuts, wage arrears and in kind substitutes reflect contract violation rather than contract renego tiation. Moreover, they impose considerable income uncertainty on workers and fur ther lower worker welfare. Why do they persist? Wage arrears and in kind substitutes tilt the age earning profile for workers, making it costlier for workers to leave their firms. Limited enforcement of wage contracts, the nature of local markets (where employer contract violation behavior can be matched by other employers), and low regional mobility are factors that induce employers and workers to persist in these 95 The contribution of wage inequality to total income inequality has been dampened by the decline in its income share. High income inequality is largely explained by the inequality in the growing share of self employment (in total household income). 68 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY practices. The concentration among particular groups (less educated, more experi enced), regions (rural), and industries (agricultural cooperatives) may reflect the lower bargaining power or job opportunities of these individuals versus other groups (younger, nonagricultural, private sector) of workers. Recent growth has not significantly increased the returns to education. However, it is important to note that wages started to increasingly reflect returns to education during the transition, as a result of market liberalization. The returns to vocational education have fluctuated over the transition period. The returns to labor market experience and job tenure declined in Russia, as in most transition countries, indicat ing low rates of returns to work experience from the socialist era. As in other transi tion countries, the returns to employment in the private sector are quite high relative to the private domestic sector and, particularly, the state. However, there is a gender gap. Women earn less than men, after controlling for socioeconomic and occupa tional characteristics. Although recent economic growth has reduced poverty rates, they still remain very high relative to CEE countries. The fall in earnings from formal employment con tributed to growing poverty among labor market participants or the working poor. Poverty rates were highest (relative to the national level) among individuals with wage arrears and those who are unemployed (particularly those in receipt of benefits). These results confirm the conclusions of the previous chapter that economic growth will be the main vehicle for increasing employment, labor productivity and, hence, real wages in Russia. However, the increasing payoff to education in Russia suggests that investment in education will also be important for sustained growth in labor productivity, and that particular attention will have to be paid to improving the relevance of vocational education for the labor market. While the large decline of real wages in Russia is mainly the result of the fall in aggregate demand and high rate of inflation realized over the transition, the prolonged decline in the level and uncer tainty of real wages, and the growth in their inequality, may also be strongly linked to the weak regulatory structure of the labor market. Less regulation of the market helps improve labor market outcomes, but complete lack of regulation -- as in Russia -- appears to have had an adverse impact on welfare, and can also have adverse effi ciency consequences. The following chapter therefore looks closely at labor market institutions in Russia, by law and by practice, and draws on Russian and international experience to understand their impact on the functioning of Russian labor markets. Chapter III Labor Market Regulation The previous chapter suggests that labor market regulation may have played an important role in determining labor market outcomes in Russia. In this chapter we discuss four main aspects of labor market regulation. These are: (a) the nature of the employment "contract," including the rules and norms governing hiring, contracting, and dismissals; (b) the wage determination process; (c) the institutions determining worker organization and collective bargaining; and (d) institutions for enforcement and dispute resolution. Simply evaluating the law and institutional setup is not ade quate for understanding the Russian labor market, as current practice often diverges markedly from formal arrangements. For this reason, for each of the above areas, this chapter discusses the regulatory structure, current practice, and international evi dence on the labor market impact of regulations. Policy options in each area are pre sented at the end of this chapter. At the time this report was being finalized, the State Duma passed a new Labor Code, replacing the previous 1971 Code (with amendments). The new legislation is relevant for each of the four aspects of regulation considered in this chapter. As the chapter was written prior to the passage of the law, the analysis reported here is based on the prior regulatory framework. What we have done is to supplement our core analysis in each area by noting how the new Labor Code might change current prac tice. In our view, the new Code does take a modest step forward in making labor mar ket legislation more consistent with a market economy, especially with respect to contracting and terminations. However, it does not fundamentally alter many fea tures of the old regulatory regime and therefore the analysis of this chapter. A. Background Before turning to the four specific aspects of regulation addressed in this chapter, this section briefly describes the overall regulatory system, the robustness of evidence on recent practice, and the applicability of international evidence. Regulatory framework. In reviewing the regulatory framework, it is important to note that a number of laws, resolutions, and decrees come into play. The most important is the Labor Code, which establishes the general framework for labor con tracts, and outlines guarantees and privileges and the role of trade unions. The Code covers all employees and all forms of organizational ownership. As we have already noted, a new Labor Code has just been passed, replacing the 1971 version (with amendments). This new Code is the culmination of a number of years of difficult 70 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY debate. In addition to this Code, contracts are further regulated by Government reso lutions, including "On Confirmation of Recommendations on Conclusion of Labor Contract in Written Form" and "On Standard Form of Labor Contract" (1993). Between 1991 and 1997, new industrial relations laws were approved, including the Law on Collective Bargaining Procedures (1992), the Law on the Order of Resolution of Collective Labor Disputes (1995), and the Law on Trade Unions, Their Rights and Guarantees of Their Operation (1996). The important development in the wage reg ulation area was the abolition of the Unified Tariff of Wages and Salaries in 1992.96 Despite all of these initiatives, by no means have Russian labor laws and regula tions fully adjusted to the realities of the market economy, even after a decade of tran sition. The main goal of labor market regulation is to protect employees from dis crimination and to ensure that all workers are employed according to their qualifica tions (Denisova, Friebel, and Sadovnikova 1998a). There are strong rules governing most aspects of the employment relationship within the enterprise; regulations still provide for trade unions to assume certain functions that are viewed as managerial prerogatives in most market economies; and the law enshrines extensive guarantees and privileges for particular groups that are far beyond what is found in Western labor laws. New legislation, including the 2001 Labor Code, has only slightly moderated these facts. Limited consensus exists regarding the role of Government and the private sector in providing employment and social protection. This debate reflects a larger political debate about the role of Government in a modern market economy. Many pro mar ket voices argue that this is a serious problem and that major reforms still need to be put in place to curtail guarantees and privileges and to reduce direct Government intervention in the labor market. On the other hand, there are still influential voices arguing for the state to maintain a strong regulatory hand to protect workers from economic restructuring and managerial arbitrariness.97 This clash of views exists in all debates on labor policy, and has seriously hampered most reform initiatives includ ing, until now, a new Labor Code. Nonetheless, as we have already noted, the Labor Code still reflects a view of the labor market that is more appropriate for a planned economy than a market economy. Recent practice. Understanding regulatory and institutional aspects requires going beyond what is on paper and looking at employment practices on the ground. 96 The other major piece of legislation is the Employment Law, which defines the state role in the labor market through the Employment Service, active labor programs, and unemploy ment benefits. This Law is discussed in Chapter 4 of this report. The Government has committed to introducing a new Employment Law to reflect announced changes in policy in these areas. 97 This is a controversial question with two very different perspectives what Freeman (1993) has called the "institutionalist" and "distortionist" views. The "institutionalist" view sees job security arrangements, minimum wages, and collective bargaining as providing important social protection for workers, as instruments for encouraging productivity growth (through training and the accumulation of firm specific skills), and as means of moderating the effects of downswings in aggregate demand. The "distortionist" perspective emphasizes the advantage of market processes and is concerned that these institutional forms of regulation impede adjust ments to economic shocks, discourage hiring, and favor "insiders" (that is, regular workers). LABOR MARKET REGULATION 71 For this reason, the chapter tries to piece together reality by complementing national data with evidence collected from various smaller independent surveys of establish ments and workers. While smaller surveys have limitations (see below), they offer the only real channel for understanding the situation. The results of most of these surveys have been published elsewhere.98 We also produce new and more recent data based on three surveys conducted by the Russian Academy of Sciences.99 The surveys we draw on do flesh out the reality of employment relations in Russia. However, these data are sometimes qualitative and based on samples that are relatively small and uneven in their coverage. This limits the capacity to make reliable estimates for sub sectors of the enterprise population (for example, regional, industrial, form of own ership, and so on). International evidence. In reviewing international evidence on the impacts on employment outcomes, the chapter relies primarily on countries in the OECD region where most of the analysis has taken place. We also take advantage of a recent study that compares advanced CEE reformers with labor market institutions in OECD countries. The chapter describes the range of approaches used and what is known about their labor market implications.100 Interpreting the international evidence requires a few qualifications. First, national contexts (history, culture, institutions) vary a great deal, and labor market impacts of a given law or practice in one country may be quite differ ent from another. Second, the actual arrangements for a specific aspect of the regula tory regime can be very difficult to capture. For example, simply looking at what is pro vided in the legislation may provide an incomplete or inaccurate picture if enforcement is weak or if nonformal practices in reality take precedence.101 Third, the current prac tices in Russia often are "outliers" when considered within the parameters of CEE and OECD countries -- consequently, it is not always clear how their experience would apply in Russia. Nonetheless, keeping these caveats in mind, the international evidence can offer important insights on reform options to encourage employment and earnings growth in the market economy. While many of the current preoccupations in Russian labor policy may seem far removed from practice in developed countries, the longer term reform strategy should be designed with that practice in mind. 98 For a description of the published surveys we have relied on, see Clarke (1999). 99 Details on the methodology for the new surveys we draw on are provided in Tchetvern ina (2000).The first is the RLFS. This has been carried out on a longitudinal basis a number of times beginning in 1994, with the latest wave in 2000. The longitudinal nature of this survey offers a unique view of trends; however, the panel now has a small sample size (n = 85) because of attrition and the composition of panel firms is not universal (for example, it excludes new firms in the private sector; in nonmanufacturing industries, and so on). The 2000 RLFS has been replenished by a new group of participating enterprises making the complete sample 308 enter prises. The second survey of enterprises was conducted in late 1999 in five regions. It covers 278 enterprises in a wide range of sectors. Sample selection, however, was not based on a random selection methodology. The strength of this survey is that interviews in each establishment cov ered managers, union leaders (in the 180 enterprises with unions), and employees in each work place (n= 2,213). The third source of data is surveys of trade union activity in enterprises in 1995 and 1998, again with interviews of managers, union leaders, and employees. 100 This evidence is largely drawn from Betcherman, Luinstra, and Ogawa (2001). 101 See OECD (1999a) and Bertola, Boeri, and Cazes (2000) for discussions of these problems. 72 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY B. Labor Contracts Regulatory framework (pre 2001 Labor Code). The legal framework in Russia has been geared heavily toward formal, permanent, open ended contracts (table AIII.1). Contracts can be signed for an indefinite period or for a fixed term (not more than five years). There are numerous restrictions on the use of fixed term contracts. However, an employer must have a specific reason for offering a fixed term contract: if the job will be for a finite period; if there are particular working conditions (for example, working in extreme conditions); or if this is the preference of the employee. Although temporary agency work does take place, this form of employment is not covered in the Labor Code. Since 1992, there has been a statutory requirement that all new individual labor contracts be in written form. The Labor Code also places restric tions on the use of overtime and shift work. These restrictions on fixed term con tracts, use of temporary agencies, and overtime and shift work are excessive by the standards of most OECD countries. Probationary periods upon hiring are permitted. This is a common way in which managers can assess the suitability of new employees without entering into the full obligations of a standard employment relationship. However, there has been a three month maximum for probationary periods unless the trade union agrees to a period of up to six months. This three month maximum is relatively short for screening pur poses.102 Furthermore, employers cannot use probationary periods for certain classes of workers, including youths (under 18 years of age), persons graduating from educa tional institutions, and disabled workers. Especially in the case of young workers, who typically will not have much of an employment record to guide prospective employ ers, this ban on the use of probationary periods seems inappropriate. The Labor Code restricts managerial discretion in the deployment of labor more than is the standard in Western countries. Transferring employees to other work within the enterprise requires the consent of the worker and two months' notice. There are restrictions on the temporary transfer of workers as well. These restrictions limit internal (functional) flexibility in the enterprise. As a general rule, international experience suggests that the law should allow employers to place workers where they will be most productive. Of course, this principle should not preclude unions and managers from voluntarily negotiating collective agreements that guide labor deployment. There are also special protections limiting the work that women can do. All women are prohibited from performing arduous work; employers should accommo date pregnant workers and mothers with children under 3 years of age by reducing norms of output or servicing or by transferring them to less demanding jobs (while retaining previous salary). Certain restrictions exist in terms of assignments that can 102 On the other hand, it should be noted that the use of probationary periods can be abused. Workers on probation may receive lower wages, less employment protection, and fewer benefits. Anecdotal evidence suggests that some employers terminate workers at the end of their probationary period and replace them with new probationary employees. An effective appeal procedure is necessary to protect against this. LABOR MARKET REGULATION 73 Table III.1. Forms of Labor Contracts by Sector, ISITO Survey, April 1998 State Budgetary Privatized De novo Average sector entities firms firms Distribution, percent Open ended employment without contract 77 73 72 34 67 Open ended contract or agreement 14 14 18 29 18 Fixed term contract between 1 and 5 years 5 10 4 6 6 Fixed term contract less than 1 year 3 2 4 9 4 Labor contract to perform certain work 1 1 5 2 Oral agreement 1 1 1 18 4 Total 100 100 100 100 100 Source: Clarke (1999, table 5.1). be given to pregnant women or women with small children (for example, no over time, night work, business trips, without consent). While regulations such as these may be motivated by social protection objectives, they do not reflect modern realities and they may harm employment and career prospects for these workers. The new Labor Code introduces some modest changes in the regulatory frame work governing labor contracting, but in many areas it does not alter the status quo. It does increase the flexibility to hire workers on fixed term contracts, especially in firms with fewer than 50 employees. The new Code also extends the use of probation in positive ways. However, the Code does not make any marked improvements in either the deployment of labor or in terms of the protections of certain categories of workers, including women. Future reforms will be necessary to provide employers with the similar scope to deploy workers that their Western counterparts have. Also, the protection of female employees cannot be provided to the extent it is in the Code, without making women uncompetitive in the labor market. Other policies, outside of employment protection legislation, are required to meet such social objectives. Recent practice. As discussed in the previous chapter, wage employment is the dominant form of employment in Russia. The legal framework creates a preference for permanent and full time contracts in wage employment. Nearly 70 percent of jobs in the state, privatized, and de novo private sector have permanent, open ended contracts. Table III.1 indicates that fixed term contracts are not the norm in any sec tor. However, the greater flexibility allowed employers by these contracts is increas ing their prevalence among particular occupational groups. This is especially true for managerial and professional categories (Tchetvernina 2000, Denisova, Friebel, and Sadovnikova (1998b). Denisova, Friebel, and Sadovnikova (1998a) cite evidence sug gesting that the majority of managers think that the practice of fixed term contracts should be expanded. Clarke (1999) has found that employees on fixed term con tracts are no more disadvantaged than other workers and, in some way, report higher levels of satisfaction. He concludes that Russian employers do not use these contracts to reduce the job security of lower grade workers. The legal framework also accounts 74 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY for the preference for full time positions, and discourages part time jobs. As noted in the previous chapter, the share of part time jobs is very low. Making flexible contract forms legal would therefore bring more of the work force into the formal economy While the law has prescribed written labor contracts since 1992 -- presumably to ensure enforcement -- most contracts are nonetheless oral. The common practice is still that a worker is hired on the basis of his or her application, and employment is confirmed by issuance of an order (prikaz) signed by an enterprise director. There fore, most contracting is informal, which raises concerns about its enforceability. The informality in contracting is especially true in the private sector but, more surpris ingly, it is also the case for many workers in other sectors of the economy. According to ISITO household data for April 1998, about 90 percent of workers employed in the state, budget, and privatized sectors were in permanent arrangements, but the vast majority did not have a written contract. In the de novo private sector, there was also a significant incidence of hiring solely on the basis of an oral agreement. Finally, infor mal hiring arrangements also dominate secondary employment where more than half (54 percent) of workers reported that they had been hired on the basis of an oral agreement only (Perova and Khakhulina 1997). In the full 2000 RLFS sample of 308 enterprises, which almost certainly underestimates the use of informal hiring, 15 per cent reported at least some informal hiring, with the share rising to almost one third for limited liability companies.103 Written contracts, where they do exist, are often inconsistent with the law. They frequently do not contain provisions on wages and working conditions that should be included by law. It is noteworthy that a significant minority of firms (about 10 per cent of privatized firms and 10 ­ 25 percent of de novo private sector firms) do not include provisions for overtime, regular wage payments, or paid leave in their con tracts (table III.2). These figures rise with respect to items that are not necessary to have in contracts in Russia but that would be standard in contracts in most developed countries.104 According to Clarke (1999), a substantial number of contracts do not explicitly define job duties: roughly one fifth of employees in state, budgetary, and privatized firms reported that their duties were only defined verbally, while this figure rose to 52 percent in the de novo private sector. Thus, whether oral or written, agree ments between workers and employers in Russia are generally outside the law. As a result, workers have virtually no recourse to any contract enforcement protection and employment based social programs. Contracts can be either written or oral in many OECD countries. Allowing oral contracting can help small employers and employees transact quickly. Drawing up written contracts can be time consuming for such arrangements. However, written contracts confer protection to large employers who would like to create uniformity in contract provisions across all their subsidiaries and branches. Employees also pre fer written contracts because they are easier to enforce; enforcement of an oral con 103 The sample is heavily weighted to state and privatized enterprises and away from de novo private sector firms where informality is most prevalent. 104 Tchetvernina. (2001) compares these incidences reported by employers with reports by employees. They find that the coverage is generally similar. LABOR MARKET REGULATION 75 Table III.2. Provisions Stipulated in Contracts (Permanent and Fixed term) by Property Form, Employer Reports (n = 278), 1999 State enterprises Privatized firms De novo private firms Perm. Fixed Perm. Fixed Perm. Fixed Percent of enterprises with provision in contract Compulsory contract provisions Wage (salary) 97.1 99.9 86.2 100.0 79.5 88.1 amount Conditions and 91.3 85.7 89.4 90.6 75.6 79.1 schedule of work Conditions guaranteed by law but not compulsory for contracts Regular wage pay 76.8 71.4 80.9 79.2 76.9 73.1 ments Paid leave 99.9 85.7 96.8 86.8 91.0 74.6 Paid sick leave 87.0 77.1 91.5 81.1 83.3 58.2 Payment for over 59.4 42.9 68.8 58.5 50.0 50.7 time hours Conditions of dis 69.6 60.0 83.0 84.9 66.7 76.1 missal Source: Tchetvernina. (2000). tract may be quite costly in practice. Whatever the form of contracting -- written or oral -- it is most important that the contract is legal, flexible, well understood by both parties, and enforceable. The importance of contract enforcement and its practice in Russia are taken up in a separate section later in this chapter. International evidence. In the area of contracting, researchers have focused on the contracting rules (permanent, fixed term contracts, and temporary agency work) for employing nonstandard workers. These generally include employees on fixed term contracts and temporary agency workers. Most research tests the impact of these contracting rules on employment and unemployment outcomes for workers. It is hypothesized that creating incentives or disincentives for certain types of con tracting will have impacts both on the level of employment and the composition. The main objective of contracting rules (such as the use of permanent contracts only) is to enhance job security by making dismissal costly to employers (in this case, by restricting hiring of nonpermanent employees). However, these rules can also have the unintended effect of raising costs of a worker to employers, and thereby creating hiring disincentives for employers. If these very strict regulations are enforced, they protect jobs for incumbent employees while limiting opportunities for the unem ployed and new entrants (for example, youths, women re entering). The regulation of fixed term and temporary agency employment generally per tains to (a) the types of work (for example, occupations) for which these forms of employment are legal, and (b) the maximum duration allowed. These rules are gen 76 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY erally stipulated in national or subnational labor codes. There is considerable varia tion across industrialized countries, with Anglo Saxon countries having the least restrictive arrangements and Southern European ones the most. Most advanced CEE countries fall between the two extremes (Riboud, Sanchez Paramo, and Silva Jau regui 2001). During the 1990s, there was generally a loosening of restrictions, with many countries broadening the use of fixed term contracts and temporary agency work (OECD 1999b). TableAIII.2 provides examples of current arrangements in selected OECD countries. The most extensive study of the labor market impacts of different contracting arrangements was carried out by the OECD (1999a), based on the experience of its member countries. According to this analysis, strict limitations on the use of fixed term and temporary agency contracting are associated with: ·Lower aggregate employment rates; ·Lower employment for women and young people; ·Higher levels of self employment (as a share of total employment); ·No impact on aggregate unemployment levels; and ·Lower flows into unemployment but longer average unemployment durations. This international evidence, then, suggests that the restrictions on flexible contracts (limitations on fixed term contracts and lack of legality of temporary work contracts) will -- all else equal -- reduce employment rates, especially for women and youth, and increase the difficulty for unemployed workers and new entrants to find jobs. In summary, excessive restrictions on flexible forms of contracting in law induce employers to engage in such contracting in the informal sector. The weak enforce ment structure allows this to happen. Easing of excessive restrictions on nonstandard contracting could help to bring some employment "out of the shadows." According to OECD data, this would particularly help more vulnerable groups, such as women and youths. The new Labor Code takes some promising steps in this direction. How ever, the Code does not make any marked improvements in either the deployment of labor or in terms of the protections of certain categories of workers, including women. Future reforms will be necessary to provide employers with the similar scope to deploy workers that their Western counterparts have. C. Dismissals and Terminations Regulatory framework (pre 2001 Labor Code). Enterprise restrictions on termi nations have been considerable. For example, employers can terminate labor con tracts for the following reasons: (1) the enterprise is liquidated or requires a reduction in personnel; (2) the employee is not suited to the job requirements; (3) the employee regularly does not fulfill job requirements; (4) idleness (including absenteeism); (5) failing to return to work after a period of leave; (6) previous employee rehired; (7) employee showing up at job under the influence of alcohol or drugs; and (8) employee stealing state or public property. LABOR MARKET REGULATION 77 Terminations falling under (1) staff reductions, (2) employee not suited to the job, and (5) failing to return to work require the consent of the trade union. The union is given 10 days to approve these types of dismissals. While advance notification of ter minations is an appropriate option by the standards of industrialized countries, the possibility of trade union veto is a substantial and unusual restriction on the right of the employer to adjust work force size. It may be more appropriate to use an appeals procedure where labor can question the legality of the layoff decision. Women, young people, and the disabled have additional guarantees on firing. On the employee side, workers on permanent contracts can terminate their con tracts with two weeks' advance notice. The employer is obliged by the end of this period to fulfill all contractual obligations (that is, to pay back wages). In the case of fixed term or one off contracts, on the other hand, employees have to complete the job during the period stipulated by the contract. They can renege on the contract only on the grounds of sickness or incapability of fulfilling the job, or some other strong reason. Denisova, Friebel, and Sadovnikova (1998b) argue that this makes fixed term contracts particularly unfavorable for workers. There are various regulations governing mass terminations. Advance notice requirements are three months for informing the trade union and two months for informing the affected workers. Trade unions must approve these layoffs, and they can ask employers to explore various alternative employment opportunities for affected workers. In response to an application by the trade union, local authorities have the right to delay mass layoffs for up to six months. There are also limitations on mass layoffs for certain state owned enterprises being privatized (Denisova, Friebel, and Sadovnikova 1998b). Severance requirements depend on the nature of the termination. If an employee is dismissed because of being unsuited for the work, because the previous employee is rehired, or where the contract is terminated because of violations of the regulations by the enterprise, an employee gets not less than two weeks' salary. In the case of staff reduction, including mass layoffs, an employee gets one months' wage and is contin ued to be paid when looking for a new job for up to three months if he/she has reg istered at the Employment Service and has not been placed in a job during that period. For employees working in the far North or those having the same status and for specific categories of employees (that is, those working in closed regions), the sev erance payment can be for up to six months. The new Labor Code appears to provide some significant improvements over the termination clauses in the old Code. In particular, it gives employers more flexibility to adjust to changing market conditions, which is necessary in a market economy. In supporting this flexibility, we emphasize the importance of effective unemployment benefit and employment programs to support unemployed workers in the labor mar ket (see next chapter). The new Code appropriately obliges the employer to provide the trade union with advance notice of termination.105 It is certainly appropriate to have such advance notice obligations in a market economy. It is critical, however, that 105 Maleva et al. (2001) also find that the new Labor Code imposes considerable costs on employers. 78 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY trade unions not be given veto power over these dismissals -- although the right to appeal dismissals that are alleged to be counter to the Code is appropriate. This trade union role -- which apparently is more limited than was the case under the previous Code -- may seem inappropriate given traditional practices in Russia, but it is neces sary to give employers the necessary flexibility in restructuring workplaces to meet the needs of the market economy. Hopefully, procedures for dismissals and appeals can be carried out quickly and efficiently within the framework provided for by the new Code. Current practice. As noted in the previous chapter, the composition of sepa rations in Russia is strikingly different than in OECD countries. Specifically, in Rus sia the share of quits is a much higher share of total separations than are redun dancies. However, the distinction between voluntary quits and separations is far less clear in Russia than in OECD countries. As conjectured widely in the literature, separations classified as "voluntary" and employee initiated are often induced by employers through prolonged administrative leaves, wage arrears, reduced hours, or other forms of deteriorating working conditions. Many affected workers, having no future with the firm and no sources of income, eventually are forced to quit. Once again the lack of enforcement mechanisms facilitates this form of labor adjustment. While this adjustment may keep enterprise employment levels (although not necessarily payroll costs) artificially high -- because separation is delayed -- these practices are inefficient for rationalizing the work force and impose large transactions costs for firms. One reason for the high share of voluntary quits may be high costs of layoffs (union consultations, for example). Workers have to be paid back wages; obtain sev erance, and are eligible for continued use of social services through their previous firms. A visit to Vladimir oblast and discussions with firm management revealed that the provision of social services to laid off workers was the main reason that the firm forced workers to voluntarily quit rather than lay them off. Other evidence on the reasons for layoffs comes from employee surveys. Table III.3 reports on reasons why workers separated from their previous job by sector. Almost one half of those leaving de novo private sector firms cited dissatisfaction with pay. On the other hand, personal reasons were cited more frequently in the state and privatized sectors. What role do severance requirements play? The Russian arrangements do depart from international standards in that severance amounts are not linked to seniority. For short tenure workers, the combination of notice and severance requirements may be overly generous, and obligations should be differentiated according to length of service to change this.106 However, as we will see below, with this one exception, severance obligations cannot be considered onerous compared to most other transi tion countries and even with OECD countries. Severance obligations, though, may still be a major problem in the sense that many firms that need to restructure may be cash strapped and have no funds for severance payments. This may also encourage 106 However, many redundant workers do have long tenure. LABOR MARKET REGULATION 79 Table III.3. Reasons for Leaving Previous Job, by Sector of Previous Employment. Kemerovo Oblast and Komi Republic, October 1997 Sector of employment State Privatized De novo private Total (n = 1384) (n = 71) (n = 195) (n = 1650) Distribution (percent) Closure of the enterprise 6 3 12 6 Dissatisfaction with social benefits 12 1 8 11 Dissatisfaction with pay 26 18 45 28 Fear for enterprise stability 3 3 6 3 Personal and family reasons 40 37 21 38 Made redundant 13 38 9 14 Total 100 100 100 100 Source: Goskomstat Labor Force Survey, cited in Clarke (1999, table 5.5). them to avoid outright redundancies and use the alternative cost containing strate gies, such as wage arrears or administrative leave, discussed above. However, not all firms face cash constraints. Case studies of enterprises have revealed that some man agers do offer financial incentives to voluntary quits (Denisova, Friebel, and Sadovnikova 1998b). The real constraints imposed by the labor laws on employment should not be overstated. Employer reluctance to lay off workers, especially in the budgetary and privatized sectors, seems to be significantly influenced by poor incentives and repu tational risks. Many of these firms -- and all the more so in one company towns -- face widespread expectations about protecting their employees, especially given the current reality of poor employment opportunities and weak social protection. In interviews with 70 managers who had surplus labor but nonetheless did not plan lay offs, 46 percent reported that the main reason was "to preserve the collective" (Tch etvernina 2000) International evidence. The key issue for policymakers concerns how difficult and costly it is for employers to terminate regular (that is, permanent) employees for economic reasons.107 The case for restricting employer termination rights is similar to that for limiting contracting for nonstandard employees, discussed earlier. By making dismissal for economic reasons more difficult or costly, these employment protec tion rules are intended to increase job security. However, the tradeoff again is that employers may be reluctant to hire workers if they face constraints in dismissing 107 "Regular" employees are meant to cover those with a permanent or indeterminate posi tion. It excludes fixed term or temporary workers. The discussion does not include dismissals for "noneconomic" reasons such as discrimination, union organizing, or job performance. 80 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY them for business reasons down the road. As a result, we can expect that strong job security rules that are enforced will lengthen tenure and reduce turnover but will have a negative effect on new hiring of regular employees. Restrictions on terminations can take various forms, including: (a) what is con sidered to be a justifiable reason for termination; (b) severance obligations; (c) advance notice requirements; and (d) necessary administrative procedures for laying off workers (including the role of trade unions). There may also be special require ments in the case of mass layoffs. These restrictions are often found in national or subnational labor codes but, depending on the country, the degree of job security can also be defined by court decisions, sectoral collective bargaining agreements, or even unwritten industrial norms. For example, in many countries (especially in Northern Europe), there are no severance requirements stipulated in the labor code but sever ance obligations are imposed in collective agreements. There are significant variations within the industrialized world in terms of the protection offered to regular workers. According to OECD (1999a) rankings, South ern European countries generally have the strictest arrangements while the Anglo Saxon countries have the least restrictive -- especially the United States.108 Advanced CEE countries fall in between the two extremes (Riboud, Sanchez Paramo, and Silva Jauregui 2001). During the past decade, there has been no clear trend in this aspect of labor market regulation: Some countries have eased restrictions, a few have strength ened them, but in most, arrangements have remained relatively unchanged. Table AIII.3 provides examples of termination arrangements for some OECD countries. Measuring the degree of job security protection afforded to regular employees is difficult, a fact that has been emphasized by many researchers (Betcherman, Luinstra, and Ogawa 2001). However, some consensus exists on its effects, based on studies of industrialized countries. Strict limitations on termination (for economic reasons) of regular employees, including generous severance pay and limitations on part time and flexible employment, are associated with:109 ·Lower labor turnover rates (hires plus separations); ·Lower aggregate employment levels; but greater numbers of long tenure jobs; ·Lower labor force participation rates; ·No clear impact on unemployment levels; but longer average unemployment durations; ·More self employment as a share of total employment; ·More nonstandard employment (for example, part time or temporary), although there is less consensus on this; and ·Positive employment effects for skilled prime age males but lower employment for women, young people, and less skilled workers. 108While U.S. employers have no statutory limits on dismissal rights, in reality they do face some constraints because of court decisions and collective agreement provisions. 109There is a fair degree of consensus (though not complete) in the research findings in industrialized countries. The conclusions have been drawn from a range of studies including OECD (1999a), Nickell and Layard (1997), Elmeskov, Martin, and Scarpetta (1999), Lazear (1990), and Di Tella and MacCulloch (1999). LABOR MARKET REGULATION 81 Strict job security for regular workers reinforces the same trends identified earlier in our discussion of hiring and contracting. A labor market with rules to protect job security -- such as one with rules that restrict nonstandard forms of employment -- has more stable jobs but also more long term unemployment and nonparticipation than labor markets without these protections. Together, restrictive hiring and firing regulations increase the protection available for incumbent employees but reduce access to formal, paid employment. The greatest risk seems to be that these rules worsen inequality by protecting "insiders" at the expense of more vulnerable "out siders." The key debate concerns the magnitude of these impacts. At least in developed countries, the employment effects appear to be smaller than many economists would assume (e.g., OECD 1999a).110 However, research in Latin American countries, where employment protection rules tend to be very strong, has generally found much larger negative impacts on employment and inequality, including in Latin America (Heck man and Pages 2000) In summary, statutory employer obligations toward permanent employees have been substantial (specifically regarding termination rights) in Russia, although this may change somewhat with the new Labor Code. Where obligations are large, the international experience indicates that the result is more informalization, and (by promoting voluntary quits rather than layoffs) potential reductions in the productive efficiency of enterprises. In OECD countries, employers often overcome high protec tion accorded permanent employees in the labor law through the use of fixed term and temporary contracts. In Russia, as discussed above, these options are restricted, and employers have therefore resorted to wage arrears, administrative leave, and con tracting in the informal sector. The new reforms appear to moderate the excessive termination conditions. However, it is important that employers have the right to adjust their work forces to economic and technological realities. Unions should be consulted and given legitimate avenues of appeal but, ultimately, staffing should be at the discretion of employers.111 In the final analysis, however, explanations for employment rigidities, such as the low layoff rates, in Russia must go beyond any impositions of the law, particularly given that there appears to be significant mobility in some parts of the labor market. Reputational risk, poor incentives to firms to allocate labor efficiently, and weak enforcement of the law may also be important for explaining lack of downward adjustment of output to employment. This suggests that while legal reform is impor tant, it will not in itself reverse the unusual patterns we have observed in the func tioning of the Russian labor market. 110 Again the measurement problems must be acknowledged and, in particular, the capac ity of researchers to fully capture what is actually happening in the labor market. In this regard, it is noteworthy that the Di Tella and MacCulloch (1999) study, which uses a qualitative measure of employment protection, finds stronger impacts than virtually any of the other studies that attempt to use more formal, quantitative measures. 111 Maleva et al. (2001) find that new code still imposes considerable costs on employers. 82 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY D. Wage Determination Regulatory framework (pre 2001 Labor Code). Salaries in the U.S.S.R. were based on the unified Tariff Schedule of Wages and Salaries, which established wage supplements and coefficients depending on the region, occupation, and character of the work. In the budgetary sector, wages are still set centrally accord ing to the Unified Tariff Table. This calculation is made by multiplying the mini mal monthly wage by a coefficient that corresponds to the employee's qualifica tions. However, salaries outside the budgetary sector were deregulated in 1992 Enterprises may set wages independently and are bound only by minimum wage regulations and any applicable collective agreements (see below). There are some other restrictions. For example, the law states that workers in the North must be compensated with higher wages, given the arduous living conditions in that region. This practice is markedly different than in OECD countries, where wage setting is a function accorded to the market and collective bargaining arrange ments, rather than to the state. The new Labor Code largely continues existing wage regulations. What is new is that the Code now stipulates that the minimum wage for the whole territory of the Russian Federation cannot be lower than the subsistence minimum defined for a working age individual. Although the state cannot guarantee such a level of a minimum wage immediately, it is assumed as a long term goal per se. The sub sistence minimum -- an absolute poverty line based on a minimum basket of goods and services -- is not an appropriate benchmark for a minimum wage. Dif ferent countries have different practices in establishing a minimum wage. In many countries, for example, it is negotiated by social partners and established thereafter by the Government taking into consideration many other aspects of the labor market. Moreover, since the cost of living and labor market conditions differ enormously by region, a single federal minimum wage may not be appro priate for Russia; certainly, it will not reflect the large regional variations in sub sistence minimums. The new Code also states that in order to increase the level of real earnings, wages should be indexed according to a consumer price index. In a market economy, wage levels (including the minimum wage) should be negoti ated by social partners and individually between the worker and the employer. Especially in a period of deep economic recession, in order to maintain employ ment and avoid bankruptcies, it may be difficult to keep the level of real wages intact. Current practice. Wage practices differ significantly by form of ownership. These differences include wage levels, benefits offered, the basis for determining wages, and the importance of variable pay (for example, bonuses) and unreported wages. However, some wage practices are universal. First, management almost always controls wage determination. As we will discuss later, collective bargaining rarely takes place and, with generally slack labor market conditions, only highly skilled employees have any power to bargain as individuals. Second, another important aspect of wages concerns nonpayment. This includes some practices we have already discussed, such as unpaid administrative leave and wage arrears. There LABOR MARKET REGULATION 83 Table III.4. Official Minimum Wage and Average Monthly Wage, 1995 2000 Years Official minimum monthly wage (Rbl.) Percent of average monthly wage due 1995 42.6 9.0 1996 72.7 9.2 1997 83.5 8.8 1998 83.5 7.6 1999 83.5 5.3 2000 (August) 132.0 5.7 2001 (Quart. 1 3) 300.0 9.7 Source: Russian Economic Trends, October 2000 (tables 5 and 6). is also substitution of non monetary for monetary compensation in the form of in kind payments discussed earlier.112 Sectoral wage setting. In the last chapter we noted that the private sector pays a wage premium over the public sector. This is consistent with evidence from many transition economies, for example, Bulgaria, Poland, and Romania (World Bank 1997, 1998, and 2001) There are various reasons why wages are highest in the private sec tor. Clearly, labor productivity may be higher. However, as Clarke and Kabalina (2000) argue, a higher private sector wage may also reflect a compensating differential because enterprises in the private sector offer less stability, require a more disciplined work environment, and provide fewer social benefits than employers in the state and privatized sectors. In fact, employees in private sector firms are often denied legally prescribed benefits such as paid leave, sick leave, and health and maternity benefits. This is collaborated by Tchetvernina (2000) in their survey of employees (table III.5). Pay differentials also reflect variations in how different types of enterprises set wages. As noted above, salaries outside the budgetary sector have been deregulated since 1992. In reality, wage deregulation has developed furthest in the de novo private sector (Clarke 1999). The private sector also relies much more on performance as a basis for setting pay levels. As table III.6 indicates, payments in the state and budget ary sectors are determined in 80 to ­90 percent of cases according to time based wage rates. In de novo private firms, on the other hand, almost half of respondents report that individual, collective, or enterprise outcomes (profit sharing) determine their wages. Informal payments. As noted in the previous chapter, officially reported wages do not capture full wages paid to workers. Wage payments, especially in the private sec tor, are often made in two parts (Clarke 1999). First, there is the official, reported wage that is very low. This practice is abetted by the extremely low statutory mini 112 In the RLFS panel, 11 out of the 85 firms reported non monetary payments in 2000. For these firms, these represented about 10 percent of total wages. At least for this panel of enter prises, the incidence of non monetary compensation was much higher in 1996 and 1997 (Tch etvernina et al. 2001). 84 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table III.5. Percentage of Employees, by Sector, Reporting Guarantees Stipulated by Legislation or Contract Are Not Fully Provided, 1999 Guarantee either not provided State Privatized De novo private or only partially Paid leave 1.6 2.3 22.6 Sick leave 8.0 8.8 37.8 Overtime 29.6 47.3 50.1 N 558 771 884 Source: Tchetvernina et al. (2001). mum wage. Second, there is the actual payment that is much higher than the reported wage. Actual payments can differ substantially from reported wages. Goskomstat (1997) has estimated the difference at 20 percent. Based on their 1999 survey of employees, Tchetvernina (2000) finds that over one third of private sector employ ees earn more than their registered wage. In 10 percent of these cases, actual pay ments are at least six times the official level (table AIII.4). This practice offers tax advantages to employees on undeclared earnings and also allows them to receive compensation in the form of current cash, as tax rates are very high and future social benefits are uncertain. This practice raises concerns. For work ers, their employment situation becomes vulnerable to arbitrary managerial discre tion since they risk losing the unofficial wage component. From a public policy per spective, this two tier wage practice hurts public revenues and contributes to the gen eral compliance problem characterizing labor relations. Tariff structure. The Unified Tariff Table for public administration workers, using the norm of the minimum wage grade (tariff) for the whole salary grid, compresses the remuneration scale of budget employees.113 This may have implications for the quality and quantity of staff in that sector, an issue that is being taken up in the pub lic administration reform program of the Government. What is surprising is that despite its deregulation for non state sector, this tariff system remains important for setting wages in the other sectors of the economy. In the 1996 Survey of managers of industrial joint stock companies and state enterprises, 95 percent reported that they used the tariff system in setting wages. Among the panel of 85 manufacturing enter prises (with no de novo firms), the role of the tariff system has continued to be impor tant. In 1996, 68 percent of the firms reported that they either applied the state tariff scale directly or used it as the basis for their own wage scale; the corresponding figure for 2000 was 71 percent (Tchetvernina 2000). The continuation of this practice runs contrary to expectations. One would expect that the use of the tariff would lose importance over time, rather than contin uing or even increasing. It is unclear why this has happened. Perhaps employers and employees do not have a better way of determining wages given poor signals coming from the labor market. There are potential adverse impacts. The farther wage setting 113 Minimum wage grade was raised to Rbl. 450 on December 1, 2001. LABOR MARKET REGULATION 85 Table III.6. Forms of Wage Payment by Sector, Kemerovo and Komi, 1997 Form of wage payment State Budgetary Privatized De novo Total Percent of employees Piece wage, individual 9 5 13 19 11 Piece wage, collective 6 2 11 11 7 Time wage 81 91 68 53 76 Mixed (piece wage and time wage) 4 1 6 7 4 Percentage of profits (sales) 1 1 2 10 3 : Clarke (1999, . 5.11). is from market price, the less efficient is the market in allocating labor to its most pro ductive use. This makes investment/allocation decisions by employers and employees suboptimal. For example, employees would be less likely to invest in human capital if wage differentials do not reflect productivity differentials. International evidence. Here, we focus on the determination and application of the minimum wage. Like the other aspects of labor market regulation discussed in this paper, the role of minimum wages is controversial. The underlying idea is quite simple -- to set a floor on what employers can pay in order to ensure that employees receive a "fair, living wage" and thus to support the incomes of low wage workers and their families. While minimum wages can boost the earnings of low income employ ees, they can also lead to unemployment where the minimum wage is above the mar ket clearing level and where it is actually binding.114 The different views on minimum wage policies essentially hinge on the relative weight attached to these positive and negative effects. The controversy has heightened in recent years because of conflict ing evidence regarding the actual employment impacts of increases in minimum wages.115 Minimum wage regulations can have several dimensions: (a) the level set; (b) cov erage; (c) differentiation in the level (for example, by age, sector, region); (d) how the level is adjusted to reflect inflation; and (e) how the level is set (for example, by Gov ernment or by the social partners). Most (but not all) industrialized countries do set minimum wages but there is considerable variation in the details. The level set illus trates this variation. In OECD countries, in 1997 the adult minimum wage as a per centage of full time mean earnings ranged from 28.8 percent in Spain and 34.9 per cent in the United States to 51.1 percent in the Netherlands and 55.3 percent in France (OECD 1998). During the past decade, there has been a general decline in minimum wage levels, both in real terms and as a percentage of average wages (OECD 1998, 1999b). In CEE transition countries, the minimum wages are lower than 114 For a concise review of the theory regarding the employment impacts of the minimum wage, see OECD (1998). 115 The literature summary in this chapter is based on OECD (1998). 86 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure III.1. Minimum Wage as a Proportion of Average Wage, Transition Countries Minimum wages as a percent of average wage 0,45 0,40 0,35 0,30 0,25 0,20 0,15 0,10 0,05 0 dnaloP ainauhtiL aina aine moR aivtaL airagnuH avodlo aissuR cilbupeR hcezC natsikebzU aigroeG natsikijaT natsine suraleB M mrA eniarkU kilbupeRzygryK mkruT natshkazaK Source: World Bank (2000b). in OECD countries and are approximately 30 percent of average wage (Rashid and Rutkowski, 2001) (figure III.1). Not surprisingly, the labor market impacts of minimum wages depend heavily on the level at which they are set and how well they are enforced. In some countries, the level is too low to be binding (that is, to affect wage and employment decisions). The general trend toward declining real minimum wages presumably has led to weaken ing of its impacts on employment and earnings. Most of the empirical research on the impacts of minimum wage is based on the experience of industrialized countries, especially the United States. Careful consider ation is needed when generalizing this experience to Russia, where enforcement is weak and the minimum wage level is set too low to matter in the labor market. How ever, we can review the industrialized country experience to understand the effects of minimum wages when they are binding and enforced. Much of the debate about minimum wages concerns their impact on employ ment levels. Currently, at least in the United States, there is a lot of controversy about this effect (for example, Neumark and Wascher 2000; Card and Krueger 1995, 2000). An international review by the OECD (1998) based on nine member countries concluded that there is a significant negative employment effect for teenagers -- in the neighborhood of a 2 ­to 4 percent decline in employment -- for a 10 percent increase in the minimum wage. This impact then diminishes and effectively disappears for the prime age group. On an aggregate basis, then, researchers tend to find modest or insignificant employment effects; however, negative effects become more significant once analysis focuses on workers actu ally constrained by the minimum wage (for example, youths and other low wage workers). LABOR MARKET REGULATION 87 Can minimum wages reduce inequality and poverty? On the distributional side, studies (almost exclusively in industrialized countries) have found that higher mini mum wages do reduce the dispersion of earnings and the incidence of low pay. They also tend to narrow wage differentials between demographic groups (for example, age and gender). In some developed countries with large numbers of "working poor," increases in minimum wages have had modest impacts on poverty; however, the OECD (1998) has concluded that minimum wages can play only a relatively minor role compared with other factors (for example, macroeconomic conditions, gen erosity of public assistance). At its current level, the Russian minimum wage can hardly be characterized as an instrument of wage policy. Its role in alleviating poverty is extremely limited, and it does not provide any adverse effects on unskilled workers. In summary, Russia has a very low minimum wage that is not currently binding in any sense. Given its low level, the minimum wage could be raised substantially to address poverty among low end workers without creating negative employment incentives, even for young or unskilled workers. There are several caveats. The mini mum wage will become a more relevant policy instrument as the economy is formal ized and enforceability improves. Under these conditions, the ultimate level of mini mum wage should be evaluated against the average wage and kept sufficiently low (about 30 percent or so) so that it does not create adverse work incentives. Regional differentiation in the minimum wage should be considered given the large variation in average wages in Russia. The provisions in the new Labor Code that create public obligations for guaranteeing and creating a single nationwide minimum wage, and keeping its real value fixed, are worrisome in this regard. The continued use of the tariff, despite its deregulation, suggests that wages do not convey important signals about worker productivity to employers or about labor sup ply and demand. Indeed, despite recent gains in wages as signals of worker produc tivity, wage practices such as wage coefficients for hiring Northern workers, nonre porting of wages, and the compressed public wage scale continue to make wages a noisy indicator of the opportunity costs of labor in Russia. The new Labor Code does not make any significant changes in this area. E. Trade Unions, Employer Organizations, and Collective Bargaining Regulatory framework (pre 2001 Labor Code). In the Soviet era, virtually all workers belonged to trade unions.116 However, the function of unions differed greatly from what they typically do in market economies. Their major role was to pro vide social services, recreation and culture, housing, consumer goods and services in short supply, and sick pay. They were also part of the official apparatus and had vari ous functions associated with the administration of the labor and social insurance 116 For a discussion of trade unions before and in the early years of the transition, see Hoffer (1997). 88 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY systems. At the enterprise level, unions were a partner of management and mobilized workers for production. In fact, enterprise directors could be union members. Collec tive bargaining over the terms of employment (including wages) did not take place. With the transition, unions lost their "quasi state" functions (for example, to introduce draft legislation, impose penalties for labor law or safety violations, decide labor disputes, administer social insurance). During the 1990s, legal reforms were introduced to provide for the basic industrial relations concepts in a market econ omy. These included the right to form independent unions, collective bargaining rights for unions and employers, the right to strike (but not lockouts), and the exclu sion of employers from union membership or bargaining for workers. A tripartite commission was also established (discussed below).117 As Hoffer (1997) points out, with the changes in the labor regulatory regime, the Russian model changed from one where the state structured all aspects of the employment relationship to one where the state role was to establish minimum stan dards with management and labor to "flesh out the framework provided by law" through the negotiation of the specific terms of employment. In principle, this is an appropriate model that follows the practice of advanced countries. However, the capacity to carry it out in the current Russian context is inadequate because of the weaknesses of labor institutions, including unions, management, and (as we will dis cuss below), enforcement and dispute resolution institutions. A framework for collective bargaining exists in Russia, as in other market economies. The Law on Collective Agreements defines a framework for agreements at different levels. At the highest level, the General Agreement sets general principles regulating labor relations in the Russian Federation. It is signed every year by the Russ ian Tripartite Commission, which includes representatives of the Government, the unions, and employers. The Law on Collective Agreements also provides for tripartite regional, sectoral, and professional agreements. Agreements also can be negotiated at the level of the enterprise. As we will see in the next subsection, however, little real bargaining occurs, and the wages and working conditions are rarely determined in any real or enforceable manner by collective agreements In the new Code, provisions remain for collective bargaining at all of these levels. The Code does change procedures for determining bargaining representatives for employees. These new rules specifically pertain to what is considered a "local union" as well as to how a bargaining representative is selected where multiple trade unions exist. These provisions may have the effect of limiting the opportunity for small and independent unions to represent workers. Current practice. Although membership is no longer universal, the most com monly reported union density rate of 75 percent is nevertheless among the highest in the world (ILO 2000a).118 The trade unions are consolidated into trade union centers, 117 The Constitution guarantees that everyone has the right of association, including the right to create trade unions. Russia has ratified the ILO conventions 87 and 98 regard ing freedom of association and collective bargaining. 118 This figure represents union members as a share of the total nonagricultural labor force. The ILO (2000a) reports this density for 1995. LABOR MARKET REGULATION 89 Table III.7. Who Protects the Employees? Opinions of Employees, Employers, and Trade Union Leaders, 1999 All enterprises Enterprises with trade unions Employees Employers Employees Employers Union leaders Distribution (percent) Employer/manager 38.9 64.0 34.7 62.4 26.7 Trade union 6.2 9.1 11.2 16.8 49.3 Workers themselves 25.9 6.5 24.9 5.6 12.2 Labor contract ­ 10.4 ­ 11.2 ­ Nobody 22.6 3.9 23.6 3.2 4.6 Other 6.4 6.1 5.6 0.8 7.3 Total 100.0 100.0 100.0 100.0 100.0 N 2213 278 1049 132 132 Source: Tchetvernina (2000). the largest being the FNPR (Federation of Independent Trade Unions of Russia), which was founded in 1990 as the ancestor of the official Soviet Central Council of the Trade Unions. Other trade union centers emerged during the 1990s as new inde pendent unions were formed.119 The high union density figure greatly overstates the health of the labor movement in Russia and its capacity to represent workers. Surveys suggest that membership levels are much lower than the cited numbers; according to some studies, membership actually declined by 25 percent in the 1990s (Wesolowsky 2000. There is no doubt that many economic and social trends are acting against unionization. Most notably, unions are almost unknown in the de novo private sector. (table III.7). The financial situation of the trade unions is also difficult because of sharp declines during the 1990s in membership dues received (Hoffer 1997).120 How well have unions represented workers?121 According to the 1999 survey car ried out by Tchetvernina (2000), both employees and employers rated unions poorly in terms of their performance in representing the interests of workers (table III.7). In unionized establishments, only 11.2 percent of employees and 16.8 percent of 119 Recently, Russian trade unions have become integrated into the international labor movement because the FNPR and two smaller unions All Russia Labor Confederation (VKT) and Russia's Labor Confederation (KTR) have become affiliates of the International Confedera tion of Free Trade Unions, the world's largest trade union body. 120 Trade unions, especially the FNPR, do have substantial assets in, and receive important income from, real estate from the Soviet era. 121 These latter unions originated largely as protest movements. While they played impor tant roles in the last years of the Soviet era and in some industries during the transition (for example, mining), they have not been able to consolidate themselves as a real national alterna tive to the Russia's Federation of Independent Trade Unions (FNPR) (Hoffer 2000). 90 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY employers identified unions as the primary protector of employee interests. Given these responses, table III.7 suggests that union leaders may have an unrealistic sense of their own role; still, one half did not believe that their unions were the primary protectors of workers. According to these results, then, workers must rely on them selves or management for protection. A sizable proportion of employees believes that nobody is representing their interests. Despite the provision in Russian law for full collective bargaining rights at national, sectoral, regional, and enterprise levels, little real bargaining occurs. The nationally applicable General Agreements largely consist of nonenforceable state ments of intent on social and labor policies (Denisova, Friebel and Sadovnikova 1998a; Hoffer 1997). According to Denisova, Friebel and Sadovnikova (1998a), sec toral agreements are often formal and merely reproduce legislative norms.122 While some regional agreements have more substantive content, wide variations exist (Hof fer 2000). These framework agreements have generally not been an instrument for collec tive bargaining of wages and working conditions. Most observers point to the lack of employer representation as a major stumbling block. In many instances, agreements are signed between governments and unions with no representative of the employ ers.123 This lack of representation reflects both tradition (where the state was the employer in labor agreements) and weak employers' organizations.124 The upshot, however, is that employers typically do not take responsibility for the content of these agreements and, even where terms of employment are stipulated, enforcement is a major problem. At the level of the enterprise, the available evidence suggests that contracts do lit tle to represent worker interests. In a 1999 establishment survey, 48 percent of union leaders said that their collective agreements did not protect the socioeconomic inter ests of their workers (Tchetvernina 2000). This lack of protection partly reflects the contents of the agreements, which do not always specify the terms of employment (including wages). It also reflects the inability of unions to ensure their enforcement. In the 1999 survey of enterprise trade union leaders, only 20 percent reported no vio lations of the collective agreement. The most frequently violated provisions were wage arrears (cited by 59 percent of union respondents); wage increases (24 per cent); work safety and conditions (21 percent), and benefits (11 percent) (Tch etvernina 2000) Simply in terms of coverage, 1996 Ministry of Labor data indicated that only about 18 percent of enterprises were covered by collective agreements. And this figure masks wide variations by region and type of enterprise. Collective agreements are less prevalent where economic restructuring has occurred more rapidly. They are rare in 122 Sectoral agreements are generally not binding for enterprises that do not sign the agreement. 123 For example, only 17 of the 58 sectoral agreements in 1996 were signed by employer representatives (Denisova, Friebel and Sadovnikova 1998a). 124 Employers' associations are beginning to emerge although they have a long way to go, especially in terms of representing the private sector. According to Denisova, Friebel and Sadovnikova (1998a), more than 60 employers' associations exist, with about 30 to 35 func tioning in practice. LABOR MARKET REGULATION 91 smaller enterprises and in the de novo private sector. Just slightly more than 5 percent of Moscow's registered enterprise trade unions had managed to sign a collective agreement that would be recognized as a valid document by the courts (Hoffer 2000). He reports that the city's trade union federation explains the low level of collective bargaining as the result of local union representatives not knowing their legal rights and being too close to management. Tchetvernina. (2000) found in 1999 that 80 per cent of enterprise union leaders favored the traditional Soviet practice of including management in unions. Hoffer (2000) also argues that management intimidation is an important factor in the low level of union activity and collective bargaining. There are currently a number of cases before the ILO's Committee on Freedom of Associa tion.125 International evidence. Collective bargaining can play an important role in determining wages (and other conditions of work). It is well known that, ceteris paribus, unions can raise wage pressures, and wages bargained collectively are gener ally higher than those bargained individually. Economists also focus on how respon sive wages determined through collective bargaining will be to labor market condi tions. The characteristics of the bargaining process (that is, the structure in which bar gaining is carried out) can matter as well as the extent of bargaining. So researchers have attempted to look at both dimensions. Studies have tended to use two measures of the extent of bargaining: trade union density (union members as a percentage of the work force) and collective bargaining coverage (percentage of workers having wages determined by collective bargaining). They have also looked at two dimen sions of bargaining structure: the degree of centralization and the degree of coordi nation. There are major differences across countries in terms of the extent of bargaining and the bargaining structure. These differences reflect both industrial relations laws and culture and practice. The extent of collective bargaining is shrinking in many countries, at least on the basis of union membership trends (table AIII.6). In terms of bargaining structure, there has perhaps been some shift away from centralized and coordinated approaches to more enterprise level bargaining. In CEE transition coun tries, for example, most bargaining occurs at the level of the firm. However, this is not a universal trend (ILO 2000a) Economists also test the hypothesis that representation of worker voice through unions and collective bargaining can reduce discrimination. And by instituting dis pute resolution mechanisms, reducing arbitrary management decisions increases job tenure and investment in training. It can also help improve work safety conditions. These positive effects of unions would contribute to greater labor productivity. How ever, unions can have monopolistic tendencies, pushing up wages more in their own sector than in the economy as a whole, thus increasing wage disparities. Still, this is not always negative if "discrimination" of some workers leads them to accept lower wages than other workers with similar skills. In addition, unions can sometimes resist 125 According to reports by the ICFTU, there have been numerous cases of obstruction of lawful union activity, including alleged incidents of union activists being murdered. (See ). 92 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY reforms, such as restructuring of enterprises, as in Latin America. Unions may not always resist reforms. For example, Poland and South Africa are examples where unions had a very positive impact of driving their respective countries toward politi cal and economic freedom. An extensive survey of the literature on the economic and employment impacts of unions and collective bargaining on the basis of the experience of OECD countries (Aidt and Tzannatos 2000) and other countries (World Bank 1995)126 and Nickell and Layard (1997) includes the following conclusions: ·Collective bargaining increases wages for covered workers by 5 to 15 percent (depending on the country). The size of this premium increases when total compensation is measured because unions also bargain for better benefits. ·This wage pressure, all else being equal, could raise unemployment, but is offset in some countries by effective coordination of wage bargaining by unions and firms. However, this coordination can be fragile and break down. If it does, as in the case of Sweden in the 1990s, then strong unions can have very adverse impact on wages. ·At the aggregate level, bargaining coverage (but not union density) tends to be associated with higher real wage growth but lower employment (and in some studies, higher unemployment). ·Unions and collective bargaining compress the wage distribution and particu larly the differential between skilled and unskilled workers. Countries with union (density) tend to have lower earnings inequality. ·There is no consistent evidence on the impacts of the degree of centralization in collective bargaining. ·Unions can also reduce discrimination against women, ethnic groups, and other minorities. ·Job tenure is longer in firms with unions, and more training is carried out in unionized firms, promoting growth in labor productivity. ·Unions increase compliance with worker safety and health standards. When combined with overall improvement in industrial relations, this helps increase labor productivity. ·Unions can also engage in monopolistic behavior and opposition to reform, and reduce productivity growth. In the United States and Britain, unions are nega tively associated with productivity growth. In continental Europe, this pro ductivity impact is minimal, suggesting that effective coordination can negate the adverse impact of unions. ·While coordination between employers' organizations and unions seems to have improved macroeconomic and labor market performance in the 1970s and 1980s, the evidence is less clear in the 1990s. ·Competitive product markets and laws that give workers the right to opt for the union of their choice or not at all also enhances the positive impact of unions and negates their adverse effects. 126World Bank (1995). LABOR MARKET REGULATION 93 In summary, Russia has made some progress in making the transition in industrial relations from a regime designed for the planned economy to one appropriate to a market economy. For example, some important pieces of the legal framework are now in place. However, there is a long way to go, particularly in terms of developing the institutions that underpin effective industrial relations. Unions or the bargaining structure do not adequately reflect the voices of employers or workers. International research has demonstrated that worker voice, embodied in the true representation of workers and employers in the bargaining process, can improve training and health and safety in the work place, thereby contributing to productivity gains and improve ments in worker welfare. In the new Code, provisions remain for collective bargain ing at all of these levels. The Code does change procedures for determining bargain ing representatives for employees. These new rules specifically pertain to what is con sidered a "local union" as well as how a bargaining representative is selected where multiple trade unions exist. These rules may have the effect of limiting the opportu nity for small and independent unions to represent workers. Reaching the goal of modern industrial relations will require true worker and employer representation in unions and collective bargaining. However, unions can also raise wage pressures, and all else being equal, raise unemployment. Ensuring that product markets are competitive should help contain wage pressures by unions. There is no consensus in the literature about whether centralized or decentralized bargaining is more efficient. But given that Russia now has a very centralized regime that is not achieving genuine bargaining outcomes, encouraging more decentralized bargaining where there is no existing institutional inertia might be considered. Achieving this objective will require developing capacity of unions and employers, as well developing enforcement and dispute resolution institutions. We discuss these institutions below, in the final section of this chapter. F. Enforcement and Dispute Resolution Regulatory framework (pre 2001 Labor Code). The principal responsibility for monitoring and enforcement of labor regulations rests with the Federal Labor Inspectorate under the MLSD.127 Table AIII.7 summarizes the activities of the labor inspectorate between 1994 and 1998. It shows a large increase, especially during the early years of this period, in the inspections carried out and in the number of viola tions found. However, it is generally understood that labor legislation is still violated on a massive basis. Note from table AIII.7 that, despite the large numbers of infringe ments (more than 2 million a year), only a relatively small number of employers are actually penalized in some way. The resources of the Federal Labor Inspectorate are inadequate for fulfilling its mandate. As table AIII.7 indicates, in 1998 there were just 4,720 staff labor inspectors to cover a work force of 65 million workers and more than 700,000 establishments. 127 The Labor Code also provides for labor inspectorates to be established by other bodies, including trade unions. 94 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY The new Code does not appear to make major changes in this area. A positive aspect of the approach is that most conflicts are intended to be resolved at the enter prise level, which should minimize costs and time requirements. On the other hand, the system creates a cumbersome practice of reconciliation of differences at the enterprise level. The timetable for hearing and resolution of labor disputes is very tight. Each dispute should be heard within 10 days after it is filed, and should be exe cuted within 3 days but after 10 days allocated for an appeal. On many occasions, the worker can and has to resolve a dispute directly in court, These include periods when the worker is not satisfied with the resolution of the commission, the issue of rehabilitation to work is on the agenda, the date and cause for dismissal are appealed, the worker is on transfer to another job, or the workers is on payment for an idle period. Although the Code provides labor inspectors and inspectorates sig nificant privileges and rights to monitor the execution of labor legislation, their role as mediators, conciliators, and arbitrators of labor disputes is diminished if not non existent. Current practice. Weak enforcement in Russia is evident in practice. As noted in the previous chapter, nonpayment of contractual obligations, or wage arrears, spread to nearly 60 percent of all workers in 1998 and, despite declining, continues to affect a significant share of the work force. Other contract violations are also evident. In 1999, about 48 percent of the individuals on administrative leave did not get any cash compensation during their absence despite regulations requiring they receive two thirds of their regular pay (Goskomstat 2000b) Most worrisome, contractual viola tions are more prevalent among workers with the least bargaining power. Thus, lim ited enforcement means that some workers are more affected than others. Legal restrictions are also binding on particular firms. Formal termination proce dures seem to apply relatively more to more "visible" entities. Clarke (1999) finds that the share of employees in state and budget entities that believe they can be easily dis missed without formal grounds is much lower than the share in privatized firms, and both are well below the share of such employees in de novo private firms (figure III.2). A recent study by Pinto et. al. (2001) of three regions also found that the labor law was binding on the labor reallocation decisions of half the surveyed firms. While not conclusive, this evidence suggests that limited enforcement mechanisms do not cre ate a level playing field for all firms. Why is enforcement weak? There are many reasons for the weak enforcement of legislation and the basic rights of workers. Certainly, these problems are indicative of the more general problem of compliance that plagues many aspects of life in Russia. Moreover, in the past it has been exacerbated by the current high levels of unem ployment and weak demand for many types of labor that afford employers a great deal of discretionary power in managing their work forces. As well, there are specific problems related to labor market regulation. One involves the Soviet tradition of sub ordinating workplace concerns, such as occupational health and safety.128 128 In that tradition, which still exists, dangerous or unhealthy working conditions are com pensated for by special wage premiums, additional leave, free meals, and early retirement. Many workers have (albeit short run) incentives not to seek improved conditions. LABOR MARKET REGULATION 95 Figure III.2. Percentage of Employees Reporting They Can Be Dis missed without Any Formal Grounds by Sector, Year 60 50 40 30 ercentageP 20 10 0 ate State All Budget atizedv priv Pro ov no De Source: Clarke (1999, table 5.3). More importantly, the institutions for the enforcement of legislated worker rights and labor contracts, and for the effective resolution of labor management disputes are ineffective or nonexistent in Russia. Parties can pursue claims through the legal system. However, labor courts do not exist and cases must be brought to civil courts. Although procedures stipulate that labor cases be handled quickly (within a month at most), they often take much longer. Moreover, in most situations, employees do not have legal representation for financial reasons (Denisova, Friebel, and Sadovnikova 1998a). The courts also face a major challenge in simply handling the huge volume of labor cases. The flood of unpaid wage complaints exacerbated this situation in the mid to late 1990s. In 1997, 2 percent of the labor force was involved in lawsuits with their employers -- 97 percent over unpaid wages. In that year, the civil courts heard more than 1.3 million wage payment cases, representing one third of all lawsuits. In these unpaid wage cases, the court almost always has decided in favor of the employee (99 percent in 1997).129 However, court decisions were often not enforced, 129 The court decisions regarding other types of labor disputes were also generally in the employee's favor but not overwhelmingly so. For example, between 1994 and 1997, two thirds of dismissal related cases were decided in favor of the employee (Denisova, Friebel and Sadovnikova 1998a). 96 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY and many workers did not receive compensation (Denisova, Friebel and Sadovnikova 1998a). The difficulty of this situation was compounded by the fact that a substantial share of unpaid wages was in the government sector. As chapter I indicated, eco nomic growth (hence improvement in the bargaining power of workers) and the criminalization by the Government of wages arrears in 1999 has led to a significant decline in wage arrears both in absolute terms and as a share of the total overdue payables of enterprises, but arrears still remain sizable in 2001 and it is obvious that this problem has not disappeared. Procedures also exist for enterprise level mediation and arbitration committees to address the enormous number of alleged labor violations and other workplace dis putes. These committees include representatives of both the workers and the enter prise. However, available evidence suggests that neither employers nor employees consider these committees to be effective mechanisms for resolving disputes. Table AIII.8 summarizes data from Tchetvernina (2000) on which mechanisms the parties have used to resolve a labor dispute -- most often related to wages. As the table shows, there are some differences between employers and employees. However, there is agreement in that only a very small number have relied on the available dispute res olution institutions: labor management dispute committees; trade union commit tees, or the courts. Not unexpectedly, survey data suggest that this situation does not lead to reason able outcomes, at least from the worker's perspective. Note from table AIII.8 that 44 percent of employees indicated that the conflict had not been resolved. In another survey by Tchetvernina (2000) with a much larger sample of employees (n = 2213), 37.2 percent reported that they do not seek assistance anywhere when faced with a labor conflict. Finally, the Federal Law on Collective Bargaining Procedures gives employees the right to organize strikes when disputes have not been resolved or when employers have not fulfilled the terms of the resolution. Lockouts, however, are not permitted. In recent years, the major reason for strikes has been unpaid wages, and the level of strike activity has roughly reflected the trends in wage arrears, peaking in 1997 and declining in the past couple of years (figure III.3). In their survey of enterprise trade union leaders, Tchetvernina (2000) have found that only a small number view strikes as an effective means of defending workers' interests. However, between 1998 and 1999, the percentage taking this view increased significantly (from 12.5 percent in 1998 to 25.2 percent in 1999). Perhaps the relative success of labor strikes over the wage arrears issue has had a demonstrable effect on union leaders. International evidence. Compliance with labor laws and valid labor contracts and the resolution of disputes represent important elements of the labor market regulatory framework in OECD countries. Why are these institutions important aspect of labor mar ket regulation in OECD countries? Enforcement of contracts (written or oral) improves both consumption and production efficiency. If contracts are not enforced, then there is a time inconsistency problem. At any time after the contract is negotiated, either party that finds that it can obtain better terms through reneging and renegotiating terms will do so. Renegotiated terms would favor the party that has the greatest bargaining power at any one point in time. This is consistent with evidence on contract violations in Russia. The LABOR MARKET REGULATION 97 Figure III.3. Number of Person Days Lost in Strikes, 1995 99 7000 6000 5000 (000s) lost 4000 ys da 3000 Strike 2000 1000 0 1995 1996 1997 1998 1999 Years Source: Russian Economic Trends, October 2000, table 5. ensuing income uncertainty for workers and employers reduces investment in human and physical capital and work effort for workers (Rashid and Townsend 1994). Perhaps most importantly, labor contracts are the most important contracts for most individuals. When those contracts are not respected and enforced, this reduces confidence in other labor and non labor contracts into which the individual might enter. North (1990) has argued that these are critical institutions in promoting imper sonal exchange, which in turn explains much of the differences in economic growth and performance. Nonenforcement of labor contracts in Russia are, by this reasoning, inimical to the healthy development of a market economy. Unlike the other issues covered in this chapter, there is little empirical work on the economic or employment impacts of different enforcement arrangements or dispute resolution. However, there is an emerging best practice.130 Enforcement. International best practice in enforcement has been undergoing important changes over the past decade. Much of the innovation has taken place on the occupational health and safety front but it applies to a wide range of compliance issues. These changes reflect new ideas about the economic impacts of compliance -- that is, growing emphasis on the longer term competitive advantages of healthy, safe, and legally compliant workplaces as opposed to the short run benefits of undercut ting competitors. New approaches are also emerging in response to the increasing complexity of enforcement in the labor field and to the stretched resources of inspec tion services everywhere. Innovations involve an emphasis on technical assistance as opposed to sanctions, using enterprise compliance plans as benchmarks for improv ing conditions, and involving the social partners. 130 This section is drawn from Betcherman, Luinstra, and Ogawa (2001). 98 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY The new approaches do not exclude sanctions because the threat of their imposi tion remains essential for demonstrating the rule of law. However, the experience in OECD countries, particularly with respect to occupational health and safety, under lines the benefits of involving employers in developing their own policies and imple mentation plans. This can save on the resources required of the inspectorate and it increases the likelihood of compliance by employers who "own" the strategy. Inspec tors can then judge the enterprise's performance against its own plan that is specific to its needs and circumstances (as opposed to the generality of legal requirements). Some countries require all enterprises over a certain size (for example, 50 employees) to prepare an annual plan on improving working conditions or to report each year to the labor inspectorate on progress made (Hammer and Ville 1998). This evolving approach to compliance still places important obligations on Gov ernment. It must develop a clear framework of rights, obligations, powers, structures, and mechanisms for enforcement. An effective administration and field inspectorate is necessary, although the role of inspectors is changing with the transition from sanc tions to technical assistance and support. Rather than simply inspecting premises and prosecuting statutory violations, inspectors take on a more "service minded" approach in which they work with the enterprise to resolve concerns and agree on a plan of action. Advisory, educational, and mediating skills become more important (Hammer and Ville 1998). One important set of issues concerns the resources required for enforcement and the appropriate balance between protection and costs. Many OECD countries have developed standards based on "reasonable practicability" and "disproportional meas ures" -- that is, that costs of prevention must be in proportion with the risks (Von Richthofen 1999). In terms of the source of funds for financing labor protection, the general practice is that costs should be borne by employers. In many countries, pre ventative activities and inspection services are financed from the social insurance fund. The ILO recommends against funding these activities through fines because this will inhibit the inspectorate's promotional/educational role and create incentives for sanctions (Hammer and Ville 1998). One challenge faced by all countries relates to where inadequate resources (for example, limited finances, stretched inspectorates) should be targeted. Hammer and Ville (1998) contend that large employers and the worst violators need to be the pri ority. To reach the rest, including the huge numbers of smaller enterprises, govern ments are trying a number of alternatives: involving other organizations (such as chambers of commerce, trade unions), using the media for national educational campaigns, targeting inspections, and establishing advisory services or accrediting private advisory service providers. Some governments (for example, Zimbabwe, Japan, the EU) subsidize or provide tax incentives to small and medium size enter prises engaging in preventative measures and/or during the transition to new regu lations. Dispute resolution. Effective dispute resolution relies on three key principles. First, prevention is always better than resolution. As discussed above, adequate enforce ment of labor laws goes a long way toward preventing labor disputes. Second, if a dis pute is unavoidable, the parties to a dispute ought to attempt to resolve it themselves. LABOR MARKET REGULATION 99 Third, if a dispute cannot be resolved, third party intervention ought to involve the disputing parties as much as possible (Heron and Vandenabeele 1999). In OECD countries, there has been increasing experimentation with approaches to dispute resolution that improve accessibility and minimize cost and time burdens. These innovations generally involve a move away from court based procedures and adversarialism and toward alternative noncourt approaches that emphasize fact finding, conciliation, and arbitration. Many of these newer approaches to dispute res olution build on the expertise of industrial relations specialists as opposed to legal experts. There is also growing interest in approaches that place the primary responsi bility for the resolution of disputes with the social partners (that is, management and labor), with the Government playing a role of catalyst and resource (for example, through an advisory service such as the British Advisory, Conciliation, and Arbitration Service) (Thomason 1993). The dominant innovation in many countries over the past half century has been the introduction of administrative labor tribunals as an alternative to litigation. Typi cally, legal review of the decisions of these tribunals is available through the court sys tem. One trend with administrative labor tribunals has been to promote an "inves tigative" rather than "adversarial" approach. This is especially successful where the representation of one party (usually the employee, especially in nonunion situations) either is absent or weak. Investigative approaches are characterized by (a) the active role of the tribunal authority or mediator in prior investigation and leading discussion during the hearing; (b) the lack of legal, adversarial processes such as cross examina tion of witnesses; and (c) representation by industrial relations specialists instead of lawyers (Clark 1999). Another related innovation is "extra judicial conciliation," which has been used, for example, in Chile to expedite dismissal related disputes. Labor inspectors can hear dismissal claims in order to determine the legality of the dismissal and the amount of wages/severance/welfare benefits due. Hearings inspectors can summon both the employer and employee to appear. This process has been cited as a "good practice" in labor law administration by the ILO (2000b). Alternative dispute resolution approaches (that is, that do not involve court pro cedures) have a number of advantages. They can be fast, informal, and simple without requiring expensive technical expertise. As well, they give the parties control of the process; for example, both generally must agree on "neutrals" such as conciliators, mediators, and arbitrators. By definition, alternative dispute resolution is less antago nistic and can preserve working relationships. Since settlements are not usually in the public record, they also protect the privacy of the parties. There are some disadvantages, however. One is a lack of transparency. Also, due process considerations, including rules of evidence, the right to representation, the right of appeal, and other basic court procedures, are not a part of these alternative dispute resolution approaches. These considerations have been relevant in the tran sition countries of Eastern Europe where both workers and employers have been concerned about the predictability and legitimacy of resolution outcomes -- and, as a result, have tended to continue to rely on court procedures. However, to address such concerns, some countries and organizations have developed guidelines such as 100 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY the Due Process Protocol adopted by the American Bar Association to ensure that the rights of each party are protected. Despite these concerns, alternative dispute resolu tion is largely considered a fair and effective means of resolving employment disputes and of reducing the backlog in courts and government sponsored labor tribunals (Zack 1997). In summary, enforcement and dispute resolution pose major challenges for Russ ian policymakers, employers, and labor. The failure of the existing institutions per petuates an environment where too many employers can violate laws and contracts with impunity; where there is little access to viable means to resolve disputes; and where employees too often expect that their concerns will be dismissed. The conse quences of the weak institutional framework for industrial justice are exacerbated in a slack labor market, and while disputes and contract violations dissipate when eco nomic activity increases labor demand, workers remain vulnerable to their re emer gence in times of economic slack. The new Labor Code does not appear to signifi cantly change this framework. Summary and Conclusions Labor market regulation in Russia is restrictive in law but not in practice. Labor mar ket regulation in Russia has been and largely remains unrealistically strong and inap propriate for a market economy. It imposes a lot of costs of worker protection on employers. However, for many firms and workers, in practice labor regulation is com pletely bypassed, so that the labor market is virtually unregulated. Has the Labor Code impeded restructuring in Russia? The weak enforcement of the restrictive labor law, coupled with evidence of significant labor flows before and after the crisis, indicate that the Labor Code has not been a major factor in preventing labor redeployment in Russia. Reputational risks of employers and other poor incentives to managers for laying off workers may also be at work in reducing the pace of layoffs earlier observed in Russia. The weak regulation of the labor market may have helped employment flexibility, but it also appears to have had tradeoffs in promoting infor malization of the economy (through its avoidance), lower worker productivity, and reduced worker welfare (for example, low wages, and growth of in kind substitutes and wage arrears). What can be done? The enforcement of restrictive law is not the solution. Rather, reducing excessive restrictions and increasing enforcement should be the focus of future efforts. The new Labor Code provides some improvements but more needs to be done, including providing more freedom to employers in deploying their work force. Moving to a flexible labor code that is fully enforced. A strict labor code without enforcement leads to violations of labor rights and reduces the welfare of workers below acceptable levels, and impedes labor productivity. A strict code with full enforcement will improve worker welfare but impose high costs on employers and restrict the ability of the labor market to adjust to economic realities, also limiting eco nomic productivity. The challenge for Russia is to move from a labor regulatory frame work that is restrictive and not enforced, to one that is flexible and fully enforced. LABOR MARKET REGULATION 101 The following priority actions would reduce excessive protection to workers offered by the legislative framework within the firm but, at the same time, would begin to strengthen the role of institutions in allowing workers a voice to ensure that basic rights are protected. These changes need to be complemented by a strong enforcement regime (dispute resolution, labor inspectorates). Social protection for workers, beyond the basic rights offered through labor legislation and more effective industrial relations, could be achieved through active and passive labor market pro gramming. Reform strategies in this area must therefore be made in concert with those in the social protection area. (We take up social protection for workers in the next chapter). Considering the existing laws, institutions, and actual practices, and in light of the international experience, priorities could include the following: ·Reducing excessive rigidity in the Labor Code. The new Labor Code appears to make important progress in this area by removing the union veto on dis missals and implementing advance notice and effective appeals procedures. Some progress also has been made in providing for more flexible hiring arrangements, especially with respect to fixed term contracting. More could still be done. Increasing flexibility in hiring and dismissals should bring more employment "out of the shadows," and international experience tells us that it should most help vulnerable segments of the workforce (for example, women and youths). It is true that these amendments will reduce formal job security and, as noted above, it is important that they be coupled with improvements in the social protection system for workers (see Chapter IV). ·Continuing to increase minimum wages. The current level plays little role in determining wage floors. Higher minimum wages (given the low base) are unlikely to have negative employment effects, and would reduce poverty among low wage workers. The level of minimum wage should not exceed a low share of average wage (for example, 25 to 30 percent) to ensure that work disincentives are prevented. However, the minimum wage will not be an effec tive policy instrument until the economy formalizes and enforcement improves. The linkage of the minimum wage to the subsistence minimum could lead to fiscal and incentive problems, particularly in low wage regions. Also, policymakers will need to consider how to accommodate the wide regional variations in labor markets and costs of living. ·Reducing the influence of tariff in wage setting. The tariff has been uncoupled from nonbudgetary sector wages; but its continued relevance as a wage setting guidepost is evidence of poor functioning of the labor market. As such it bears further investigation. The establishment of higher wages for particular areas, such as the North, is a legal requirement that is inconsistent with market prac tice and should be gradually phased out. ·Developing institutions to allow worker voice, improve work conditions, enforce contracts, and resolve disputes, thereby raising worker productivity. (a) Allow true worker and employer representation in unions, and eliminate management representation of workers, which would help improve work conditions. 102 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY (b) Consider decentralized bargaining approaches in collective bargaining, if the centralized approach is not yielding efficient bargaining outcomes. (c) Increase the resources available to the Federal Labor Inspectorate and build its capacity to provide technical assistance and advisory services to enterprises. (d) Establish alternative dispute resolution mechanisms based on professional third party mediation, conciliation, and arbitration services outside the court system. The development of competitive product markets should help contain wage pressures exerted by unions. The chapter finds that the debate over labor market reform in Russia is a con tentious one, but may offer a false choice. The debate divides those who want to see more social protection from those who want to see more labor market flexibility. In a sense this is a false choice: By instituting a more realistic and enforceable, flexible, formal, regulatory regime with a modernized safety net, the equity and efficiency con cerns of both groups could be alleviated. Achieving these outcomes will also require the development of a broad consensus regarding the need for and direction of labor market reforms. Chapter IV Safety Nets for Workers Older, less educated workers, with previous work experience and hence, obsolete skills, are the most vulnerable labor force participants in Russia. These workers con stitute the majority of the unemployed and, once unemployed, find it very difficult to find a job. As noted earlier, households headed by workers with wage arrears and the unemployed had much higher poverty rates than the national average. Most OECD countries establish unemployment protection programs to protect workers against the loss of income and skill as a result of unemployment. Russia has also established the two main programs found in OECD countries: (a) an unemployment benefit (passive) program, providing temporary cash assistance to the unemployed; and (b) Active Labor Market Programs (or ALMPs), including, among other things, training, job counseling, and public works. The objective of these programs is to prevent poverty among the unemployed and, by allowing workers to find a job better suited to their skills, to increase their productivity. This chapter discusses four aspects of these pro grams: financing, distribution, efficiency, and administration. It includes a brief review of social support programs used to facilitate restructuring in Russia. International experi ence in unemployment programs is also provided throughout the chapter. A. Background Russia has two public programs for assisting the unemployed: (a) a passive labor mar ket program that provides short term income replacement or unemployment bene fits in case of job loss; and (b) an ALMP that imparts training and other services to help the unemployed re enter the job market. In 1999, about 2.1 million individuals participated in both these programs. 131 The unemployment benefit program provides a sliding scale of benefits (expressed as a proportion of past wage) for a maximum duration of 12 months. Ben efits are subject to maximum and minimum thresholds. The unemployment benefit 131 Unemployment assistance programs have been the responsibility of the MLSD since 1996, after the dissolution of the self standing Employment Service. The MLSD central staff pro vides policy level oversight, while operations are conducted through a network of branches, including 99 regional and 2,444 local offices. These regional offices manage passive and ALMPs and offer a regionally differentiated program mix. In addition to these programs, redundancy measures to protect workers include the obligation by enterprises to provide prenotification of mass layoffs in agreement with their labor unions and to provide severance pay. 104 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY program is fairly typical of unemployment programs in developed countries, although the eligibility conditions of the Russian program are broader. (A detailed description of the Russian unemployment benefit program is provided in box IV.1.)132 ALMPs implemented in Russia include training, job creation (wage subsidy, and so on), public works schemes, counseling, and job information. The programs are administered through the federal employment service. The main purpose of these programs is to assist the unemployed to rejoin the work force. ALMPs in Russia are similar to those found in many developed countries around the world. Like other countries, the Government has used a combination of ALMPs and cash benefits to facilitate enterprise restructuring, most notably in the coal sector. B. Evaluation of Unemployment Protection Programs in Russia The following sections evaluate the unemployment protection program in Russia according to four main criteria: financing, distributive and efficiency impacts (and tradeoffs between these two program objectives), and program administration. Some issues to be addressed in each area include the following: · Protection vs. efficiency: A natural tension between protection and efficiency exists in theory and practice. Protection can be defined as the maintenance of living standards or protection from poverty. Efficiency can be defined in terms of the balance between benefits that are low enough to reduce work disincen tives and those that are high enough to allow workers to take enough time to find jobs in which they are the most productive. Too much protection can dampen efficiency objectives -- and reduce incentives to work, but programs that focus on efficiency alone can thwart protection and job search objectives. Therefore, program coverage has to be carefully reviewed and benefit levels carefully tempered to ensure that the program provides protection, encourages job search, and does not create unintended adverse work incentive effects. · Two other aspects of efficiency have to be considered. First, programs that pro vide explicit protection to laid off workers should separate job protection (financed by employers) from protection that is financed by the state. Second, programs that facilitate restructuring should lead to future productivity gains that offset short term costs. · Publicvs.privateroles:Employmentprogramsshouldbedesignedtoaddressthe rationale for public provision and financing. There is general agreement that the public provision of protection against unemployment can be used to pool risks over a large number of people more effectively than risk management using only 132 Not all countries have unemployment insurance based benefits as in Russia. Other pro grams followed worldwide include unemployment assistance programs (means tested or flat amount) that are supplementary to unemployment insurance accounts or stand alone pro grams and individual savings accounts. Salient characteristics of these programs are provided in table AIV.1. SAFETY NETS FOR WORKERS 105 Table IV.1. Employment Fund Budget (Percent of GDP) and Arrears (Millions of Rubles) 1994 1995 1996 1997 1998 1999 2000 Incomes 0.51 0.43 0.35 0.36 0.33 0.32 0.31 Expenditures 0.30 0.37 0.34 0.35 0.32 0.29 0.29* Surplus 0.21 0.06 0.01 0.01 0.01 0.02 0.02 (incomes over expenditures) Arrears 1542.5 2843.3 3661.3 1618.6 (as of 01.01 of each year) *0.25 percent of GDP was actual expenditures (table AIV.3). Source: MLSD. informal family and community networks. However, the public sector need not always be involved in the delivery of services. In the case of ALMPs, for example, public financing may help the low income unemployed obtain training, but pri vate firms may be the most effective in providing services. · Administration/financing: Do programs reflect the financial and administra tive constraints of the country? Programs that are complex to administer require monitoring (cross checking of information) and efficient information systems to be effective. If these do not exist, programs that require them have limited effectiveness. Program financing. Public expenditures on unemployment protection pro grams have been declining since 1998 (table IV.1). In 1999, the Government spent a total of 13 billion rubles on active and passive employment programs, amounting to 0.21 percent of GDP. This is consistent with Russia's level of income, but lower than program expenditures in advanced CEE countries (approximately 1.1 percent of GDP) and much lower than in OECD countries (average 2.4 percent) (table AIV.3). The expenditures trend in CEE countries is mixed. Unemployment benefit expen ditures as a share of GDP have declined in Hungary and Slovenia, are increasing in the Czech Republic, and are stable in Poland (Vodopivec, Wrgtter, and Raju 2000). Composition of expenditures. The composition of program expenditures has also been changing over time, with the share of expenditures on benefits increasing (table IV.2). In 1999, nearly 64 percent of total program expenditures went to passive pro grams (0.16 percent of GDP) and a much smaller share, 20 percent (0.05 percent of GDP), was devoted to ALMPs. The remaining expenditures were allocated to adminis trative expenses of both programs. The lower share of expenditures on ALMPs relative to unemployment benefits is typical of both CEE and OECD countries (table AIV.2). Large regional differences in composition of unemployment program expendi tures can be observed. The largest share of ALMP expenditures are in Moscow and Orel 106 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Box IV.1. The Unemployment Benefit System in Russia Russia's unemployment benefit system was established in 1991 to protect workers and other eligible labor force participants against loss of income and skill from layoffs and other causes of unemploy ment. Its main parameters are as follows: Eligibility. Every registered unemployed person in Russia is eligible for unemployment compen sation, including those who are laid off. As in other European countries, and transition countries, there are special groups who are eligible for benefits. Voluntary quits, new entrants and re entrants, and individuals near retirement are eligible for benefits. However, the number of eligible groups is quite large in Russia relative to other countries, and includes individuals dismissed for disciplinary problems, long term unemployed, dropouts of training programs, and re entrants to the labor mar ket. The law also contains employment quotas, for example, for persons with disabilities. Therefore, perhaps more than other systems, the Russian system does not distinguish between social assistance and standard unemployment benefits. For redundant workers and voluntary quits, benefits are a function of individuals' past wages. Specifically, eligible individuals are entitled to receive unemployment benefits equal to 75 percent of their wages for the first three months of unemployment, 60 percent of their wages for the next four months, and 45 percent of their wages for the next five months. The duration of benefits is 12 months (of the past 18). The benefit replacement rate and duration are on the high end for CEE countries (see annex IV). For redundant workers, there is a waiting period equal to the number of months of severance pay received. There is no such waiting period for voluntary quits (unlike prac tice in most CEE countries). Minimum and maximum. Unemployment compensation is subject to maximum and minimum constraints. The minimum benefit must be at least equal The minimum benefit must be at least equal to the level of 20 percent of the regional subsistence minimum but not less than Rbl. 100. The max imum benefit is the regional subsistence minimum. Other eligible categories receive 20 percent of the subsistence minimum. The duration of benefits (six months), and maximum (20% of subsistence) and minimum thresholds (minimum wage) are also lower for this group of workers. Privileges. Also specific to Russia, some workers receive greater privileges, such as higher com pensation for living in the North; or for having been subjected to the Chernobyl accident. In addition to cash benefits, dismissed workers can obtain other provisions, including housing, medical, and pre school services at former places of work. Work incentives. As in other CEE countries, the unemployment benefit program includes mech anisms to addresses moral hazard problems. Benefits can be terminated or suspended for some time for fraud and abuse, as well as lack of participation in training, refusal to participate in employment and what remains quite contentious public works after three months of unemployment. Financing Until 2001, when general revenue funding of benefits was introduced, this system was financed via a classic, OECD type, pay go, unemployment insurance system. It was financed through the collection of "contributions," of 1.5 percent of payroll (down from 2 percent in 1996) within each region. To allow consumption smoothing across the country, regions were supposed to send 20 per cent of their revenues to the federal level for distribution to deficit regions. in the Central region, and Amur in the Far East -- each with somewhat over 40 percent of their expenditures spent on ALMPs. By contrast, Nenets, Karelia, and Astragan directed less than 20 percent of their employment fund expenditures toward active programs (see table AIV.3). The regional financing of passive labor market programs is negatively correlated with ALMPs, suggesting tradeoffs between the two types of expenditures.133 One hypothesis explaining the disparate regional provision of ALMPs is that only regions 133 The correlation coefficient between expenditures on active and passive programs is 0.834 703 35. SAFETY NETS FOR WORKERS 107 Table IV.2. Share of Benefits in Total Unemployment Program Expenditures Share of benefits for the unemployed in: Employment Mandatory insurance Aggregate expenditures Fund revenues contributions by employers of the Employment Fund 1993 6.0 6.7 10.3 1994 13.3 18.3 17.8 1995 26.8 38.9 29.2 1996 45,7 58.6 47.6 1997 54.4 63.0 56.8 1998 56.0 63.7 58.3 1999* 55.3 60.2 59.6 * Nine months. Source: Data on formation and spending of the Employment Fund from the MLSD. have funding above and beyond what is required to pay unemployment benefit can offer these programs. Another explanation is that some regions opt for ALMPs rather than unemployment benefits in order to reduce `benefit dependency. One reason advanced for regional inequities in financing of unemployment pro grams is that, prior to 2001, contributions were decentralized to the regions. Contri butions for unemployment were collected at the regional level, and the regions were responsible for passing on 20 percent of their contributions to the center and paying out benefits. This amount was not always obtained, and insufficient revenues were distributed to tax poor regions134 For this reason, starting 2001, benefits are based on general revenue financing. It is expected that this change will improve financing of benefits and regional incidence of arrears. There are three caveats. Our analysis finds that the distribution of contribution arrears across regions in 1999 was virtually uncorrelated with the distribution of payment arrears to the unemployed.135 Thus, the hypothesis that benefit arrears were related to local revenue problems is not substantiated. It remains to be seen whether federal financing of unemployment benefits will reduce the level and inequality in arrears. General revenue financing has not reduced arrears or regional disparities in the child allowance program.136 Finally, regions may tend to 134 By contrast, social insurance and health insurance will be funded by a combined social tax. 135 Correlation coefficient is 0.166 177 887. 136 As noted earlier, per worker contribution arrears by region normalized for average wage rates are only weakly (0.18) correlated with per worker wage arrears normalized for average wage rates. 108 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.3. Replacement Rate of Unemployment Benefit Ratio of unemployment benefits to: Subsistence minimum for Average wage Average per capita income working age persons 1993 27.7 9.7 12.6 1994 47.0 18.5 19.7 1995 36.3 20.3 18.6 1996 42.4 19.8 20.6 1997 61.8 26.7 27.3 1998 68.3 32.1 34.7 1999 40.2 25.5 16.7 Sources: Calculations based on 1999 data from the MLSD and Goskomstat (1999, pp.20 21). overstate their budgetary needs for ALMPs and Benefits because they are no longer responsible for financing these programs. Thus, regional financing and auditing mechanisms will have to be introduced so that regional arrears and inequities are minimized. The main reason for limited and uncertain financing does not appear to be insuf ficient budgetary resources. Rather, social protection against unemployment in Rus sia has more often taken the form of employment guarantees and enterprise delivery of social benefits and services. In some transition countries, the share of resources going to subsidies ranges between 2 and 3 percent of GDP.137 The Government has therefore given less attention to the adequate and timely payment of unemployment benefits. In addition, social spending has been plagued by misallocation of resources. A reallocation of expenditures from nontargeted spending could help improve the financial sustainability of the program. For example, if SIF (Social Insurance Fund) recreational benefits not related to poverty (sanatoriums and resort vouchers) had not been in effect in 1999 and 2000, these savings could have been reallocated to employment programs (table AIV.4). In that case, 0.52 percent and 0.46 percent of GDP would have been spent on unemployment benefits. Distributive impact: Adequacy of support. The Russian unemployment benefit program has a system (legal) benefit replacement rate that is similar to or even higher than many CEE and OECD countries (table AIV.1/box IV.1). However, inadequate program financing has made the effective replacement rate (the ratio of benefit to average wage) of benefits much lower than the system rate. Aside from arrears in financing, other factors such as low reported wage or workers, or contribution record 137 Riboud, Sanchez Paramo, and Silva Jauregui (2001). SAFETY NETS FOR WORKERS 109 Table IV.4. Minimum Unemployment Benefit Unemployment Ratio of the minimum unemployment benefit to: benefit recipients (percent) receiving minimum Subsistence mini Average wage Average per capita benefit mum for working income (percentage) age persons 1993 -- 26.9 10.4 13.5 1994 -- 20.3 8.0 8.5 1995 49.2 16.1 9.0 8.2 1996 47.3 19.7 9.2 9.6 1997 47.0 20.3 8.8 9.0 1998 47.6 16.9 7.9 8.6 1999 48.1 8.3 5.3 3.5 Note: Before 1995 no such indicator had been calculated. Source: Calculations based on 1999 data from MLSD and Goskomstat (1999, pp.20 21). of workers, could also explain the low level of benefit. The (expected) benefit replacement rate was 25.5 percent of the average wage in 1999 (about 40 percent of the minimum living standard)138 (table IV.3). With double digit inflation and benefit arrears, the average replacement rate has varied from month to month. For example, the replacement rate for the first quarter of 2000 was reduced to 22 percent of the average wage. The replacement rate for benefits has also declined in CEE transition countries, but in these countries benefits are not subject to arrears. The reported benefit level also includes arrears paid to unemployed as well as ben efits paid as a result of court decisions in favor of benefit claimants. (Tchetvernina 2000) Therefore, the adequacy of benefits (share of benefit as proportion of average wage) in Russia is likely to be much lower (and with a lower expected value) than implied by the reported replacement rate. The unemployment benefit structure is currently quite flat. In 1999, about half of all unemployed received the minimum benefit (about 5.3 percent of average wage). The actual paid maximum benefit was also much lower than the legislated maximum and is about 56 percent of the average wage (table IV.4). Actual replacement rates in CEE transition countries are in the same range as in Rus sia (table AIV.5), but average replacement rates in OECD countries are much higher -- 138 The share of benefits in total consumption, a measure of benefit adequacy, obtained from RLMS 1998 household survey data was surprisingly high. However, very few households report receiving benefits. In some cases, results are based on fewer than 10 observations -- and results are not robust (table AIV.12). 110 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY about 60 percent.139 However, the benefit replacement rates for most CEE countries have fallen over the past decade, as has the maximum duration of benefit, as countries have tightened programs in response to fiscal problems.(Figures AIV.1 and AIV.2). Coverage. In August 2000, the unemployment rate based on the RLFS was 10.1 percent, while the registered unemployment rate was 1.0 percent140. This difference is substantial. In other words, out of an estimated 7.1 million unemployed persons, less than a million, or 14 percent, were registered with the employment offices (table IV.2 above).141 Most registered unemployed are women, perhaps reflecting the lower opportunity cost of women for applying for benefit. In contrast, as noted earlier in Chapter I, women constitute roughly half of the survey unemployed. The coverage of the unemployment benefit program, or share of benefit recipients to survey unemployed (80 percent of registered unemployed receive benefit), is much lower than for similar programs in CEE countries (tables AIV.6 and AIV.7). In CEE countries, regis tered unemployment rates are much closer to survey rates. In OECD countries, a much higher share of survey unemployed (versus Russia) receives benefits. For example, benefici aries equal 82 percent of the unemployed in Australia and 89 percent in Germany. In some OECD countries such as Austria and the Netherlands, and CEE countries such as Slovenia, the number of registered unemployed is actually higher than the number of unemployed. There is considerable variation among OECD countries. The lowest registration ratios are in Greece (30 percent), Japan (36 percent), and the United States (34 percent). Program coverage in Russia has declined since 1996, after increasing until 1995 (table IV.5). Several factors may be responsible for this development. Prior to 1998, program applications likely declined because of low and uncertain benefits. For the same reason, employment offices may have reduced the share of applicants granted unemployment status. Since 1998, improvement in economic conditions, coupled with tightened eligibility conditions introduced in 1999, are likely to have further lowered the rate of application, as well as registration. A calculation of a generosity index of unemployment benefits for Russia, which takes into account both the replacement and coverage rates shows that the country has a much lower generosity index (3.0) than the average for CEE (12.1) or OECD (26.3) countries (Vroman 2001).142 139 The range is quite substantial even within OECD countries. New Zealand and Australia and Ireland have replacement rates of approximately 30 percent, while Sweden and Denmark have replacement rates of 80 percent of average wage and above. Source: OECD Employment Outlook (1995, 1999); IMF World Economic Outlook (1999). 140 The coverage rate is the number of registered unemployed relative to survey unemployed. The number of registered unemployed is used to assess coverage of both ALMPs and Unemployment benefit programs. However, most registered unemployed (over 80 percent in 2000) receive benefit. 141 Household budget survey data confirm the low coverage found in administrative data. According to the 1998 RLMS, less than 1 percent of all households receive unemployment ben efit (as compared with 8 percent for child allowances and 37 percent for old age pension). Table AIV.11 household data (using a different definition of the poor and therefore not strictly com parable) for CEE countries indicate that coverage in those countries is generally higher than in Russia (the exception is the Slovak Republic). 142 The generosity index is a product of the coverage of the program and its replacement rate multiplied by 100 (Vroman 2001). SAFETY NETS FOR WORKERS 111 Table IV.5. Trends in Applicants and Registered Unemployed Number of applicants. Share Number Share of of nonworking of applicants registered Total Jobless applicants registered as unemployed (percent) unemployed among nonworking applicants (percent) 1992 2.3 2.3 100 1.0 43.5 1993 2.1 2.1 100 1.4 70.0 1994 3.2 2.8 88.8 2.5 78.1 1995 5.1 3.9 76.8 3.3 82.9 1996 5.3 4.4 83.6 3.5 78.9 1977 4.6 3.8 82.8 2.8 73.6 1998 4.7 3.8 80.9 2.7 70.0 1999 4.3 3.3 76.7 2.1 63.6 Source: Annual reports "Data on job placement of applicants to employment offices of the Russian Fed eration". The data for January -- December of each year are presented in the T 2 (Job placement) Forms of the MLSD. There is considerable variation in regional coverage (registration rates) in Russia. In 1999, seven regions had a 30 percent or better registration rate, while 10 regions, including St. Petersburg, had less than a 10 percent registration rate (table IV.4). Regression results show that the regional registration rates vary positively with the regional variations in the RLFS unemployment rates (table AIV.8). Regions with higher survey unemployment rates tend to have higher coverage rates. However, the RLFS unemployment rate explains only a small portion of the regional variation in registration. Other factors are significant as well. Registration is lower in areas in which unemployment benefits are smaller relative to average incomes. Benefit gen erosity is therefore an important factor in explaining regional variation in coverage rates and low rates of registration overall. A positive link between benefit generosity and the registered unemployed rate has also been found for CEE transition countries (Vodopivec and Raju 2001) After accounting for survey rates and benefit levels, the registered unemployment rate also varies positively with the ratio of Employment Service staff per unemployed person. Thus, the staffing of the program has an important impact on its ability to attract and serve clients, independent of the level of benefit. Broad regional group ings account for additional variation in registered unemployment rates. In particular, regional registered unemployment rates are higher in the North, the Volga Vyatka region, and Eastern Siberia. After adjusting for other factors, the coverage is lower in the Northern Caucasus than in the Central region. 112 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Consumption smoothing. The low level of benefits and the high uncertainty in their payment means that unemployment benefits are likely to provide negligible consumption smoothing. A recent study by Richter (2000) shows that Russians tend to consume less from benefits with large arrears (such as child allowances) because they are considered transitory income, and consume more from more permanent, or less arrears prone benefits, such as pensions. This finding suggests that unemploy ment benefits, which are also subject to considerable arrears, are not likely to have contributed to consumption smoothing in Russia. If financed adequately, unemployment insurance programs can be successful in smoothing consumption. In the United States, studies show that the income levels of recipients were only 3 to 8 percent lower than those of nonrecipients with similar characteristics and that, without these benefits, consumption would have fallen sub stantially.143 The unemployment insurance system also performs well under idiosyn cratic sectoral and regional shocks. It can also act as a stabilizer in times of recessions, reducing the magnitude of the downturn. By inference, a well implemented unem ployment benefit system could have positive effects in smoothing consumption in Russia. Poverty alleviation. The 1998 RLMS data indicate that unemployment benefits have a poverty reduction impact among household that receive them, but the small number of individuals covered by the program means that the overall poverty reduc tion impact is very small. Without unemployment benefits, the poverty rate among the few households with children that received unemployment benefits would have increased by 8 percentage points to 83 percent. Furthermore, the poverty gap for recipient families was reduced from 40 percent to 25 percent. While families who actually receive benefits are among the poorest, and tend to stay poor, unemploy ment insurance still has had a positive impact on their welfare (table AIV.9). These results should be interpreted with caution, however, because of the small number of observations in the data set. The poverty reduction of unemployment benefit programs in CEE countries is mild and varies considerably (table IV.6). While strict comparison with Russian results is not possible (table IV. 6 uses a different poverty line), the results show that outside of Poland and Hungary and Slovenia -- where programs are more generous, poverty was reduced by only 2 percent for the countries studied. The small impact is the result of the insignificant share of poor individuals eligible for the program. Among those poor who were eligible, unemployment benefits accounted for almost a third of all income in Poland, and between a third and a quarter in Hungary and Slovenia. How ever, unemployment formed a small share of income in Bulgaria, Estonia, and the Slo vak Republic.144 143 Hamermesh and Sleznick (1995) and Gruber (1997), cited in Vodopivec and Raju (2001). 144 Lokshin and Ravallion (2000) suggests that safety nets helped individuals cope with the 1998 crisis. The incidence of poverty was reduced relative to that which would have been obtained if such public programs (even excluding pensions) did not exist. However, slightly greater funding would have helped to reduce poverty more significantly. Richter (2000) finds that keeping trans fer levels at the 1994 levels would have reduced poverty by 10 percent in 1998. However, neither study isolates the impact of unemployment benefits on poverty alleviation. SAFETY NETS FOR WORKERS 113 Table IV.6. Poverty Impact of Unemployment Programs in Select Tran sition Economies, Mid 1990s (Percentage) Bulgaria Estonia Hungary Latvia Poland Slovak Slovenia Rep. Poverty 1.1 0.5 14.8 2.2 16.7 2.7 6.8 reduction145 Targeting146 17.4 31.1 4.9 12.4 6.8 0.5 16.0 Source:Vodopivec and Dhushyanth (2001). Of course, limited poverty reduction offered by unemployment benefit programs is not very surprising if unemployment programs are designed to smooth consump tion and not to alleviate poverty. That is to say, earnings related benefits are designed to help maintain living standards during temporary periods of income loss, rather than to provide benefits targeted to the poor. Incidence of public expenditures. In Russia, in 1998, approximately 26 percent of total spending for the unemployed accrued to the bottom 20 percent of the popula tion, indicating that the poor receive a share of transfers that is slightly higher than their share in the population (Table AIV.14). (Once again, the small number of obser vations means that this evidence should be interpreted with extreme caution.) Evidence from transition countries (with the same caveat on comparability of results as above) also suggests that -- other than in Estonia -- the poor do not receive the largest share of spending on unemployment benefits (table IV.6). While the share of unemployment benefits collected by the richest quintile exceeds the share col lected by the poorest quintile in quite a few countries, the overall effects are neutral or may be progressive, because unemployment insurance contribution rates are earn ings related. Unemployment benefits are also not an important tool for income redis tribution in developed countries. The effects of benefits are progressive in about half of the OECD countries, and neutral in the other half (Vodopivec, Wrgtter, and Raju 2000, table AIV.16). This is perhaps not surprising. Unemployment programs are not by definition pro poor, as benefits are proportional to past income. These are prima rily consumption smoothing programs. There is little empirical evidence on the distributive impact of other types of unemployment benefit programs. Available evidence, summarized in table AIV.17, suggests that the distributive impact differs by type of program. Evidence from Argentina suggests that public works are also very pro poor. Training programs and public works are progressively distributed, while individual savings account (ISA) programs are regressively distributed (in Columbia). Severance pay, which is available only to formal sector workers, seems to increase the advantage of formal sector 145 Change in poverty headcount brought about by unemployment benefit receipt, in per cent). Poor are defined as individuals with consumption less than 50 percent of median. 146 The share of unemployment benefit received by the poor, in percent. 114 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY workers versus those working informally, limiting any distributive impact (De Fer ranti, Perry, and Serven 2000). Further evidence from Vroman (2001) indicates that the Australian means tested unemployment assistance system is very progressive, perhaps because these benefits are not contribution and wage related. Roughly 70 percent of cash benefits are paid to those in the bottom three deciles of the income distribution.147 Efficiency impact: Unemployment benefit programs. Because benefits provided by the unemployment program in the Russian Federation are largely negli gible, adverse work incentive effects are small, if any. Adverse incentive effects imposed by unemployment benefit systems can be quite significant, however, in countries in which actual benefits are more generous. In OECD countries, unemployment benefits reduce the probability of recipients leaving unemployment to take up employment -- leading to an unemployment trap.148 These negative effects of duration of benefit on probability of exit from unemployment to employment have also been found for CEE countries (Bulgaria, Czech Republic, Estonia, Hungary, Poland, Slovak Republic, and Slovenia), although the effects of replacement rate are much less pronounced. For these reasons, recom mendations have often been provided to tighten program benefit and eligibility con ditions so that the efficiency costs of the programs do not outweigh their consump tion smoothing impact (World Bank 2001a).Interestingly, there does not seem to be much impact of unemployment benefit on increasing intensity of job search, improv ing job matches, or entry into regular jobs. Nonetheless, the incentive impacts of unemployment benefit schemes are not all the same. For example, work disincentive effects will differ in terms of unemploy ment insurance and means tested unemployment benefits. Disincentive effects may be lowest in ISAs, because individuals receive benefits that represent a return on their own contributions, unless contributor entitlements exceed their ISA balances. The ability of countries to reduce adverse work incentives also depends on administrative capacities as well as benefit levels (see administrative section below). The efficiency aspects of unemployment benefit programs are provided in table AIV.18. Has the lack of a well funded unemployment benefit system in Russia reduced managers' incentives to lay off workers, impeding economic efficiency? Haltiwanger and Singh (1999) also show that a generous compensation has helped facilitate down sizing in other countries. In Russia, clearly the restructuring of the coal sector might not have taken place without a generous benefit package, including severance pay, back wages, and unemployment benefit (see below). There is considerable evidence from the United States that the availability of benefits strongly increases the probabil ity of temporary (rather than permanent) layoffs.149 According to a theoretical model 147 Vroman (2001) uses 1995 data to show that the bottom three deciles receive total trans fers expenditures equal to 20.8 percent in Italy and 58.0 percent in Australia. The top three deciles in Australia received 7.4 percent of transfers, the lowest percentage among all the coun tries studied. 148 This section draws on Vodopivec, Worgotter, and Raju (2000). 149 Clark and Summers (1982), Feldstein (1978), and Topel (1983), referenced in Vodopivec and Raju (2001). SAFETY NETS FOR WORKERS 115 Table IV.7. International Assessment of Unemployment Benefit Programs Scheme Financing Strengths and weaknesses Unemployment Payroll tax Allows risk pooling; and provides consump insurance tion smoothing. Performs well where labor market institutions encourage flexibility, informal sector is small, and there is strong administrative capacity to monitor program and control incentives. Benefits/taxes must be kept low to avoid adverse incentive effects. Unemployment General revenues Means tested benefits: Is very progressive but assistance may have disincentive effects similar to unem ployment insurance if benefits are too high. Where means tested requires strong adminis trative and monitoring capacity and low infor mal sector. Flat benefit: Regressive; potential to work well in countries where administrative capacity is weak and informal sector activity is high Duration of benefit and replacement rate should be set low to avoid adverse incentive effects. Individual Savings Worker contributions Works well in low or middle income coun Accounts tries. Avoids disincentives to work, has good self monitoring features, but does not cover poorer and/or informal sector workers. Largely untested. Severance payments Financed by firm Unfavorable option; strong negative effi ciency effects limits hires; limited risk pool ing; politically contentious. Public works General revenues Can reach informal sector workers and poor for income support. Where administrative capacity is weak entails large non labor costs; is often tem porary in nature; and does not help increase wage or employment prospects (see ALMP sec tion below). : Betcherman (2000). proposed by Blanchard (1997), a generous unemployment benefit that raises the cost of new hires (through a higher payroll tax rate) could reduce the impact of labor real location and growth by dampening job creation. However, if program benefits have been tightened over time, as in the case of CEE countries, these initial adverse effects may well decline. Forteza and Rama (2000) find that higher mandated benefits do not impede recovery, once economic reforms have been implemented. In Russia, economic restructuring has occurred despite the lack of an effective unemployment benefit program. However, in strategic sectors, such as coal, it has required a very generous severance package to lay off workers. Given that employ ment declines have not been as great as output declines might dictate, and there is evidence of surplus labor, particularly in certain regions and some industries, and fur 116 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ther restructuring may be required, and an effective unemployment package may help facilitate restructuring. Thus, in summary, in addition to allowing consumption smoothing, a well funded unemployment benefit system would protect workers than the current system ­ a goal well worth achieving. While the evidence on the efficiency impact of unemploy ment benefits is mixed, a well funded unemployment may well increase economic efficiency by facilitating a more efficient allocation of labor and facilitating layoffs. Another advantage of an unemployment benefit system could be to complement more flexibility in the Labor Code, by transferring social protection activities from firms to the public sector. However, the generosity of such a system should be such that it does not dampen job creation and reduce incentives to work. As noted above, serious work disincentive effects and the persistence of unemployment can result from to generous a programs. A summary of the international assessment of unem ployment benefit programs, from both efficiency and distribution aspects, is pre sented in table IV.7. Efficiency impact: ALMPs. The efficiency of Russian ALMPs is not known, that is, whether they have the ability to provide program participants with a job or higher wages (as compared with a program in which they do not participate). Rigorous pro gram evaluation, which identifies impacts on program participants relative to a con trol group of nonparticipants (with roughly same characteristics), has not, as yet, been conducted in Russia. The current Government places a strong emphasis on such eval uation, however, and has recently initiated the collection of administrative data to assist in program evaluation. While better administrative data are useful, full program evaluations require control groups and microdata files to derive meaningful results. The Government also proposes to introduce profiling of the unemployed to identify individuals at the greatest risk of long term unemployment, and matching them with the ALMP programs that would most improve their chances of finding a job. Impact evaluation of programs. The quantitative evaluation of ALMPs that has been carried out in selected OECD and transition countries has yielded interesting results. A brief review of the impact assessment of each program is presented in table IV.8 (Dar and Tzannatos 1999).150. The research suggests that ALMPs require substan tial administrative capacity in terms of design and implementation and can be rela tively expensive to be effective. Evaluations of their impact on employment prospects and wages have been mixed at best. Specifically, most training and retraining programs tend to be no more effective than job search assistance. These studies also show that public works programs do not necessarily lead to continued employment but can be rationalized on the basis of community development or as antipoverty measures. In some countries, they can also provide employment to informal sector workers. Some evidence from a recent evaluation of ALMPs in the Czech Republic, Hungary, Poland, and Turkey indicates that, while the results differ in some ways, OECD coun try findings are generally supported in transition economies as well (Fretwell, Benus, 150 Dar and Tzannatos (1999) also discuss methodological issues relating to the evaluation of ALMPs. In their summary assessment, they give particular emphasis to scientific approaches where outcomes for program participants are compared with outcomes from a control group. SAFETY NETS FOR WORKERS 117 and O'Leary 1999). The implications of this research are that the aggregate gains from ALMPs tend to be modest at best, but that targeted groups can potentially benefit from certain types of well designed programs, albeit at high cost. Given that these evaluations are based on the experience in advanced industrialized and CEE transi tion economies, their applicability to a country like Russia is an open question, and can only be answered once evaluations are completed. A rigorous evaluation of ALMPs has to address the following questions: ·Deadweight loss: Are outcomes no different than they would have been without the program? ·Substitution effect: Do workers in subsidized jobs just substitute for other unsub sidized workers that would have been hired anyway? ·Displacement: Do firms with subsidized workers take business away from other unsubsidized firms? Administrative data. Despite the absence of program impact evaluation, some evidence on the performance of Russian ALMPs, however inconclusive, can be gleaned from administrative data. The data suggest two trends (table AIV.3). First, the share of program resources spent on job creation/preservation programs has decreased, while the share of resources spent on training and public works has increased. The decline in job preservation/creation expenditures is likely to be a positive development. Job preservation programs provide funding for job mainte nance, while job creation programs directly provide 12 month average salaries for workers. Tchetvernina (2000) finds that these programs are costly for employers. Enter prise managers estimate costs of creating one job as one year of payroll per worker. Such high costs can have the potentially harmful effect of preserving obsolete jobs and/or creating nonviable employment. Nevertheless, despite declining, these pro grams still remain substantial. In 1999, employment offices helped preserve and cre ate 50,000 jobs, and seven federal targeted programs resulted in the creation of another 250,000 jobs. Indeed, spending on job subsidies is much higher in Russia than in both OECD and CEE countries (table AIV.19). Second, administrative data show that out of the total resources spent on ALMPs in 1998, 38 percent was spent on job creation and another 41 percent was directed toward training and retraining. The remainder was spent on early retirement and other programs, including public works (table IV.9). The least costly program on a per recip ient basis was public works; while job creation/training programs were the most expensive. These finding are consistent with those frequently observed in OECD coun tries. Thus, Russia spends the majority of resources on high cost programs that have been found internationally to be no more effective than job search assistance. Efforts to improve cost effective, job information services have been initiated in Russia, and the availability of regional, job information databases is increasing. In 1999, 40 regional employment services contributed to establishing the job bank, and 70 regions used these data compared with only 50 in 1998. The job bank contained information on 10,000 vacancies compared with 2,700 vacancies in 1998. Informa 118 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.8. Effectiveness of Active Labor Market Programs: Interna tional Evidence Program Appears to Help Comments 1. Job search Adult unemployed Relatively more cost effective than other labor market assistance/ generally when economic interventions (for example, training) -- mainly employment conditions are improving; because of the lower cost; youths do not benefit services (19) women may benefit usually. Difficulty lies in deciding who needs help more. in order to minimize deadweight loss. 2. Training of Women and other No more effective than job search assistance in long term disadvantaged groups. increasing reemployment probabilities and postin unemployed (28) tervention earnings, and is two to four times more costly. 3. Retraining in Little positive impact -- No more effective than job search assistance and sig the case of mass mainly when economy nificantly more expensive. Rate of return on these layoffs (12) is doing better. programs usually negative. 4. Training for No positive impact. Employment/earnings prospects not improved as a youths (7 ) result of going through the training. Taking costs into account, the real rate of return of these programs is negative. 5. Employment/ Long term unemployed High deadweight and substitution effects. Impact wage subsidies in providing an entry analysis shows treatment group does not do well (22) into the labor force. as compared with control. Sometimes used by firms as a permanent subsidy program. 6. Public works Severely disadvantaged Long term employment prospects not helped: pro programs (17) groups in providing gram participants are less likely to be employed in a temporary employment normal job and earn less than do individuals in the and a safety net. control group. Not cost effective if objective is to get people into gainful employment. 7. Microenterprise Relatively older groups, Very low take up rate among unemployed. Significant development the more educated. failure rate of small businesses. High deadweight and programs (15) displacement effects. High costs -- cost benefit analy ses rarely conducted but sometimes show costs to unemployment insurance budget higher than for control group. Source: Dar and Tzannatos (1999). tion on open vacancies was reported by about 4,000 employers, compared with 1,300 in 1998. However, a recent evaluation of job bank database use indicates that, while the data bank is promising for intraregional data sharing, the nationwide exchange of job vacancy information is limited in practice and needs to be improved (Tchetvern ina 2000). Table IV.10 shows that the declining trend in registered unemployed was mainly among voluntary quits or laid off workers and secondary or vocational graduates -- another important development in the composition of ALMP clients. The number of students, pensioners, and workers in search of secondary employment among regis SAFETY NETS FOR WORKERS 119 Box IV. 2, ALMPs in Russia: A Brief Overview A large range of ALMPs were established in Russia in the early 1990s. This box focuses on only the main programs: Job Creation/Preservation. Job preservation/creation (financial support) were the main ALMPs (in terms of expenditure share) until the mid 1990s. Over time, the emphasis of ALMPs has shifted to vocational guidance and training. The move away from these programs reflects both financial and economic considerations. The programs are very costly to employers; the use of funds is difficult to monitor, the substitution and displacement impacts of these programs are not well known. Many aspects of these programs were modified from 1995 onward. However, these programs still remained important in 1999, comprising more than 40 percent of total expendi tures. In 1999 employment offices helped preserve and create 50,000 jobs, while seven federal tar geted programs created an additional 250,000 jobs. However, no rigorous scientific evaluation of these or other ALMPs (listed below) has yet been completed. Vocational training. The share of registered unemployed referred for vocational training has increased. Referrals tend to be younger individuals (aged 16 29) and women. In some cases, local employers, who receive remuneration for this effort, provide training. The placement rate of trainees has increased over time; and their re registration rate has decreased. By the end of the 1990s, the share of those placed after training exceeded 90 percent. But this statistic is biased. Often, training is only offered to those unemployed who provide written guarantees of employ ment after completion of the course. In regions that have had to give up this practice because of litigation, the placement rate is much lower (40 percent). Vocational guidance. Vocational guidance includes providing information, vocational coun seling services, and psychological support to the unemployed. The employment offices have pro vided these services only since the mid 1990s. Over time, both unemployed and employers par ticipating in the program have increased. Interestingly, the share of unemployed receiving vocational guidance has declined. Anecdotal evidence suggests vocational guidance and consult ing services help individuals find jobs, although exact numbers are not available. Early retirement pensions. As in some CEE countries, the duration of unemployment benefits is extended for workers with length of service that entitles them to old age pensions. The maxi mum duration of benefit is 24 months (in 36 months). Pensions are paid from the pension fund, and reimbursed from the Employment Fund (now general revenue). A large proportion of women and long term unemployed participate in the program. The share of retired beneficiaries among the total registered unemployed has increased over time. However, the number of retirees has been declining. Only 1 percent of early retirees received benefits in 1998 1999, down from 4 percent in 1995. However, a growing share of pre pension age unemployed is participating in public works, and some are being placed in jobs through employment quotas. Targeting vulnerable unemployed. In the late 1990s, ALMPs were developed to help specific categories of unemployed, such as young graduates of secondary vocational schools and long term unemployed, acquire job seeking skills. These programs include, for example, job clubs, the New Start program and the Youth Practice program. The latter two programs provide youths with job seeking skills (writing CVs, calling employers) and practical training (with wages paid by both employers and the employment service). Job placement after participation in Job Seekers Clubs (66 percent) is high, but quite low in both Youth Practice (33 percent) and New Start. Public works. The objective of public works is to help the long tem unemployed maintain their skills and work habits. A large share of women and long term unemployed participate in these programs. Participants are paid low wages (about 30 to 40 percent of the wage of per manent employees and according to their qualifications; table IV.8). Public works are manda tory for certain categories of the unemployed. A previous experiment to make public works participants give up their registered unemployment status and receive a wage equal to benefit led to a reduction in program participation. Source: Tchetvernina (2000). 120 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.9. Employment Fund Expenditures on Active Policies, 1992 99 1993 1994 1995 1996 1997 1998 1999* Vocational training, vocational 8.5 19.4 24.9 36.5 42.8 45.8 47.1 guidance Public works 1.9 3.9 7.7 16.8 17.5 15.0 9.2 Social adjustment 0 0 0 3.0 6.5 2.8 2.5 Job creation/preservation 88.7 75.1 66.7 43.7 31.9 35.6 40.6 R&D 0.9 1.5 0.6 0 1.2 0.9 0.6 * Nine months. Source: Data from the MLSD (Tchetvernina 2000). tered unemployed actually increased between 1997 and 1999. Recently initiated tar geted programs to assist the long term unemployed tend to be focused on youths (see box IV. 2). ALMPs have shifted to younger workers rather than older, experienced, and less educated workers who comprise the majority of the long term unemployed. This is a mixed blessing. The success rate of ALMPs with younger workers may be higher, mak ing programs more cost effective; but the program is not targeting older workers, who have the most difficult time getting jobs. Finally, administrative data on program effectiveness include placement rates of specific programs (table IV.11). These data suggest that placement rates for training graduates have increased over time. But this result must be interpreted with skepti cism because many trainees are already guaranteed a job prior to joining a program, which biases this statistic. In regions where a job guarantee is not a prerequisite for obtaining training, placement rates are in fact quite low. Therefore, administrative statistics on program impact are not a useful guide to the success of ALMPs in Russia A beneficiary survey covering only four Russian regions found that 70 percent of those served by the training centers rated them "very useful" (World Bank 2000). Employment services also appear to have been very useful in assisting coal sector workers learn about their eligibility to receive benefits and services (see below). But since these smaller surveys are inconclusive, a more rigorous evaluation of ALMPs would be the best means to assess the efficiency of ALMPs. Lastly, the Russian Employment Service does not contract services to private providers. Private service providers appear to have been growing over time, including ones with firms as clients and ones that accept applications from individuals. Private services are largely concentrated in Moscow, which has higher average wages and more foreign enterprises than any other part of the country. These private providers appear to be concentrated in providing jobs by "head hunting" for highly qualified professionals for private firms. As private sector providers increase in number, they could be used by employment services to contract out the provision of particular services. Profiling. Employment Office staff do profile workers in order to discuss their suit ability for particular programs. However, the Government is currently considering SAFETY NETS FOR WORKERS 121 Table IV.10. The Socioeconomic Characteristics of Applicants, 1993 99 (Percent of the Previous Year) Reason for separation 1995/ 1996/ 1997/ 1998/ 1999/ 1995/ 1999/ 1994 1995 1996 1997 1998 1992 1996 Persons having paid 136 88 69 105 92 177 75 employment Students 377 84 97 113 117 1327 128 Pensioners 100 117 106 103 85 313 92 Teenagers -- 85 90 111 112 -- 111 Based on: Annual reports "Data on job placement of applicants to employment offices of the Russian Federation". The data for January -- December of each year are presented in the T 2 (Job placement) Forms of the MLSD (Tchetvernina 2000). more empirical profiling methods to both identify which unemployed are susceptible to long term unemployment and to match them to ALMP programs that will best improve their employment prospects. This type of profiling is used in many OECD countries (box IV.3). There are two main methods of profiling the unemployed: (a) statistical model based programs that use multivariate regression techniques to iden tify individuals most vulnerable to long term unemployment, and (b) characteristics based programs that mainly rely on the judgment of case workers. A recent evaluation of profiling programs in OECD countries, notably the United States, Canada, Australia, and the United Kingdom, finds that profiling based on sta tistical models or judgmental "characteristics screening" by staff can result in two types of error (OECD 1998).151 The first type, errors of exclusion, fails to identify vul nerable groups. The second type, errors of inclusion, identifies nonvulnerable groups as vulnerable and includes them in the program. Although the predictive powers of statistical estimation are found to be quite low, evidence from the United States sug gests that a model based approach leads to better selection of groups at risk than characteristics screening. Countries are now refining their models, adding more explanatory variables, and using more available data. Certain tradeoffs inherent in the model based approach are important from a pol icy perspective. If the probability threshold for defining groups at risk is to be kept low (to include many unemployed), to ensure that all groups have access to ALMPs, the risk of individuals being excluded will certainly be reduced, but inclusion error may be increased. On the other hand, if ALMPs are to be offered to all at risk workers, the prob ability threshold for being at risk could be kept high. While a high probability thresh old would reduce inclusion errors, it would increase the risk of exclusion. In this case, judgmental screening could be introduced to capture excluded workers, although these procedures would have to be used judiciously. In addition, the report finds that: The probability values assigned to the unemployed should not be the only method used to determine appropriate program placement. More in depth analysis is needed. 151 OECD (1998). 122 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.11. Job Placements of Unemployed Completing Training Pro grams, 1993 99 Percent of individuals Number Percentage placed after training of those not placed of those reregistered 1993 64.3 28,964 40.8 1994 59.2 65,713 -- 1995 62.9 99,131 43.7 1996 74.3 59,911 50.0 1997 82.9 31,129 28.8 1998 90.0 22,722 34.2 1999 91.2 21, 331 28.6 Source: MLSD; Tchetvernina (2000). There is as yet no evidence on the macroeconomic impact of profiling, or whether the substitution or displacement effects, would offset any microeconomic impacts. However, a full evaluation of job profiling and its cost effectiveness has not yet been undertaken. The only evidence available is on the cost effectiveness of ALMPs. A spe cific assessment of profiling would require the development of new evaluation meth ods and an extended follow up period. 4.55 In other words, while profiling is an interesting option for identifying at risk groups -- and scientific evaluations work best -- its cost effectiveness is not yet well known. This is an area in which further work is required. Any experiment in Russia on profiling should be undertaken with a built in evaluation so that cost effectiveness and program impacts can be assessed. The implementation of profiling may require greater sophistication and financing than is possible given the administrative capac ity and financial resources available to Employment Services Russia. In summary, the general absence of program evaluations in Russia makes it difficult to assess the performance of ALMPs. Their composition indicates a much greater focus on training and job creation/preservation than on programs that have been determined through international experience to be more cost effective, such as job counseling or job information services. Profiling may help identify at risk groups in Russia, but the cost effectiveness of this mechanism is not well known, and its introduction should be monitored with care. If profiling will reduce the effectiveness of job counseling and information services, its introduction should be reconsidered. The political pressure applied to employment services to reduce open unemployment (maintain/create jobs), rather than helping the unemployed find jobs, represents a serious area of concern from the standpoint of the functioning of an efficient labor market. Social support for enterprise restructuring. The Russian economy still requires significant restructuring, which may involve mass layoffs in particular indus tries. This section discusses social protection strategies for laid off workers in restruc SAFETY NETS FOR WORKERS 123 Box IV.3. Profiling to Reduce Long Term Unemployment Profiling of the unemployed is used in many OECD countries to identify those vulnerable to long term unemployment and to match them with cost effective ALMPs that would improve their chances of employment. Profiling methods differ across OECD countries. United States. Profiling is based on early identification of those likely to become long term unemployed, particularly repetitive job losers. A model has been developed to predict which unemployed persons will exhaust their benefits. Those identified are required to take part in cost effective job search assistance. Australia. Profiling based on a statistical model is used to identify those running a risk of long term unemployment on the basis of age, education, and other variables. In 1995, such methods identified 5 percent of the unemployed; another 10 percent were identified based on judgmental factors. Belgium. Since 1993, the employment service has targeted the unemployed under age 46 without complete secondary education and unemployment of 10 months or more. These per sons are required to participate in an action plan consisting of steps such as job search coun seling, training, or subsidized employment. Great Britain. During the initial interview, the job seeker and counselor agree on a re employment plan as a condition of receiving benefits. Job seekers are provided job search or job review seminars. After unemployment for more than 24 months, the unemployed must attend a New Start course. More intensive assistance is given to those under age 25 in order to prevent long term unemployment among youths. Source: Prokopov and Maleva (1999). turing industries, one company towns, and remote areas in Russia. Findings from international experience complement this evidence. Restructuring the coal industry. Workers in particular industries, such as coal, are likely to face greater reductions in employment than workers in the economy overall. This is certainly true of transition economies and in developed market economies as well. For example, a shrinking demand for coal, and consequently coal miners, has been experienced in Great Britain, France, Germany, and the United States, as other forms of energy have become more economical than coal. This situation has affected Russia rather later than the OECD countries. Strong unions and high wages and benefits have typified the coal industry in many countries. Unionization started in the mines because of the terrible working condi tions in the 19th century, and continued strong as hazardous circumstances prevailed in even the best run mining operations. Furthermore, long term health risks have long been recognized, in particular black lung disease, which has led to earlier retire ment ages and other special programs to assist miners. Russia has been no exception to these trends. In addition, because of high wages and family traditions, in many countries there is a pattern of sons following fathers into the mines. In other words, there is not a long history of alternative employment among mineworkers. Consequently, restructuring in the coal industry is a particularly delicate and difficult issue for any government to address. Two surveys of the social impact of coal restructuring, one each in 1996 and 2000, which cover several coal mining regions in the Russian Federation, have been com 124 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY missioned by the World Bank152 Findings indicate that the self reported status of the living conditions of laid off miners improved between 1996 and 2000, with the pro portion of miners saying that they "had trouble making ends meet" declining from 45 percent to 36 percent. The demographic pattern of lay offs was more heavily con centrated among women than the overall pattern of employment in the industry. In 1996, about one quarter of lay offs were women, while in 2000 that proportion had risen to 40 percent. Overall, women make up a large share of mineworkers. The over representation of women may reflect the elimination of increasing proportions of "surface" jobs later in the mine closure process. Over time, workers also obtained better information about their rights and enti tlements, reflecting in part the effectiveness of Employment Services. Workers in the 2000 survey appeared to be much better informed about their rights and entitlements than those in 1996. Benefits include two types of severance pay: (a) standard sever ance equal to three months' average salary, and (b) 15 percent of salary per year employed for workers at retirement age. In both survey years, about three quarters of the respondents reported receiving the required two months' advance notice of lay offs. In 2000, a considerably higher percentage of workers reported receiving infor mation about their rights compared with 1996. Those receiving an old age pension tend to report a relatively better financial situation than other laid off workers do, while those receiving a disability pension do not. Laid off miners are also eligible for wage arrears. In 1999, more than 95 percent of laid off miners knew about their entitlement, and a similar percent said they were eligible to receive back wages. Nonetheless, as in all programs in Russia, there were regional differ ences among those surveyed. In 2000, the incidence of wage arrears was higher among those workers laid off at closing mines compared with those laid off at continuing oper ations. This finding suggests that the continuing operations in 2000 have managed to develop a business strategy that will ensure that they continue in operation, hopefully, on a profitable basis. Miners laid off at continuing mines also appear to be in a better financial situation than others are, but this may result from regional and demographic differences, such as additional earners, higher education, and re employment in mining, as well. Part of the social impact survey focused on the economic situation of miners laid off because of mine closings.153 The proportion of unemployed in 2000, at 38 percent, was substantially smaller than in 1996, at 66 percent, possibility indicating a more robust and flexible labor market. Nonetheless, the 2000 rate is more than three times higher than the 10 percent unemployment rate reported for the country as a whole. One reason may be that they have significantly higher wages, and thus unemployment benefits are much higher for them. Three quar ters of the unemployed had registered at their local employment offices, a pro portion many, many times greater than that of the population overall.154 152 This section of the report is based entirely on a study (forthcoming) on the social impact of the coal sector restructuring (ECSIE) The regions are Kemerovo oblast, Rostov oblast, Tula oblast, and in 2000, the Komi republic. 153 The 2000 survey also looked at lay offs from downsizing. 154 These high proportions of registered unemployed are not reflected in overall oblast fig ures, however. SAFETY NETS FOR WORKERS 125 While we have no research on the reason for this finding, it may be related to the fact that severance pay and back wages are enforced among coal miners and to the active role that Employment Services played in this process. Similarly, in 2000 more than 70 percent of the unemployed were receiving employment benefits on time and in full. The recipiency rate of unemployment benefit is also in considerable contrast to the rest of the population. Laid off miners in 2000 frequently found other jobs within the coal sector. This may have led to the decline in the numbers of unemployed who indicated that they would like retraining, which dropped to 41 percent in 2000 from just over half in 1996. Furthermore, some 54 percent of respondents believed that retraining would not be effective. This is an interesting finding, as training programs have not been found to be effective in many cases in both OECD and transition economies. The sur vey also found that younger people were more likely to want to start a business. This is also a finding that needs to be considered carefully, particularly since results of other studies indicate that conditions for business development need to be improved in Russia. In summary, the financial options offered miners have probably assisted them in their period of unemployment. However, once again no quantitative assessment has been conducted, and the impacts of these programs (other than allaying the political costs of transition) are not well known. Furthermore, retirement options have removed miners from the work force, but at a cost to the pension fund. Nonetheless, the higher unemployment rate indicates that the measures used cannot have reduced social stress among miners completely. This should not be a surprise, however, because this is an issue with which all developed market economies with mining operations contend. The availability of severance pay, back wages, and unemploy ment benefits for mine workers, however, is far higher than that received by other unemployed persons in the population. In other words, the role of ALMPs and sever ance packages, including back pay, retirement benefits, and unemployment benefits, has been important in downsizing strategic industries, as it often is worldwide Out migration from one company towns. Restructuring the North requires a further reduction in the population, although significant population shifts already have taken place naturally. The high cost of moving, particularly given the rigidity of the Russian housing market, has made out migration for some groups quite difficult. As noted in the first chapter, research and surveys have shown that the greatest obstacle to migration is the need for sufficient cash to purchase housing in a new region, as mortgage markets are not yet well developed. Other considerations, such as availability of employment and education, are secondary to migrants. If opportunities to move to more hospitable areas of Russia are facilitated for the unemployed, retired, and others, subsidies and costs to the federal and local governments could be reduced. These costs include the mainte nance of an artificially large population in the North. Population redistribution could also improve the financial footing of fledging private companies in the North by lower ing taxes and enhancing the welfare of the remaining population. The objective of the recently initiated Northern Restructuring program of the Government addresses this problem by supporting the restructuring of three munic ipalities, including supporting the out migration of up to 25,000 people. The partici 126 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY pating municipalities represent the diversity of economic and social circumstances of the Russian North. One is the center of a mono industrial coal producing area in which the industry has been severely downsized since the government significantly reduced subsidies for coal producers, liberalized prices, and initiated coal sector pri vatization. In this case, the closing of a small, nonviable settlement will be tested. Another municipality is a mono industrial city of about 250,000. The major enterprise has been privatized and is restructuring. In this case, a large proportion of nonwork ing people (pensioners, disabled, or unemployed) will be assisted. The third area is a gold producing area where work is seasonal. There, enterprises are starting to hire a seasonal workforce, and settlements are being consolidated at the district center. The government will provide eligible groups with migration assistance allowances in the form of housing certificates to those who voluntarily participate in the program. Thus, allowances are specifically tied to the purchase of housing. In addition the migrants will be provided transportation to the new location for their families and household belongings. Who should be eligible for assistance? The eligibility criteria for the Migration Assistance Program were developed in cooperation with the local authorities and stakeholders in each municipality and are tailored to reflect the local priorities. The selection criteria were intended to reduce costs and improve productivity, and vary considerably across regions. The results of this program should help inform similar initiatives to restructure enterprises or facilitate migration in other remote one com pany towns, both in the North and elsewhere in Russia. International experience. Social support for laid off workers has been critical for facilitating restructuring around the world.155 Some social support programs for labor retrenchment are directly linked to the privatization of state owned enterprises (for example,, in transition economies in Eastern Europe and Central Asia), while oth ers (for example, in North America and Western Europe) are part of an ongoing process of economic change and renewal. However, the design and use of social sup port programs vary considerably across countries and are greatly influenced by the economic environment, including the level of unemployment, and type of general social support programs already in place where economic restructuring is occurring. These programs can take various forms: they can be voluntary or involuntary; com pensation packages can be standard or tailor made, and they may or may not include ALMPs. A brief survey of a sampling of such programs follows. Western Europe. In Germany, a new institution, the Truhandanstalt (THA) was established to deal with rapid privatization of some 8,000 state owned companies, with a workforce of 4.1 million. The privatization program had a rapid, severe impact on employment. Labor reductions were achieved by early retirement; job placement in new private firms; employment creation schemes, including wage subsidies, pub lic works, and retraining; plus unemployment benefits. Special employment compa nies and counseling services were also established to employ and retrain workers. In the United Kingdom, British Coal divested a total of 204,000 workers, mostly over the 155 This section draws from Fretwell, David, T. Beck, and E. Johannsson. (1995); Hoeven, Robert and Gyorgy Sziraczki (1997); Kikeri, Sunita.(1998). SAFETY NETS FOR WORKERS 127 age of 50, who accepted lump sum redundancy payments. In addition, British Coal Enterprise Ltd. was established to assist employees in the sector, and their families, in developing skills and securing new employment with a special emphasis on helping displaced workers start small businesses. In Sweden, the Uddevalla shipyard was downsized via normal turnover, early retirement, a freeze on recruitment, and assis tance with job search and retraining. The KLAB mine was downsized by normal and early retirement, severance, and retraining. Eastern Europe. Privatization in transition economies often has taken place in a difficult environment with the economy contracting and unemployment increasing. In Poland, restructuring in the coal sector has been supported by a Miners Social Package, which includes lump sum payments and early retirement assistance, com bined with active labor programs (for example, small business assistance, counseling, retraining) and local economic development assistance to affected communities. Early retirement pensions were also used in many other CEE countries to facilitate layoffs early in the transition, but were quickly abandoned as their long term fiscal costs became apparent. In Macedonia, severance payments were combined with active labor programs to assist workers affected by restructuring of 25 large loss mak ers. Approximately one third of affected workers participated in the latter services. An added complication, particularly in the CIS transition economies such as Russia, is the connection of a large number of community services (for example, schools, hos pitals, heating, and housing) with state enterprises. When these enterprises are liqui dated or downsized, the social assets must be disbursed in a manner to ensure that essential community social services are maintained. Latin America. In Brazil, a varied set of income support and other support pack ages were used to retrench workers in labor in six state owned enterprises, including three banks and thee utilities, between 1995 and 1997. A parallel program was carried out by several states. Severance payments were a core element of the program, plus extended medical benefits, retraining, help for business start ups, and job search assistance to affected workers. Economic restructuring in Latin America has some times been carried out in a manner and in an environment that has increased employment, thus making divestiture more palatable to labor. For example, in Chile, employment in 10 state owned companies privatized between 1985 and 1990 increased 10 percent because of overall headway achieved in economic growth and investment by the firms involved.156 In Argentina, starting from a base of 222,000 employees in 13 major public enterprises in 1990, employment was reduced to about 42,000 by 1993 by transferring 66,000 workers to private firms, retiring 19,000, and providing 95,000 with severance payments. Retirement, generous severance, multijob holding phenomena, and reactivation of the economy and expanding labor market, are credited with deterring labor opposition.157, 158 The social support programs used internationally combine both passive and active labor market elements. Income support programs include unemployment 156 Larroulet (1992). 157 World Bank (1993). 158 Guasch, Luis (1996). 128 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY benefits, social assistance, pensions, and regular mandated severance. They some times include special job loss compensation (lump sum) in addition to regular sever ance, early retirement schemes either through the regular pension system or through additional social assistance payments to workers near voluntary retirement age. For older workers, programs also may include temporary community employment until retirement age or administrative leave at some percentage, 75 80 percent, of salary until retirement. ALMPs provided to workers displaced because of restructuring gen erally include intensive job placement assistance such as remote job search. As noted earlier, these are the most cost effective programs of all ALMPs. However, different types of on the job training or institutional training, small business assistance, and public works are also provided in the case of structural unemployment. As noted above, while these programs may mitigate the social and political cost of layoffs, they tend to be very costly with very moderate wage or employment impact.159 Empirical work on enterprise restructuring conducted by Haltiwanger and Singh (1999) finds that the generosity and extent of programs reflects the underlying causes of restructuring. When retrenchment was perceived as a one time event to address low worker productivity, compensation typically consisted of severance and enhanced pensions, and retrenchment was voluntary. On the other hand, when retrenchment was perceived as part of a fundamental, radical transformation of the public sector, including a restructuring of the labor market, such as that of transition countries, programs were richer. Provisions for severance and enhanced pensions were accompanied by worker safety net measures such as unemployment benefits, job placement services, and worker retraining. In addition, these programs more often included a mandatory component. Severance pay was the most common instrument (used in 68 percent of projects), followed, in turn, by enhanced safety nets (63 percent) and enhanced pensions (29 percent). The authors also find that for every dollar spent on severance pay, an additional 1.2 dollars was spent on enhancing safety nets and 2.2 dollars on enhancing pensions.160 159 Fretwell and Wilson (1999). 160 For political reasons, voluntary retrenchment programs have become increasingly popular (Rama 1999). However, standard voluntary retrenchment programs, offering benefits primarily based on years of experience, may lead to severe adverse selection problems, because the most pro ductive workers often have superior labor market opportunities outside the public sector. Special tailor made programs could be designed to increase the efficiency of downsizing by disclosing worker characteristics. For example, the use of confidential individual bids for exit compensation, with safeguards to prevent collusion, has been proposed to lead to such disclosure (Jeon and Laf font 1999). Unproductive workers would propose the highest bids, all else being equal, as they would stand to lose the most from separation. Because determining the right menu may be difficult in practice, Rama (1999) recommends the use of other, simpler procedures as well, however. Another possible cost effective procedure would determine severance pay on the basis of welfare losses arising from the worker's separation ("indexing"). Severance pay can be indexed to a wide selection of observable worker attributes, including present wages, job security, gender, years of past service, expected duration of the unemployment spell, and prevailing wages that the separating worker can expect to earn in the private sector. For state owned enterprises in Egypt, Asaad (1999) finds that a tailor made program could reduce total compensation by 31 percent in comparison with a standard program, and that severance pay providing higher payments to long tenured work ers is likely to overpay them (from Vodopivec and Raju 2001). SAFETY NETS FOR WORKERS 129 In summary, the review presented in the preceding paragraphs demonstrates that a broad range of social support programs often accompany enterprise restructuring. Most successful programs include direct dialogue between stakeholders (for exam ple, the Government, enterprise management, workers, and community leaders) both before and during the restructuring program. The stress on income support (for example, severance) and other forms of support (for example, labor services) and their generosity varies depending on economic and employment conditions, and the rationale for restructuring, but most programs include both elements to varying degrees.161 While these programs may be very similar to those used to combat unem ployment that is cyclical or frictional economy wide, they are frequently provided with greater generosity than is financially feasible for the economy as a whole. Fur thermore, benefits tend to be more generous despite disemployment effects as they serve communities in crisis in which a lack of focused assistance could lead to greater social unrest. Administration of unemployment programs. The success of unemploy ment protection programs and special restructuring also depends on proper pro gram administration. The Russian Government has been active in improving its employment program operations. The organization of employment offices was upgraded by the creation of 39 model offices intended to introduce best practices of European labor services. Career counseling centers were established in 20 regions, and social partnerships to encourage job creation were developed in three pilot regions. Staff was trained, and many rayon offices across 77 regions of the Russian Federation were computerized. These new methods -- both ideas and technology -- were disseminated to other regions through training seminars and exchange of regional experiences (Tchetvernina 2000). Staffing and training. The staffing ratios and administrative costs of regions vary considerably. While average expenditures on administration and information tech nology are 18 percent across Russia, 13 regions allocate 25 percent or more of their employment program budgets to administration and information technology (IT). These include regions such as Moscow and Smolensk, which are relatively prosper ous, but also Eastern Siberia and the Far East, in which workers receive pay coeffi cients to make up for hardship conditions (table IV.3). Administrative expenses in high cost regions are not related to staff size. Total regional administrative costs per staff member (administration and IT) are higher rel ative to regional average wage rates, when such expenditures take up a greater share of employment fund resources. The basis for these high cost offices needs to be eval uated in order to improve the efficiency of resource allocation in these offices and nationwide. It may well be that administrative expenses capture provision of ALMPs as well as staff and service costs, and are not strictly related to the latter. This assessment is particularly important because staffing ratios help explain part of the regional variation in coverage of employment programs. Regions with low staffing ratios handle fewer clients, limiting the coverage of the program. The quality of staffing (inadequate training, lack of information) -- perhaps reflecting differen 161This discussion is based on Fretwell (2000) and Vodopivec and Raju (2001). 130 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY tial access to training -- may also be important in explaining regional variations in coverage. Recent studies, including a beneficiary assessment of employment offices conducted as part of the Bank financed Employment Service and Social Protection (ESSP) project, finds that Employment Office workers find that their effectiveness could be improved through a greater sharing of ideas and information across regions (Tchetvernina 2000). Monitoring eligibility and work incentives. A rigorous evaluation of the administra tive efficiency of the Russian Employment Service offices in monitoring and verifying claims and enforcing work incentives has not been conducted. However, Russian Employment Office staff are likely to face similar problems as administrative workers in CEE transition countries. A recent review of the administrative capacity in transi tion countries to monitor unemployment program eligibility found that Employ ment Service staff face considerable difficulty in undertaking this task. First, the large informal economy prevailing in most transition countries makes the task of moni toring very difficult. This is particularly the case in transition countries with consider able corruption and lack of law and order. Second, the culture of entitlement remains ingrained in both beneficiaries and employment office workers, so that eligibility is often not verified. Weak monitoring and enforcement capacity, such as the lack of technology, infor mation systems, resources, and often also the political will to monitor and enforce existing laws, makes this task even more difficult. For example, although informal employment is prevalent in many transition countries, labor inspectors catch few vio lators. In Hungary and Slovenia, benefit disqualifications are very rare. Similarly, veri fying the accuracy of self reported earnings of the benefit recipients (in countries that require such reports) is often not done -- in part because there are no mechanisms available to counselors to do so. Therefore, the adoption of unemployment insurance programs more suited to the administrative capacity, information availability, and formalized markets of OECD countries may have been slightly premature in Russia as in other CEE countries. Program dependence. International evidence suggests that the administrative capacity to implement unemployment protection programs varies by program. Flat unemployment benefits are the least difficult to administer. They do not require past wage and employment history or collection of contributions, or changes in benefit over time. Public works are also not very difficult to administer. However, means tested unemployment benefits are at least as difficult, if not more difficult, to admin ister as unemployment insurance programs. In countries where the informal economies are large, and cross checking systems don't exist, means tests are likely to be ignored, used on an ad hoc basis, and may be unreliable (World Bank 2001b). The self policing nature of the ISA system could help reduce some of the work incentive effects imposed by unemployment insurance programs. Under the tradi tional unemployment insurance system, employers in developing countries some times fail to pay program contributions. By introducing personal accounts, workers themselves monitor such payments. Of course, if workers anticipate frequent spells of unemployment, and if benefits are paid to those who exhaust their accounts, com pliance may still be a problem. Furthermore, the administrative complexities from SAFETY NETS FOR WORKERS 131 introducing individual account based ISAs are similar to those required by Notional Defined Contribution (NDC) or second pillar pension systems.162 ISA programs require the essential preconditions of funded pension systems: a functioning financial market, including adequate supervisory and regulatory capacity, so that money accu mulating in individual accounts can be managed appropriately. These preconditions are important before ISAs can be introduced in Russia. Political pressures. Another major concern about the administration of unem ployment protection programs in Russia is in the effectiveness of Employment Ser vice offices, which may be compromised by interactions between municipal govern ments and loss making enterprises (Tchetvernina, 2000). These interactions may result in the inappropriate use of ALMP resources. Enterprises with accumulated arrears to the Employment Fund are reported to have been "encouraged" (forced) to participate in "job preservation" or "youth training" programs as a way to restructure their arrears. More generally, enterprises admitted being under pressure by local authorities to maintain employment. While the decline in expenditures on job cre ation/preservation programs may have lessened this pressure, it is important that Employment Service offices focus on assisting the unemployed to find jobs, rather than containing the growth of unemployment itself. C. Policy Options In summary, Russia's safety net for workers has not been very effective in allaying the social costs of layoffs. But, an effective unemployment benefit system is important for helping workers cope with income and skill loss as a result of unemployment. It is also important if Russia is to move protection outside of firms to the public domain and to facilitate restructuring of large industry, the budget sector, and one company towns or facilitate depopulation of the North. The recent government program has made important steps in this regard. The following policy options provide some choices to the government as it moves forward to create a financially viable and effec tive unemployment protection program. Unemployment Benefit Programs Benefit design options. The three options for the unemployment benefit pro grams are: (a) benefits based on past wages and work history (currently the case), (b) 162 Vodopivec and Raju (2001) note that the risk of having high administrative costs of pri vate pension accounts in the United States as low to medium, and a similar assessment is valid also for ISA accounts, and for other countries as well. To keep the costs of private accounts low, some experts proposes that investment funds are approved and regulated by the government and subject to standard auditing controls to reduce fraud. He also proposes limits on investment charges as well as on free movements of money between funds. In such a case, most of the administrative costs would come from collecting contributions from individual workers, that is, at few extra costs in comparison to the public system. 132 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY unemployment assistance (means tested or flat), and (c) Individual Saving Accounts (ISAs). A flat benefit is best suited to Russia because it minimizes administrative requirements and is consistent with the shift toward general revenue financing. While, in theory, workers can provide documentation on their work at a firm without a contribution history, in practice, this may be subject to considerable fraud and abuse. Furthermore, workers in the informal sector who lose their jobs are unlikely to be able to obtain accurate documentation from their employers. The 1999 reforms have already moved the unemployment benefit part way to a flat system, at least for all but voluntary and redundant workers. The introduction of a flat benefit may not be a politically viable option. There is considerable preference in Russia for a benefit that reflects past wages. In this case, a simplified version of a wage based benefit might be introduced, with benefit as a fixed proportion of an individual's wages, subject to appropriate maximum and min imum caveats. The benefit could still be financed from general revenues. However, the period of wage assessment for benefit should be increased considerably from the current three months to reduce incentives for collusion between employer and employees. That is, wages paid for the final three month period could be inflated considerably above those received earlier. The success of a benefit system that depends on past wages will require improvements in collection of accurate informa tion on wages (or greater formalization of the economy). Keeping benefit levels dependent on past wages may have the additional benefit of creating incentives for workers to accurately declare past wages once the benefit level improves. Unemployment insurance (based on the collection of individual contributions) might be re introduced once the economy has formalized and administrative capac ity to monitor individual records has improved. ISAs might also be considered, per haps in conjunction with unemployment insurance. However, their introduction will require better financial markets and regulatory capacity than are currently in place. The introduction of these accounts is also risky at present, as there is limited experi ence with this system worldwide. Incentives. In the future, the government would have to ensure that "unem ployment traps" arising from a more generous level and duration of benefit are avoided. Research has shown that benefit generosity can help smooth consumption but also reduces incentives for individuals to re enter the labor force. ·Whether benefits are set as a share of individual or national wage, or are flat, the proportion of benefit should be set to ensure that it is not too high as a share of average national or regional wages. Therefore, benefit levels might be set as a low share of average wage, for example, equal to about 30 percent of the average wage and within the financing envelope (see section on financing and administration below) to avoid adverse incentive effects. ·The benefit level should be coordinated with the minimum wage. In the short run, a flat benefit equal to about 30 percent of average wage would be much higher than the minimum wage. However, over time, the level of unemploy ment benefit should be roughly equal to or somewhat lower than the mini mum wage (assuming it is roughly 30 to 35 percent of average wage in the SAFETY NETS FOR WORKERS 133 long run). If wages are a proportion of average wage, the minimum benefit could keep pace with the regional minimum wage, while maximum benefit could be some proportion of average wage (for example, 50 percent). Regional differentiation in benefit would also be important given differences in income across the regions. ·Benefits should be assessed with reference to the average wage. This is because the opportunity cost of beneficiaries is the market average wage, not the sub sistence level. Over the medium term, the average wage, rather than the sub sistence minimum, also provides better information about the fiscal resources available to the government to finance the unemployment benefit program. Evaluating benefits with respect to the subsistence minimum can be very costly if wages are lower than the subsistence level. If current average benefit were equal to the subsistence minimum, then benefits would be 60 percent of average wages. It would then be very expensive for the government to finance unemployment benefits and would also cause work disincentives for low wage workers. ·The duration of benefit might be reconsidered or reduced for voluntary quits from a year to about six months. If the reformed program is fully funded, the current duration of benefit would induce work disincentives as found in many CEE countries. Benefits/privileges for some. The current benefit program also provides sev eral guarantees and benefits on a preferential basis that might be reconsidered to make the program more targeted and less costly in the medium term. Phasing out these priv ileges is consistent with the government's policy to reduce privileges and better target benefit. The most important privileges/benefits for some workers include: ·Guarantees for dismissed workers. Guarantees to dismissed workers for housing, medical, and preschool services at their former places of work should be removed. These encourage employers to press workers to quit rather than be laid off and provide disincentives for former workers to relocate where better jobs can be found. ·Northern and Chernobyl benefits. Additional benefits for persons located in the North who had received higher compensation. Such benefits are inappropriate in a market oriented employment program because they create inequities across unemployed workers, provide work disincentives, and discourage mobility. ·Early retirement benefits spanning unemployment to pension. These schemes tend to be extremely costly, often with costs spiraling out of control. Consequently, it is better not to have early retirement provisions in the Employment Law. However, if it is desired, the scheme should be designed very carefully, with tight eligibility conditions, taking into consideration future expenditures. Eligibility options. The unemployment scheme in the Russian Federation makes many workers eligible, which turns it into a hybrid program -- somewhere in between an unemployment benefit and a social assistance system. 134 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ·Large number of benefit categories. The program in Russia covers individuals who do not qualify for benefits in OECD countries for moral hazard reasons, for example, workers who have left for disciplinary reasons, job leavers, re entrants, and training dropouts. Therefore, consideration should be given to reducing eligibility from the wide array of unemployed currently eligible under the system, both to target the system and to reduce moral hazard problems in the future. ·Voluntary quits vs. layoffs. The current practice of providing unemployment ben efits to individuals with recent work histories, such as job losers and job leavers, might be continued since there is little difference between the two groups. In the longer run, benefits to job leavers should be restricted. But such a reform must be delayed in Russia until the labor market becomes more trans parent. Program monitoring. The Household Budget Survey, if nationally representa tive and redesigned to include questions on the level of benefit and individuals' char acteristics, could be used to approximate the incidence, adequacy, and coverage of the unemployment benefit program on a continual basis. Continual monitoring of these indicators could help policymakers assess the impact of the program and make changes to improve its effectiveness, as needed. Active Labor Market Programs Contrary to popular belief, ALMPs have only moderate impact on alleviating long term unemployment. Their ability to assist high risk unemployment groups might be improved in the following areas: Program selection/evaluation. Despite their intuitive appeal to politicians and the public, ALMPs provided should be cost effective and likely to succeed. This is partic ularly the case given the weak results found across OECD and transition economies. The international evidence suggests that if resources and administrative capacity is limited, job counseling and information are the most effective ALMPs. International evidence also suggests that evaluation of programs is also important to assess their performance, prior to any replication of current efforts. For this, well designed pro gram evaluations are required, including of unemployed participants and nonpartic ipants. This work is being initiated in Russia and should be expedited. Profiling to improve matching. Profiling could also help match particular groups to programs that work best for them. However, since the cost effectiveness or impact of profiling is not well known, an experiment in Russia on profiling should be under taken with caution. A pilot approach is warranted that would include built in evalua tion so that the cost effectiveness of the program and its impact on reducing long term unemployment can be assessed. Employment quotas. Employment quotas should not be part of the ALMP pro gram. Employment requirements for training program participants should also be avoided. Evidence suggests that employment quotas are not an effective means of SAFETY NETS FOR WORKERS 135 providing jobs for persons with disabilities. In fact, employers tend to prefer to pay fines rather than having their employment policies weighted down with special rules and regulations that may make hiring persons with disabilities costly. Introducing private provision. A redesign of ALMPs toward hard to place employed would also be appropriate since a private sector employment service industry has started to develop. These services ought to be encouraged but also should be well regulated. Such private businesses are more likely to provide services for more highly qualified workers. Individual choices of program and private provi sion of training would allow a better match of individuals with program and jobs, because individuals and the market can be assumed to have better information about training needs than the Government. However, appropriate regulation of private providers is also required in order to reduce any potential abuse. Social Support Packages for Restructuring In addressing the remaining restructuring agenda, the government might consider developing a strategy prioritizing key industries and areas (North, or one company towns), while phasing in the social support required. This strategy could be informed by Russia's own experience in restructuring and on international practice. The main ele ments of this strategy might include the following: (a) identification of the enterprises to be restructured, and the demographic and work skills of their personnel; (b) agreements on parameters of a social support package (determine its scope, costs, source of financ ing, and administration drawing on existing mechanisms where possible); (c) stake holder involvement; (d) a public information campaign; and (e) monitoring and evalu ation mechanisms. The latter could help to ensure that workers are not rehired via a "back door" and that workers who have difficulty in re entering the labor market are identified early and given targeted assistance to ensure they do not slip into poverty. If enterprises have social infrastructure (schools, clinics), divestiture of these assets might also be monitored to ensure that it has been successfully transferred to municipalities. Cross Cutting Issues: Administration, Financing, and Evaluation Changing requirements for employers. The demands placed on employers to report job vacancies or job placements to employment offices should be reduced and reporting should be voluntary. Employers also should not be required to accept appli cants from employment offices or indicate why applicants have not been accepted. Furthermore, municipalities should not be able to forgive tax or contribution arrears in exchange for the establishment of training programs, hiring of unemployed work ers, or the like. These are all outside the practices of a competitive economy because they raise the cost of business for employers. Evaluate and regulate administrative expenditures. In line with policy reforms, the Government could profitably undertake a serious evaluation of staffing needs and organization. Organizational changes should ensure that staff is well trained and flex 136 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ible and that ALMPs offered are related to local employment opportunities. Informa tion exchange could be ensured through direct contacts between regional and rayon employment service specialists. Financing. Resources budgeted for unemployment benefits and ALMPS should reflect expected regional needs to avoid arrears. In addition, distribution of benefits to regions should be done in a fair and transparent fashion. A simple way to do this is to distribute finances (within the overall budget envelope) using information on regional and or survey based unemployment rates (RLFS). A simple simulation analysis varying the basic parameters of employment pro gram expenditures across regions was conducted to inform total financing decisions. The simulation is as follows: Consider several flat benefit alternatives that could have been substituted for the underfunded and overly complex system in place in 1999. For example, what if an adequately financed and redesigned program were based on one of the following scenarios: (1) Low Case Scenario: ·Registration of at least 30 percent of the unemployed; ·Payments averaging 30 percent of the average regional wage; ·Reallocation of ALMPs toward effective treatments; and ·Reduction of administrative expenses in high outlay regions. (2) Medium Case Scenario: ·Registration of at least 35 percent of the unemployed; ·Payments averaging 40 percent of the average regional wage; ·Reallocation of ALMPs toward effective treatments; and ·Reduction of administrative expenses in high outlay regions. (3) High Case Scenario: ·Registration of at least 40 percent of the unemployed; ·Payments averaging 50 percent of the average regional wage; ·Reallocation of ALMPs toward effective treatments; and ·Reduction of administrative expenses in high outlay regions. Furthermore, in each of the three cases, "high outlay regions" in terms of expen ditures for ALMPs were constrained to reduce expenses. Eight regions had expendi tures on ALMPs that were considerably higher than the average of 17.5 percent of expenditures. In fact, they averaged 38.2 percent of expenditures compared with an average of 12.8 percent for the others. The regions are Moscow, Orel, Smolinsk, Bel gorod, Ingushetia, Orenberg, Amur, Sakah, and Yakutia. The expenditures of these eight regions alone accounted for 40 percent of spending on ALMPs in 1999. The high cost region expenditures were particularly directed toward the "preservation of job places" and the "creation of additional jobs." The first provides funding for job maintenance and the second directly provides 12 months' average salaries for work ers. These are essentially job subsidy programs of one sort or another that research has shown to be inefficient. SAFETY NETS FOR WORKERS 137 Figure IV.1. Simulation of Unemployment Benefit Expenditures 0.90 0.84 0.74 0.80 0.70 0.62 0.60 0.53 GDP 0.44 of 0.50 0.35 0.40 Percent 0.26 0.30 0.16 0.20 0.10 0.00 ALMP outlays Administrative UI benefits Total expenditures expenses High case scenario Medium case simulation Low case simulation Actual expenditures In addition, regional expenditures were increased to equal actual expenditures reported for Sakha (Yakutia), measured as a percentage of simulated passive labor market program expenses. Only Sakah spent as much as 5.5 percent of outlays on programs facilitating job search for the long term unemployed, that is, persons out of work for more than six to eight months. As noted above, these programs are often the most cost effective. Lastly, administrative expenditures were constrained at a maximum of 25 percent of passive program outlays, with the exception of Moscow.163 Figure IV.1 compares actual with simulated expenditures for the three scenarios presented above compared with actual expenditures as a percentage of 1999 GDP. Under the simulations, expenditures on unemployment benefits in 1999 would have ranged between 0.35 and 0.74 percent of GDP compared with actual expenditures of only 0.16 percent. Each of the new programs would have increased expenditures for each type of activity -- ALMPs, administration, and passive programs -- but as a per centage of GDP expenditures ALMPs and administrative expenses would not be sig nificantly higher. The gains from cost reductions in high cost regions are essentially balanced by increased expenditures on more effective ALMP programs across all regions. Total expenditures in the high cost regions would have been reduced by 163A number of regions that had unusual expenditures on administration or ALMPs, based on actual outlays on unemployment benefits, did not need to cut their expenditures once their passive labor market programs were improved. 138 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY nearly 60 percent as a result of cutting inefficient programs, and their share of ALMP expenditures would have decreased from 41 percent to 15 percent of the total expenditures. In the Low Case Scenario, increased total expenditures could have been easily funded through reductions in allocations for other social programs (discussed above). In the other cases, more difficult choices would have to be made. The simula tions also presume that higher benefits lead to higher take up rates. For example, if benefits were set at 50 percent of the regional average wage (the High Case Scenario), the simulations assume a take up rate of 40 percent of the unemployed. With higher benefits, the likelihood of work disincentives increases, particularly for lower wage workers. These increases could be reflected in higherinitial take up rates or in a higher proportion of the unemployed staying on the rolls for the full period of bene fit entitlement. For that reason, Russia should consider pursuing a less generous ben efit policy until sufficient data are available to determine the behavioral responses of program participants to more modest unemployment benefits, but ones that encour age take up among unemployed and unpaid workers. Summary and Conclusions Unemployment benefit programs and ALMPs have had limited coverage and uncertain financing over the past decade, reducing their potential to protect workers against income or skill loss as a result of unemployment. A very low and unpredictable level of benefits reduces the usefulness of the system, thus putting large numbers of work ers to hardship which could be avoided. The Government has introduced general rev enue financing of unemployment benefit programs and ALMPs. However, the benefit design and ALMP strategy has not been fully defined. The chapter suggests that the following elements might be considered for the design of the safety net for workers in Russia. An effective safety net would protect workers against poverty, help facilitate restruc turing, and assist in moving protection out of firms and into the public domain. The report suggests considering the following elements in the design of the new program. Unemployment benefit design should be simple to administer, with incentives, and adequately financed. The report provides several policy options for unemployment benefit design. ·The report provides three key benefit options: (i) a flat benefit, fixed in nominal terms as some percent of average wage, and indexed to prices is one option for policy makers to consider. A flat benefit minimizes administrative require ments, is progressively distributed, and is consistent with general revenue financing. (ii) The Government could also consider simplifying the benefit for mula to one that is some fixed percent of average wage over the entire dura tion of the benefit. (iii) If these options are not politically feasible, and the Government decides to retain the current formula, the report recommends the following changes in the eligibility and duration conditions of benefit. SAFETY NETS FOR WORKERS 139 These changes should be considered whatever benefit formula option is cho sen by the Government: ·Over the medium term, the level of benefit should be set so as to minimize work dis incentives. The benefit level would remain a low share of average wage (e.g. 30 percent) to ensure work incentives. The minimum and maximum benefit lev els should be delinked from minimum subsistence and established relative to the average wage. Over the medium term, the average wage will give more reli able information on the availability of fiscal resources and work disincentives for beneficiaries than the subsistence minimum. Given large regional differen tiation in wages, differentiation of regional benefit levels will be important. ·The assessment period for benefits should be increased, and benefits established at a fixed proportion of an individual's wages (for example, 30 percent of wages) in order to ease administrative requirements for processing benefit claims. ·The duration of benefits could also be reduced to a maximum of six/nine months as in other CEE countries. A long duration of benefits, coupled with more gen erous level of unemployment benefits in the medium term, might induce longer unemployment spells. ·Benefits could be provided to fewer categories of workers, such as laid off workers and voluntary quits. Over time, as the distinction between voluntary quits and laid off workers is reduced, benefits for voluntary quits should be phased out or the eligibility of voluntary quits should considerably tightened in line with international practice. Special benefits to e.g. northern workers should be phased out as well. Targeting benefits would help save program expenditures, help the truly deserving, and reduce administration costs. ALMP strategy. The future thrust of ALMPs in Russia is difficult to determine since programs have not yet been empirically evaluated using best practice evaluation methods. Implementing such program evaluations should be expedited by the policy makers. On the basis of administrative data and international experience, however, the report indicates the following direction for ALMPs: ·ALMPs are an important complement to passive programs, such as unemploy ment benefits. They have the potential to help individuals re enter the labor market, and reduce their dependence on public support. Therefore it is impor tant that a basic level of financing for employment services is guaranteed by the budget. ·However, in countries where ALMP financing is limited, as in Russia, the focus of ALMPs should be on the most cost effective programs, such as job counseling and job information services should be increased. Emphasis on direct job cre ation programs should be reduced. Efforts to help the most disadvantaged workers (older, experienced workers, with obsolete skills) should increase. The use of employment quotas that state that individuals should have a job before being trained should be discontinued. ·Empirical profiling of users, currently being considered for introduction, may be useful for assessing what programs work best for particular groups -- but the 140 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY benefits and costs should be evaluated in Russia--on a pilot basis--prior to introduction because it is an administratively complex program to implement. ·The focus of employment services should be to help individuals find jobs them selves rather than helping preserve or create new jobs. Political pressure on employment agencies to contain unemployment is therefore misplaced. ·Private provision could be introduced as the sector develops so that market information can be used to match workers to training programs. Private providers should be regulated, however, so that potential abuse is restricted. Financing and administration. The report stresses that adequate financing of the program and its effective administration and monitoring are essential for its success. ·The report cautions that the general revenue financing of passive and active pro grams, introduced in 2001, will not necessarily reduce arrears or regional inequity of benefit. The Child Allowance Program, which is now federally financed, continues to have these problems. Therefore, adequate and certain financing of the program is required no matter the source of financing. At the same time, it is important that the program is designed to take into account the Government's fiscal constraints and that it uses scarce budgetary resources effectively. It is also important for the Government to provide a transparent allocation mechanism for transferring program resources to regions. Finally, the administration of both active and passive programs requires considerable attention to appropriate remuneration and training of staff, and their alloca tion across regions. ·The report finds that an adequately financed safety net for workers is possible in Russia. The simulated cost of the benefit program with a 30 percent replace ment rate (30 percent coverage, using 1999 data) would be approximately equal to 0.34 percent of GDP. Total costs of the program, including ALMP ben efits, would be 0.44 percent of GDP well within the scope of Russia's level of income. (These costs would be well below the costs of similar programs in advanced CEE countries of 1.1 percent of GDP in 2000). The increase in ben efits should be done gradually, as resources are released from improvements in the targeting/phase out of other social protection programs (privileges, hous ing allowances, for example). Social support restructuring. A combination of unemployment benefits, ALMP and severance benefits has proved important in downsizing the coal sector in Russia and also is widely used internationally to facilitate restructuring. It could therefore be used for downsizing in other sectors and regions in Russia (for example, regions with a high share of the industrial, overstaffed state sectors, or one company towns, or other over manned state sectors). The development of a strategy for identifying priority areas for restructuring and social programs for affected workers would be an impor tant first step in this direction.. The main elements of this strategy might include the following: (a) identification of the enterprises to be restructured, and the demo graphic and work skills of their personnel; (b) agreements on parameters of a social SAFETY NETS FOR WORKERS 141 support package (determine its scope, costs, source of financing, and administration drawing on existing mechanisms where possible); (c) stakeholder involvement; (d) a public information campaign; and (e) monitoring and evaluation mechanisms. The latter could help to ensure that workers are not rehired via a "back door" and that workers who have difficulty in re entering the labor market are identified early and given targeted assistance to ensure they do not slip into poverty. If enterprises have social infrastructure (schools, clinics), divestiture of these assets might also be moni tored to ensure that it has been successfully transferred to municipalities. ANNEX I EMPLOYMENT Figure I. 1. Changes in GDP in Transition Countries (Real GDP of 1999, 1989 = 100) Poland Slovenia Slovak Rep. Hungary CzechRep. European transition economies Albania Uzbekistan Eurasian transition economies Belarus Croatia Estonia Romania FYRMacedonia Bulgaria Lithuania Kyrgyz Rep. Kazakhstan Latvia RussianFederation Turkmenistan Azerbaijan Tajikistan Armenia Ukraine Georgia Moldova 20 40 60 80 100 120 140 Source: World Bank (2000b) 144 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY 9% 7% 20% 25% 30% Productivity 19902000 2% 1% Russia, 12% 14% 14% in Employment ages W and, GDP 29% 11% 8% 36% 39% Real Productivity Employment,, wages GDP 36% 5% 60% 57% 22% in CPI Change 2001b) (2000b, change Cumulative I.1. Goskomstat ableT Cumulative 1992­98 1998­2000 1990­2000 1990­98 1999­2000 Source: ANNEX I 145 2.5 5.4 0.7 4.7 60.5 36.0 14.2 21.8 4,84 Russia 11,15 1,51 9,64 8.1 4.0 0.8 4.8 51.6 24.8 7.7 32.5 34.40 44.78 1.43 43.35 Slovenia Countries Rep. CEE 3.6 5.8 0.6 5.2 2.7 8.3 20.0 28.3 26.04 41.20 7.19 48.39 percent) Slovak Select (in and 1.2 4.5 0.7 5.2 10.1 50.5 4.1 54.6 13.51 57.76 6.44 51.32 Poland Russia, countries CEE 8.0 2.2 1.9 0.3 32.1 14.6 21.0 35.6 42.05 34.98 14.69 49.67 Estonia Productivity advanced (2000). growth) and and 5.8 4.1 3.5 7.5 2.7 7.3 GDP 15.2 22.5 growth 4.98 30.30 7.52 37.82 ages, Hungary Goskomstat W Russia GDP positive (2001); of positive Republic 8.2 4.1 0.7 3.3 years 12.7 0.9 11.1 10.2 of 33.85 12.60 8.16 20.76 indicators; Employment,, year SilvaJauregui two Czech and 1990 first GDP market (first in since since Paramo, labor growth key Sanchez growth growth Changes of annual I.2. Riboud, ableT Growth verageA ages ages ages W GDP Employment Productivity Cumulative W GDP Employment Productivity Cumulative W GDP Employment Productivity Sources: 146 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.3. Unemployment Trends, 1992 2000 Panel A October October October October October October November November 1992 1993 1994 1995 1997 1998 1999 2000 A. Unemployment rates, percent LFS Unemploy ment rate 5.2 5.9 8.1 9.5 11.8 13.3 12.9 10.0 Gender Male 5.2 5.9 8.3 8.7 12.2 13.6 13.0 10.4 Female 5.2 5.8 7.9 9.2 11.5 13.0 12.8 9.6 Age groups < 20 years 20.4 21.1 26.5 28.7 41.4 46.3 36.5 35.1 20 24 9.7 10.6 12.8 15.3 18.9 22.5 20.0 16.6 25 29 5.5 6.7 9.2 11.4 12.7 14.2 14.0 10.7 30 34 4.1 5.3 8.0 9.2 11.7 12.9 13.1 9.6 35 39 3.6 4.5 6.7 8.2 11.0 12.1 12.0 9.4 40 44 3.1 4.0 6.3 7.0 9.3 10.7 10.3 8.6 45 49 3.1 3.8 5.4 6.5 8.3 9.6 10.0 7.3 50 54 2.9 3.2 5.2 5.8 7.7 8.8 9.3 6.2 55 59 3.8 3.7 5.3 5.7 8.1 8.8 10.0 8.5 60 72 5.6 4.5 4.9 5.2 6.9 8.7 10.7 8.0 Level of education University 3.3 3.5 4.8 5.0 5.7 7.1 7.1 5.3 Technical/vocational 4.5 5.3 7.3 8.3 10.2 11.4 10.6 7.9 General secondary 5.9 6.7 9.5 11.6 14.2 16.2 16.1 13.5 Basic secondary 6.6 7.8 10.7 12.9 17.6 19.4 19.9 16.5 Elementary 4.0 4.2 6.4 7.3 15.9 17.5 23.4 17.6 B. Duration of unemployment Average duration (in months) 4.1 5.7 6.6 7.4 8.8 9.9 9.7 9.1 C. Previous experience (percent) Had previous labor experience 79.9 81.3 83.6 83.2 88 85.9 80.6 81.9 Job losers 20.9 22.8 28.9 28.3 34 37.1 32.6 26.8 Job quitters 34.8 40.4 39.3 39.4 25.0 22.2 21.1 26.4 Other reasons 24.2 18.0 15.4 15.5 28.9 26.6 26.9 28.6 Did not have labor experience 20.1 18.7 16.4 16.8 12.0 14.1 19.4 18.1 Note: Sample includes respondents aged 15 72. Source: LFS figures reported in Goskomstat 1999a, 2000c . Panel B 1992 1993 1994 1995 1997 1998 1999 2000 Russia 5.2 5.9 8.1 9.5 11.8 13.3 12.9 10.0 United States 7.5 6.9 6.1 5.6 4.9 4.5 4.2 4.0 European Union 9.1 10.6 11.0 10.6 10.4 9.8 9.1 8.2 OECD Total 7.1 7.8 7.7 7.4 7.0 6.8 6.7 6.3 Source: LFS figures reported in Goskomstat (1999a, 2000c); OECD (2000). ANNEX I 147 Table I.4. Unemployment Growth Rates, Select CEE Countries and Russia 1993 1994 1995 1996 1997 1998 1999 2000 Czech Republic 3.9 3.8 4.1 3.9 4.8 6.5 8.8 Estonia 6.5 7.6 9.7 10 9.7 9.9 11.7 Hungary 11.9 10.7 10.2 9.9 8.7 7.8 7.1 Poland 14.9 16.5 15.9 14.3 11.5 10.6 12 Slovak Rep. 12.2 13.7 13.1 11.1 11.6 12.5 17.1 Slovenia 9.1 9 7.4 7.3 7.1 7.9 Russia 5.9 8.1 9.5 9.7 11.8 13.3 12.9 10 Source: OECD CCET Labor Market Database 1990 97. Table I.5. Composition of Unemployed by Duration of Unemployment Panel A: Russia (LFS, percent, 1992 2000) < 1 month 1 3 3 6 6 9 9 12 12+ 1992 October 26.8 29.6 19.6 7.6 5.3 11.1 1993 October 17.5 19.3 28.7 8.6 7.7 18.2 1994 October 11.9 16.7 29.9 10.3 8.3 23.0 1995 March 8.7 12.8 27.7 12.5 10.2 28.1 1995 October 10.4 14.8 26.3 10.0 8.9 29.6 1996 March 7.4 10.3 26.8 12.3 10.7 32.5 1997 October 7.8 15.9 15.8 10.7 11.6 38.1 1998 October 6.1 16.0 15.9 10.3 10.8 40.9 1999 February 4.5 12.4 14.7 10.8 11.2 46.6 1999 May 6.9 12.4 11.2 9.4 10.5 49.7 1999 August 7.9 15.7 12.5 8.6 10.5 44.8 1999 November 6.8 14.1 13.6 8.2 10.0 47.2 2000 February 6.6 12.1 12.7 9.9 9.4 49.4 2000 May 8.7 10.3 11.8 7.4 11.4 50.4 2000 August 9.0 17.9 11.7 7.6 10.7 43.0 2000 November 8.1 16.4 14.1 8.8 10.2 42.3 Note: Sample includes respondents aged 15 72. Source: LFS figures reported in Goskomstat 1999a, 2000c. Panel B: Percentage of long term unemployment (12 months or more) in total unemploy ment, selected OECD countries, 1997 Canada 12.5 Poland 38.0 Czech Republic 30.5 Spain 55.5 Finland 29.8 Sweden 33.4 France 41.2 Turkey 41.6 Germany 50.1 United Kingdom 38.6 Hungary 51.3 United States 8.7 Italy 66.3 European Union 15 51.3 Japan 21.8 OECD Total 34.7 Source: OECD 1999. 148 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.6. Unemployment, by Reason for Leaving and Duration of Unem ployment, Feb. 2000 Nov. 2000 Out of total Total (000s) Average Unemployed during (mos) time looking for a job From (mos) Less From From From 12 and 9 to than 1 1 to 3 3 to 6 6 to 9 more 12 Total: 000s 7,515 604 1,053 946 639 781 3,493 9,6 percent 8.0 14.0 12.6 8.5 10.4 46.5 Out of them, with 6,105 7.6 13.6 12.4 8.5 10.3 47.6 9.7 working experience: Out of them, leave previous job because of: Layoff, liquidation of 153 3.9 9.1 11.0 8.3 11.4 56.4 11 enterprise Leave (nonforceful) 512 10.3 18.4 13.4 9.4 9.7 38.9 8.6 Exhaustion of con 744 12.3 22.0 18.9 10.6 9.7 26.5 7.2 tract duration Retirement 258 5.3 7.1 9.4 5.0 10.9 62.1 11.4 Residence place 856 7.7 14.2 15.4 8.5 9.2 44.6 9.3 change Health reasons 154 6.6 10.7 11.1 7.8 7.4 56.8 10.6 Personal/family rea 1,361 7.8 14.7 10.3 7.5 9.7 49.8 9.9 sons Discharge from army 959 28.2 10.3 15.4 7.7 7.7 28.2 6.6 Other reasons 1,107 12.6 14.8 12.4 8.2 11.0 40.7 8.8 Without working expe 1,410 9.8 15.8 13.5 8.5 10.6 41.8 9 rience Source: Goskomstat, Labor Force Survey, November, 2000 ANNEX I 149 Table I.7. Unemployment, by Age and Duration of Unemployment, Feb. 2000 Nov. 2000 Out of total Total Average (000s.) Unemployed during (mos) time looking for a job 12 (mos) Less From From From From and than 1 1 to 3 3 to 6 6 to 9 9 to 12 more Total: 000s 7 515 604 1,053 946 639 781 3,493 9.6 percent 8.0 14.0 12.6 8.5 10.4 46.5 Out of them, aged (years): before 20 530 17.2 25.8 17.5 9.2 10.2 20.0 6.2 20­24 1,309 10.2 16.6 15.0 9.2 11.3 37.6 8.6 25­29 961 7.1 13.8 13.1 8.8 10.8 46.3 9.6 30­34 909 7.7 14.1 12.4 9.4 8.8 47.5 9.6 35­39 1,072 7.0 13.1 12.1 7.7 10.2 49.9 10 40­44 1,001 7.4 12.6 10.8 8.6 9.6 51.1 10.1 45­49 801 5.6 12.1 10.9 8.0 11.9 51.7 10.3 50­54 453 5.7 9.1 10.8 6.8 9.9 57.4 10.9 55­59 269 4.5 6.7 9.3 8.2 11.2 59.9 11.4 60­64 161 4.3 8.1 9.3 8.1 9.9 60.2 11.3 65­72 51 2.0 3.9 5.9 2.0 5.9 76.5 12.9 Source: Goskomstat, LFS, November 2000. 150 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.8. Unemployment, by the Last Working Place and Duration of Unemployment, Feb. 2000 Nov. 2000 Out of total Total (000s.) Unemployed during (mos) Average time looking for a job From 12 (mos) Less From From From 9 to and than 1 1 to 3 3 to 6 6 to 9 12 more Total: (000s) 7,515 604 1053 946 639 781 3493 9.6 percent 8.0 14.0 12.6 8.5 10.4 46.5 Out of them: With working experi 6,105 7.6 13.6 12.4 8.5 10.3 47.6 9.7 ence, as: Executives of all levels, 153 8.5 9.2 10.5 14.4 8.5 48.4 10 including top managers Specialists (highest 512 7.4 13.1 13.1 7.2 12.1 46.9 9.7 level of qualification) Specialists (average 744 7.4 11.0 10.9 8.3 10.3 51.9 10.2 level of qualification) Employed on informa tion preparation, docu 258 8.1 7.4 11.6 7.4 12.0 53.5 10.6 menting, counting Employed in services, utilities, retail trade and 856 8.3 14.6 13.2 8.4 10.6 44.9 9.4 similar sectors Qualified workers of agriculture, forestry, 154 6.5 12.3 12.3 9.1 9.7 50.6 10.1 hunting, fishery Qualified workers of industrial enterprises, craft arts, construction, 1,361 7.2 14.8 11.5 7.9 10.7 47.8 9.7 transportation, com munication, geology. Machine operators and 959 7.9 16.1 13.6 9.1 10.2 43.4 9.2 mechanics Workers without quali 1107 7.5 13.4 13.0 8.9 8.9 48.4 9.7 fication Without working expe 1410 9.8 15.8 13.5 8.5 10.6 41.8 9 rience With qualification 602 8.0 17.3 14.0 9.3 10.8 40.7 8.9 Without qualification 808 11.1 14.7 13.1 7.9 10.4 42.6 9 Source: Goskomstat, LFS, November 2000. ANNEX I 151 Table I.9. Unemployment, by Education and Duration of Unemployment, Feb. 2000 Nov. 2000 Out of them Total, (000s) Average Unemployed during (mos) time looking for a job From Less From From From 12 and (mos) 9 to than 1 1 to 3 3 to 6 6 to 9 more 12 Total: (000s) 7515 604 1,053 946 639 781 3,493 9.6 percent 8.0 14.0 12.6 8.5 10.4 46.5 Out of them with education: University (³4 years) 755 8.7 14.3 13.6 9.3 11.1 42.9 9.2 College (<4) 259 13.1 14.3 13.5 9.7 9.7 40.2 8.7 High vocational 2,001 7.5 13.1 12.4 9.0 10.3 47.5 9.7 (teknikum) Vocational (PTU) 897 7.0 16.8 13.8 8.8 10.7 42.7 9.2 High school 2,337 7.5 13.9 12.1 8.2 10.6 47.7 9.7 Secondary school 1,062 9.3 13.7 12.5 6.9 9.5 48.2 9.6 Primary school/with 204 6.9 11.8 9.8 10.3 9.8 52.0 10.2 out primary school Source: Goskomstat, LFS, November, 2000. 152 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.9a Determinants of Survey Unemployment Rates LFS Unemployment Rate Regression statistics Multiple R 0.930 R square 0.864 Adjusted R square 0.829 Standard error 2.665 Observations 79 ANOVA df SS MS Regression 16 2801.55 175.10 Residual 62 440.38 7.10 Total 78 3241.93 Coefficients Standard t Stat Error Intercept 7.01 4.19 1.67 Per capita expenditures 0.00 0.00 4.01 Percent urban 12.71 4.15 3.06 Birth rate (per 1000 population) 1.37 0.23 5.86 Industy/ GDP 4.75 1.85 2.57 Refugees/Population 182.40 34.22 5.33 Poverty rates (official data) 0.07 0.03 2.32 Northern area 3.80 1.52 2.49 North Western area 1.71 1.53 1.12 Volgo Vyatsky area 1.40 1.45 0.96 Central Tchernozemny area 2.04 1.64 1.25 Povolzhsky area 0.36 1.24 0.29 Northern Caucasus 6.32 1.58 4.00 Ural area 0.69 1.39 0.49 West Siberian area 0.08 1.36 0.06 East Siberian area 1.78 1.54 1.15 Far East. 2.06 1.24 1.66 Source: Goskomstat Data Table I.10. Labor Force Participation Rates, 1992 2000 (Percent) October October October October October October November November 1992 1993 1994 1995 1997 1998 1999 2000 Labor force participation rate, p.c. 70.3 68.1 65.4 64.8 62.3 61.0 63.8 63.2 Gender Male 77.6 75.6 72.8 72.1 69.4 68.1 70.4 69.4 Female 63.7 61.3 58.7 58.3 55.9 54.7 57.9 57.6 Location Urban 71.9 69.7 66.9 66.6 64.3 63.4 66.0 65.4 Rural 65.6 63.4 60.8 59.8 56.4 54.2 57.5 56.6 Continued on next page ANNEX I 153 Table I.10 -- Continued October October October October October October November November 1992 1993 1994 1995 1997 1998 1999 2000 Age groups < 20 years 31.1 29.3 23.2 23.9 16.5 14.3 15.4 11.9 20 24 79.4 78.0 76.9 77.2 71.1 68.1 70.0 68.2 25 29 90.3 88.4 87.1 87.0 84.4 83.4 86.3 86.3 30 34 92.9 90.9 89.5 89.0 87.0 85.9 88.0 88.7 35 39 93.9 92.8 91.2 90.6 88.9 88.4 90.0 90.5 40 44 94.6 92.9 91.4 90.7 89.2 88.5 89.5 89.9 45 49 92.9 91.6 89.7 89.3 87.7 86.8 88.7 88.0 50 54 85.1 82.4 79.1 78.4 80.1 78.8 83.5 82.8 55 59 57.5 54.9 49.7 47.9 45.8 44.2 48.7 48.3 60 72 18.3 15.2 11.7 11.3 9.9 9.1 14.6 12.8 Note: Sample includes respondents aged 15 72. Sources: LFS figures, reported in Goskomstat (1999a, 2000c). Panel B 1992 1993 1994 1995 1997 1998 1999 2000 Russia, age 15 72 70.3 68.1 65.4 64.8 62.3 61.0 63.8 63.2 Russia, working age population 84.2 82.2 81.1 80.3 77.2 76.0 78.0 77.7 United States 66.4 66.3 66.6 66.6 66.8 67.1 67.1 67.1 European Union 68.0 67.7 67.7 67.6 67.8 68.1 68.6 69.1 OECD Total 67.8 67.7 67.8 67.8 68.0 68.2 68.3 ... Notes: Labor force participation rates are not fully comparable across countries because of different defini tions of the working age population. In Russia it is 16 54 for women and 16 59 for men; in the United States it is 16 years of age and more; and in most other countries it is 16 64 years. Table I.11. Labor Force Status of the Russian Population (LFS, 1992 2000) Employment Unemployment Out of labor Total (share) (share) force Population (share) (000s) 1992 October 0.667 0.036 0.297 106 590 1993 October 0.641 0.040 0.319 107 112 1994 October 0.601 0.053 0.346 107 839 1995 March 0.589 0.054 0.357 107 846 1995 October 0.587 0.061 0.352 109 285 1996 March 0.576 0.062 0.363 109 284 1997 October 0.549 0.074 0.377 109 343 1998 October 0.529 0.081 0.390 109 354 1999 February 0.534 0.094 0.372 110 217 1999 May 0.545 0.083 0.372 110 217 1999 August 0.558 0.079 0.363 110 217 1999 November 0.556 0.083 0.362 110 217 2000 February 0.550 0.078 0.373 109 587 2000 May 0.567 0.067 0.366 110 218 2000 August 0.570 0.064 0.366 110 310 2000 November 0.568 0.063 0.368 110 310 Note: Sample includes respondents aged 15 72. Sources: LFS figures, reported in Goskomstat (1999a, 2000c). 154 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY sale Home tion 117 659 541 076 073 739 532 032 443 377 202 233 239 339 955 produc for 2 2 2 2 2 2 2 2 2 own employment 568 723 404 678 562 182 889 333 242 42 90 147 155 204 603 for 2 2 2 1 1 31 other have Subsistence consumption 038 210 962 800 027 654 565 588 745 428 agriculture 31 3800 8352 7182 4416 4820 8 7 3 6 1 2 < Didn't (000s) otalT 485 674 127 170 455 959 631 327 395 446 946 945 982 288 986 6 7 7 6 9 1 1 3 13 12 12 11 sale 772 300 990 841 480 449 063 614 209 05 48 449 928 049 377 19992000 Home tion 2 3 2 2 2 3 3 2 2 1 produc for own more hours 245 443 119 238 213 985 672 131 005 4 55 68 19 152 202 Agriculture, for 1 1 employment 31 and other Subsistence consumption 054 071 493 747 947 966 478 937 233 83 316 113 569 040 211 2000c). 4 9 4 4 9 3 7 1 2 2 1 Subsistence Had agriculture 31 < hours 10 10 (1999a, and otalT 071 814 602 826 640 400 213 682 437 1572. 137 820 193 820 485 278 7 7 7 6 14 13 15 13 01 1 3 3 1 aged Goskomstat in Production otalT 556 488 729 996 094 357 843 009 623 583 766 138 802 773 264 13 28 25 14 15 28 24 13 02 1 2 4 4 2 4 respondents reported Home .2000 includes figures, I.12. LFS February May August November February May August November years groups Sample ableT <20 2029 3039 4049 5059 6072 1999 1999 1999 1999 2000 2000 2000 2000 Feb.2000Nov Age Note: Sources: ANNEX I 155 Table I.13. Subsistence Agriculture: Working Hours, 1992 2000 October October October October October October Novem November 1992 1993 1994 1995 1997 1998 ber 1999 2000 Usual working hours 39.8 39.4 38.8 38.8 39.1 39.1 39.3 39.5 per week Distribution of employed by hours of work, p.c. < 21 hours per week 2.0 1.7 2.1 2.1 2.1 2.1 2.8 2.5 21 30 3.2 2.5 2.6 2.4 2.3 2.5 2.8 2.2 31 40 80.3 85.9 92.4 93.5 89.4 86.6 84.1 84.7 > 40 14.5 9.9 2.8 2 6.3 8.8 10.2 10.5 Actual working hours per 37.2 37.3 36.4 36.4 37.5 37.4 38.3 38.6 week Distribution of employed by hours of work, p.c. < 21 hours per week 2.2 2.1 2.5 2.6 2.6 2.8 3.1 2.7 21 30 3.4 3.2 3.7 3.6 3.3 3.6 3.3 2.6 31 40 72.3 76.8 80.2 81.3 79.0 77.2 75.9 77.2 > 40 15.4 12.2 7.1 5.9 10.5 11.9 13.9 14.1 Note: Sample includes respondents aged 15 72. Sources: LFS figures, reported in Goskomstat (1999a, 2000c). Table I.14. Difference in Employment and Unemployment Rates With and Without Self Employment Unemploy Unemploy Employment Employment ment rate ment rate rate rate (without) (with) (with) (without) 1999 Feb 0.13 5.9 0.66 0.53 1999 May 0.09 8.1 0.80 0.55 1999 August 0.09 9.5 0.79 0.56 1999 Nov 0.11 9.7 0.69 0.56 2000 Feb 0.10 11.8 0.69 0.55 2000 May 0.08 13.3 0.82 0.57 2000 August 0.07 12.9 0.80 0.57 2000 Nov 0.08 10 0.69 0.57 Source: Goskomstat. 156 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.15. Working Hours among Employees and Self Employed, 1999 Feb. 1999 -- Nov. 1999 Total < 21 21 30 31 40 >40 Average per worker Usual working hours per week All employed 60 408 1 681 1 808 50 530 6 390 39.4 Percent 100 2.8 3.0 83.6 10.6 Employees 55 966 1 310 1 450 47 929 5 277 39.4 Percent 100 2.3 2.6 85.6 9.4 Self employed 4 442 370 358 2 601 1 113 39.7 Percent 100 8.3 8.1 58.6 25.1 Actual working hours per week All employed 60 408 1 898 2124 44 760 8 847 38.0 Percent 100 3.1 3.5 74.1 14.6 Employees 55 966 1 484 1751 42 721 7 368 37.9 Percent 100 2.7 3.1 76.3 13.2 Self employed 4 442 414 373 2 040 1 478 39.7 Percent 100 9.3 8.4 45.9 3.7 Feb. 2000 -- Nov. 2000 Total < 21 21 30 31 40 >40 Average per worker Usual working hours per week All employed 62 180 1 565 1 537 52 821 6 259 39.6 Percent 100 2.52 2.47 84.95 10.07 Employees 57 928 1 306 1 465 45 254 7 427 38.3 Percent 100 2.25 2.53 78.12 12.82 Self employed 4 286 365 288 2 530 1 103 39.8 Percent 100% 8.5% 6.7% 59.0% 25.7% Actual working hours per week All employed 62 180 1 705 1 759 47 157 8 942 38.4 Percent 100 2.74 2.83 75.84 14.38 Employees 57 928 1 306 1 465 45 254 7 427 38.3 Percent 100 2.25 2.53 78.12 12.82 Self employed 4 252 400 294 1 903 1 515 40.1 Percent 100 9.41 6.91 44.76 35.63 Note: Sample includes respondents aged 15 72. Sources: LFS figures, reported in Goskomstat (1999a, 2000c). ANNEX I 157 Table I.16. Average Working Hours per Week by Industry (LFS, 1999 2000) Usual Hours Actual Hours Feb. 1999 -- Feb. 2000 -- Feb. 1999 -- Feb. 2000 -- Nov. 1999 Nov.2000 Nov. 1999 Nov.2000 Total 39.4 39.6 38.0 38,4 Industry 40.0 40.1 37.9 38.5 Agriculture & forestry 39.4 39.4 39.3 39.7 Transportation 40.4 40.6 39.1 39.5 Communications 38.7 38.5 36.6 36.7 Construction 40.3 40.6 39.2 39.7 Trade 40.8 41.0 40.1 40.5 Municipal utilities 39.8 39.6 38.4 38.4 Health services 38.5 38.6 36.6 36.9 Education 34.7 34.9 32.8 33.1 Culture and art 37.8 38.4 36.0 36.7 Science 39.7 39.7 37.9 38.0 Finance, credit, and 39.6 39.7 38.4 37.9 insurance Public administration 40.2 40.3 39.5 39.5 Other industries 40.1 40.0 38.6 39.4 Sources: LFS figures, reported in Goskomstat (1999a, 2000c). 158 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.17. Trends in Hours Worked (LMS, 1994 2000) 1994 1995 1996 1998 2000 November -- October -- October -- November -- October -- December November November December November Hours worked at the primary job Mean and standard deviation 159.8 167.1 168.4 164.0 170.3 [61,7] [63,5] [60.0] [58.9] [60.3] Year effects 6.066*** 7.508*** 4.547*** 8.969*** (2.73) (3.32) (3.04) (3.66) Hours worked in all jobs Mean and standard deviation 157.0 164.2 165.4 159.3 165.5 [70.3] [71.9] [68.9] [69.9] [72.4] Year effects 7.443** 8.729*** 2.501 7.866*** (2.97) (3.42) (1.48) (2.86) ** Significant at 5% level. *** Significant at 1% level. Notes: Sample consists of respondents aged 15 72. Standard deviations are in brackets; t statistics are in parentheses. The year effects are estimated relative to 1994 from OLS (Ordinary Least Squares) equation, in which other controls included are gender, age, years of schooling, 38 district dummies, daily time trend (interview date) and interactions between district dummies and the time trend. Table I.18. Time Budget of Industrial Workers, 1980 96 1980 1985 1990 1991 1992 1993 1994 1995 1996 Number of days in the year 366 365 365 365 366 365 365 365 366 of which: Holidays and weekends 97 97 102 105 108 107 110 110 110 Days worked 229 228 225 220 213 205 189 193 192 Days not worked 40.1 40 38.3 39.7 47.6 52.4 65.5 61.9 63.7 of which by reason: Vacations 22.5 23.1 21.6 22.7 26.6 26.8 26.3 25.8 25.5 Sickness 11.8 11.2 12.0 11.9 10.0 10.5 9.9 10.7 9.5 Absences allowed by law 4.0 4.6 3.3 2.9 2.5 2.1 1.9 1.5 1.3 Absences allowed by 1.0 0.5 0.8 1.2 4.6 4.5 4.6 4.4 4.4 administration Unpermitted absences 0.6 0.5 0.4 0.5 0.4 0.4 0.4 0.4 0.2 Work stoppage (whole days) 0.2 0.1 0.2 0.5 3.5 8.1 22.4 19.1 22.8 Sources: Goskomstat (1993, 1997). ANNEX I 159 Table I.19. Involuntary Leaves and Short Time Employment, 1995 2000 1995 1996 1997 1998 1999 2000 Involuntary leaves Number of employees (thousands) 2401 7538 5194 4742 3325 2175 Percentage of employees affected 4.8 15.8 11.5 11.1 7.9 5.2 Duration (millions of person hours) 195.7 2393.8 1652.2 1472.5 785.9 472.0 Duration per employee on leave (hours) 82 318 318 311 236.4 217.1 Duration per employee (hours) 3.9 50.3 36.7 34.4 18.8 11.3 Part time involuntary employment Number of employees (thousands) ... ... ... 4306 11.5 5.7 Percentage of employees affected ... ... ... 10.1 2728 1499 Nonworked time (millions of person hours) ... 1321.4 862.6 809.5 481.8 240.3 Nonworked time per part time involuntary employee (hours) ... ... ... 188.0 176.6 160.4 Nonworked time per employee (hours) ... ... 19.2 18.9 6.5 3.6 Note: Excludes small enterprises. Sources: Goskomstat (1999c, 2001a). Table I.20. Incidence and Persistence of Involuntary Leaves (RLMS) Expected probability of involuntary unpaid ULVDUM Sample leave (dummy) 1994 1995 1996 1998 2000 Full Unconditional Mean .112 .063 .078 .083 .033 cross (ULVDUM t) (N=4745) (N=4398) (N=4186) (N=3931) (N=4151) section Mean(ULVDUMt ½ Panel .276 .366 ULVDUMt ... ... ... 1 = 1) for t, t 1 (N=388) (N=205) Mean (ULVDUMt ½ Panel .279 .296 .170 ULVDUMt ... ... 2 = 1) for t, t 1 (N=340) (N=206) (N=218) Mean (ULVDUMt ½ Panel .276 .443 .382 .300 i ULVDUMt for t, t 1, ... i = 1) (N=388) (N=88) (N=34) (N=10) t 2 Note: ULVDUMt = 1 if an employed respondent reports unpaid leaves on his/her primary job in year t. Sample size is shown in parentheses for number of valid responses for ULVDUM; sample sizes vary pri marily because of attrition and replacement in the RLMS panel, and secondarily because of missing val ues for some respondents. Source: Calculations from RLMS. 160 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.21. Labor Market Transitions (RLMS, 1994 2000) Labor Force Status in 1996 1996­1998 With job Out of Employ Unem Total 1994 but not Labor ment ployment at work Force Employment 0.799 0.060 0.038 0.103 0.524 With job but not at work 0.620 0.150 0.053 0.177 0.051 Unemployment 0.402 0.035 0.239 0.324 0.044 Out of labor force 0.103 0.009 0.054 0.834 0.382 Total 1996 0.507 0.044 0.054 0.395 1.000 N = 5944 Labor Force Status in 1998 1996­1998 With job Out of Employ Unem Total 1996 but not labor ment ployment at work force Employment 0.782 0.057 0.053 0.108 0.502 With job but not at work 0.646 0.175 0.062 0.117 0.045 Unemployment 0.403 0.021 0.224 0.353 0.056 Out of labor force 0.103 0.007 0.045 0.845 0.397 Total 1998 0.485 0.041 0.060 0.415 1.000 N = 6073 Labor Force Status in 2000 1998­2000 With job Out of Employ Unem Total 1998 but not labor ment ployment at work force Employment 0.826 0.054 0.030 0.091 0.480 With job but not at work 0.636 0.170 0.067 0.127 0.039 Unemployment 0.482 0.028 0.185 0.305 0.060 Out of labor force 0.127 0.007 0.045 0.821 0.421 Total 2000 0.504 0.037 0.047 0.412 1.000 N = 6510 Note: Each cell measures the probability of transition from labor force status i to labor force status j. Source: Calculations from RLMS. ANNEX I 161 Table I.22. Composition of Employment by Industry, 1970 99 (Percent) Change 1970 1975 1980 1985 1990 1995 1998 1999 1985 99 Industry 33.9 33.0 32.5 32.3 30.3 25.9 22.2 22.4 30.7 Agriculture/forestry 17.6 15.6 15.0 14.3 13.2 15.1 14.1 13.7 4.2 Construction 8.9 9.6 9.6 9.4 12.0 9.3 7.9 7.9 15.9 Transport/ 8.9 9.3 9.6 9.8 7.7 7.9 7.6 7.6 22.4 communications Trade 7.7 8.2 8.3 8.3 7.8 10.1 14.5 14.6 75.9 Housing 3.2 3.6 3.9 4.1 4.3 4.5 5.4 5.3 29.3 Health services 4.8 4.9 4.8 5.0 5.6 6.7 7.0 7.0 40.0 Education, culture, 11.0 11.7 12.3 12.6 13.3 13.5 13.1 13.0 3.2 art, and science Finance, credit, 0.4 0.4 0.5 0.5 0.5 1.2 1,1 1.2 140.0 and insurance Public 1.9 2.0 1.8 1.9 2.4 3.0 4.4 4.5 136.8 administration Other industries 0.9 1.1 1.7 1.8 2.9 2.8 2.7 2.8 55.5 Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Goskomstat (2000b). 162 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY 1.9 2.0 1.8 2.0 2.0 2.0 1.9 2.1 2.2 2.1 Other industries 7.2 7.6 7.7 7.7 7.8 7.9 7.8 8.1 8.1 7.6 Public admin istration 1.4 1.4 1.2 1.3 1.2 1.3 1.2 1.2 1.3 1.3 and Finance, credit, insurance and 2.7 2.7 2.8 2.6 2.7 2.5 2.7 3.1 2.6 2.6 Culture, art, science tion 9.5 9.7 9.7 9.4 9.7 9.3 8.9 9.4 Educa 10.2 10.0 Percent) 7.1 7.3 7.4 7.3 7.0 7.6 7.2 7.2 6.6 7.4 Health services 19972000, 4.1 4.3 3.6 3.8 3.6 4.1 3.9 3.6 3.5 3.9 Utilities (LFS, industries. Trade 12.4 12.7 12.0 12.6 12.3 13.0 12.4 12.9 13.3 13.7 across Industry by tion 7.2 6.6 6.6 7.2 6.5 6.2 6.1 6.4 6.5 6.4 Construc employment of com 9.1 9.2 8.9 9.1 9.3 9.0 8.7 8.5 9.0 8.8 tions Transport and munica Employment distribution of ture 9.0 11.2 10.6 11.7 10.5 11.2 10.4 10.7 11.0 11.2 Agricul percentage the 26.1 26.1 26.1 26.3 26.9 26.4 27.5 26.5 26.7 27.8 Industry Composition shows I.23. table October October February May August November February May August November The ableT 1997 1998 1999 1999 1999 1999 2000 2000 2000 2000 Note: ANNEX I 163 Table I.24. Changes in Industrial Composition between 1985 and 2000 (RLMS, percent) Industry categories 1985 1991 1998 2000 Change 1985 00 Energy/fuel 2.62 3.01 3.52 2.98 13.74 Metallurgy 2.53 2.62 2.33 2.62 3.56 Chemicals 1.80 1.49 1.16 1.30 27.78 Machine building 10.61 10.24 7.06 7.04 33.65 Military complex 3.71 3.33 2.64 1.84 50.40 Wood processing/ building materials 4.37 4.50 3.59 3.52 19.45 Light/food 5.70 5.45 4.83 4.77 16.32 Agriculture 14.34 13.05 10.05 10.20 28.87 Transportation 8.10 7.27 7.63 7.42 8.40 Construction 8.21 7.96 6.51 6.50 20.83 Trade 7.92 8.31 9.67 10.72 35.35 Finance/commerce 0.71 1.08 4.44 5.54 680.28 Housing 3.62 3.74 5.09 5.11 41.16 Health services 5.33 6.32 8.58 7.94 48.97 Education, art, and science 13.72 14.35 13.69 13.18 3.94 Public administration 4.82 5.54 6.99 6.61 37.14 Other industries 1.91 1.73 2.21 2.71 41.88 Total 100.0 100.0 100.0 100.0 N 4506 4621 4207 4461 Source: Calculations from RLMS, reported in Earle and Sabirianova (2002b). 164 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.25. Sectoral Reallocation in Russia, Transition Matrices (RLMS, 1985 98) Trans 1985­1998 Non Agricul portation/ Public Other Industry employ ture construc services services ment tion Industry 0.378 0.010 0.052 0.048 0.068 0.443 Agriculture 0.030 0.359 0.033 0.025 0.036 0.518 Transportation/ construction 0.083 0.014 0.365 0.044 0.081 0.414 Public services 0.062 0.006 0.017 0.484 0.070 0.362 Other services 0.075 0.017 0.023 0.071 0.322 0.492 Nonemployment 0.097 0.044 0.053 0.159 0.103 0.544 Trans 1985­1991 Non Agricul portation/ Public Other Industry employ ture construc services services ment tion Industry 0.732 0.013 0.034 0.033 0.035 0.153 Agriculture 0.028 0.705 0.024 0.041 0.017 0.185 Transportation/ construction 0.070 0.031 0.692 0.035 0.047 0.125 Public services 0.041 0.008 0.015 0.764 0.031 0.142 Other services 0.062 0.018 0.032 0.048 0.644 0.197 Nonemployment 0.126 0.043 0.052 0.133 0.058 0.588 Trans Non 1991­1998 Agricul portation/ Public Other Industry employ ture construc services services ment tion Industry 0.479 0.010 0.051 0.055 0.075 0.330 Agriculture 0.022 0.468 0.042 0.027 0.043 0.399 Transportation/construction 0.077 0.011 0.461 0.043 0.088 0.320 Public services 0.047 0.012 0.017 0.567 0.074 0.282 Other services 0.058 0.021 0.029 0.064 0.422 0.407 Nonemployment 0.045 0.019 0.020 0.092 0.053 0.771 Source: Calculations from 1998 RLMS, 1998. ANNEX I 165 771 432 257 260 197 196 121 350 1999 1002 2074 1008 784 472 258 266 205 194 120 366 1998 1162 2187 1095 827 712 288 280 238 201 132 455 1997 1611 1660 1291 841 713 297 300 252 209 137 472 1996 1523 1845 1144 198599, 855 883 342 304 355 207 258 550 1995 1683 1961 1380 Industry 919 815 366 316 472 212 242 536 1994 1676 2294 1378 by 378 357 586 227 287 748 1993 1012 2973 2966 1926 1625 Enterprises 944 460 466 513 279 394 653 1991 3417 3289 1740 1310 Firms. Industrial 935 516 533 591 290 236 736 1989 3794 3468 2002 1522 of Industrial of 930 555 592 622 306 238 767 1987 3805 3667 2161 1563 Registry 903 548 567 635 319 246 768 (Employment) 1985 3668 3737 2097 1544 Goskomstat Size the from verageA materials I.26. processing industries Calculations ableT earY ood Energy Fuel Metallurgy Chemicals Machinebuilding W Building Light Food Other otalT Source: 166 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ... 27.8 29.5 26.3 36.7 48.8 26.0 31.3 40.7 42.5 22.0 17.9 22.8 17.2 21.3 rates 2000 24.5 27.0 21.8 33.2 42.8 22.5 28.6 35.8 36.4 19.1 15.4 20.3 15.6 21.1 11.6 separation percent 1999 24.9 27.7 20.7 31.2 40.8 24.0 28.7 37.7 33.0 20.0 15.8 20.7 18.1 22.4 13.1 and ing hir rates, 1998 24.5 26.8 19.5 29.5 41.3 26.6 28.6 36.7 31.3 19.6 15.0 18.0 18.5 19.3 12.6 workers; 1997 23.9 27.0 18.0 19932000 28.7 42.5 24.7 27.9 33.3 29.5 17.7 13.5 16.7 17.9 19.9 12.5 Separation 1996 25.7 28.4 18.5 32.8 45.1 25.9 28.9 31.7 30.2 20.4 15.3 17.9 22.4 19.8 12.5 employed of 1995 27.4 32.0 19.6 30.0 45.9 29.6 28.6 24.8 30.8 18.8 14.1 16.5 23.1 17.0 13.8 Enterprises, number 1994 25.1 28.8 14.5 25.5 44.1 29.6 33.2 27.1 30.5 17.2 12.6 25.1 15.2 11.1 average Medium ... the 1993 26.9 30.1 20.3 36.1 47.7 24.2 31.0 39.1 41.7 20.8 16.6 24.5 17.0 20.5 to and year 2000 24.2 27.4 18.3 34.8 40.3 21.3 27.0 33.7 37.2 19.5 16.0 25.2 14.2 17.3 13.9 the Large in 1999 21.0 19.8 14.8 28.9 32.8 18.9 26.1 30.3 37.1 21.1 16.6 22.4 12.3 19.8 13.7 during hired Rates 1998 19.9 19.2 13.0 28.0 31.3 18.3 28.1 28.5 35.2 19.7 15.1 19.1 11.0 19.1 15.3 percent workers 1997 18.9 16.9 12.4 26.1 29.0 19.5 27.3 24.8 35.6 18.9 15.4 17.7 11.1 20.7 14.2 of rates, ratio 2001a). Separation 1996 22.6 21.1 12.7 31.5 39.4 24.7 29.0 26.7 39.7 22.1 17.1 19.7 12.8 25.2 17.1 and and Hiring 1995 20.8 18.2 12.4 31.9 34.9 23.0 27.5 29.2 39.4 20.9 17.0 16.9 11.4 30.5 22.3 percentage firms. 2000b, a as Hiring 1999d, 1994 21.1 20.1 11.3 28.1 38.7 26.5 32.1 22.5 36.2 18.0 15.2 12.1 22.3 18.3 computed mediumsized (1998c, are Annual retail and arts insurance and rates large I.27. services services and and for administration Goskomstat Hiring ableT given 1993 otalT Industry Agriculture Forestry Construction ransportationT trade Communication Wholesale Housing Health Education Culture Science Finance Public Notes: are Sources: ANNEX I 167 Table I.28. Hiring, Layoff, Quit, and Separation Rates from Survey Data, 1991 98 Year Hiring Layoffs Quits Total separations 1991 0.21 0.01 0.12 0.20 1992 0.20 0.02 0.15 0.26 1993 0.20 0.02 0.17 0.26 1994 0.18 0.03 0.19 0.28 1995 0.19 0.03 0.18 0.27 1996 0.17 0.04 0.18 0.27 1997 0.18 0.05 0.17 0.28 1998 0.18 0.04 0.18 0.26 Note: Employment separations because of death, entrance to army, and retirement are not counted as quits or layoffs. Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (1999). 168 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.29. Changes in Occupational Composition between 1985 and 1998 (RLMS, percent) One and two digit occupational 1985 1991 1998 Change categories 1985 98 Officials and managers Officials 0.20 0.09 0.14 30.0 Corporate managers 0.62 0.77 1.53 146.8 Small firm managers 0.66 0.91 1.98 200.0 Entrepreneurs and independent farmers 0.00 0.21 1.79 + Professionals Physicists, mathematicians, and engineers 6.47 5.87 3.59 44.5 Life science and health professionals 1.92 2.04 2.31 20.3 Teaching professionals 3.56 3.61 4.39 23.3 Business and law professionals 1.70 1.72 2.05 20.6 Other professionals 0.86 0.74 0.80 7.0 Associate professionals Technicians 3.62 3.44 3.82 5.5 Life science and health associate professionals 2.96 3.10 3.94 33.1 Teaching associate professionals 2.34 2.74 2.55 9.0 Finance and business associate professionals 1.48 1.64 1.77 19.6 Other associate professionals 4.84 4.85 4.70 2.9 Clerks Office clerks 5.85 5.65 5.03 14.0 Customer services clerks 1.37 1.68 1.91 39.4 Service workers Personal services workers 2.43 2.36 2.88 18.5 Catering services workers 1.97 2.08 0.99 49.7 Protective services workers 1.02 1.45 3.42 235.3 Salespersons 2.72 2.78 4.56 67.6 Craft workers Extraction and building trades workers 4.09 3.66 3.90 4.6 Metal and machinery workers 12.79 12.76 9.47 26.0 Other craft workers 2.72 3.27 2.95 8.5 Operators and assemblers Stationary plant operators 3.25 3.51 3.40 4.6 Machine operators and assemblers 3.09 2.51 2.12 31.4 Drivers and mobile plant operators 14.03 13.88 11.47 18.2 Elementary occupations 11.95 10.88 11.21 6.2 Military specialists 1.50 1.79 1.30 13.3 Total 100.0 100.0 100.0 N 4527 4704 4236 Note: The last column indicates the positive or negative changes in the share of each type of occupation. Source: Calculations from 1998 RLMS. ANNEX I 169 Table I.30. Mobility Trends in Russia (RLMS, 1985 1998) Inter Share Years Sectoral Inter firm Occupational Complex industry of complex mobility mobility mobility mobility mobility flows 1985--1998 0.306 0.485 -- 0.497 1985--1991 0.149 0.239 0.260 0.219 0.153 49.2% 1991--1998 0.262 0.416 0.415 0.422 0.296 56.1% 1991--1995 0.180 0.285 -- 0.287 1994--1998 0.174 0.261 0.280 0.284 0.183 49.1% 1994--1996 0.103 0.161 0.166 0.177 0.110 47.8% 1996--1998 0.101 0.161 0.177 0.166 0.096 39.9% Note: Sectoral, inter industry, inter firm, and occupational mobility are fractions of employed respon dents who changed their sector, industry, firm, and occupation, respectively, between the first year and the last year of the considered period. Complex mobility is defined as simultaneous changes in occupa tion, firm, and industry. 170 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY more 46­60 47.18 15.79 43.39 23.26 33.53 29.28 15.93 16.01 36.81 35.39 33.47 28.60 56.29 39.67 52.40 60.55 50.24 45.56 34.76 33.53 34.02 29.82 26.79 25.21 and years 20 26­45 10.02 3.77 8.30 5.17 8.83 7.37 3.26 2.74 5.66 5.86 6.18 6.39 9.68 7.41 8.62 10.78 10.67 7.16 8.97 4.01 5.34 5.48 4.91 3.30 tenure 46­60 71.57 45.23 46.28 70.82 54.75 62.58 64.33 44.32 43.95 60.35 51.09 50.60 45.68 76.77 59.54 66.85 28.05 76.65 72.66 69.03 58.11 56.04 53.16 44.08 39.96 40.55 with more Comparisons and 26­45 38.81 23.64 22.15 36.71 26.55 33.70 36.75 21.40 19.47 34.24 30.69 29.16 26.21 42.85 29.29 30.68 39.89 44.43 37.67 33.45 36.03 25.80 28.64 23.87 20.92 16.24 employed years of 10 Fraction International 4.54 7.54 46­60 11.32 3.41 4.51 14.70 7.01 9.90 11.82 11.15 13.12 9.76 14.03 4.68 6.85 8.64 3.56 2.97 13.09 5.33 9.77 12.22 15.21 16.23 12.72 16.81 enure,T of 26­45 15.42 15.23 20.69 9.64 7.61 29.38 16.47 20.14 23.81 14.40 19.14 15.32 19.74 11.39 16.34 20.44 8.19 4.32 28.76 18.19 14.83 20.00 20.74 23.19 21.61 24.95 less arseY and year on 1 25£ 48.89 35.98 42.97 29.06 23.46 75.73 53.46 40.80 51.62 30.93 43.58 41.59 52.39 44.87 36.49 47.16 29.55 25.36 76.50 44.94 39.65 52.23 48.13 47.84 36.93 49.44 Based RLMS. 46­60 17.92 11.60 10.55 17.93 13.04 14.28 14.38 10.27 10.54 14.82 13.51 13.49 12.33 19.64 17.31 16.37 19.91 21.10 17.83 17.33 14.48 14.17 13.73 12.16 11.57 11.49 on based Measures (years) 8.39 7.04 6.21 9.09 7.72 7.44 8.02 5.89 5.77 7.87 7.17 7.26 6.85 9.00 7.42 7.59 9.62 26­45 10.18 8.17 7.68 8.25 6.87 6.83 6.05 5.83 5.27 calculations Mobility tenure and job 25£ Job 1.67 2.38 2.29 3.06 2.62 1.42 1.88 2.39 1.97 2.52 2.11 1.86 1.62 1.84 2.37 2.18 3.26 2.57 1.22 2.14 2.56 2.02 1.63 1.55 1.95 1.52 (1999) Age Mean I.31. States States Burgess Kingdom 1994 1996 1998 2000 Kingdom 1994 1996 1998 2000 ableT omen France Germany Netherlands Italy Japan Spain Sweden United United Russia France Germany Netherlands Italy Japan Spain Sweden United United Russia W Men Sources: ANNEX I 171 Table I.32. Job Destruction and Job Creation Rates over Time (Percent), Select Transition Countries Bulgaria Czech Rep. Hungary Poland Romania Slovak Rep. Russia Job creation 1989 92 0.2 1.5 2.2 1.2 6.7 1.6 0.8 1992 94 1.5 4.5 1.0 6.1 n.a. 3.7 2.5 Job destruction 1989 92 25.0 10.2 19.1 14.9 11.2 15.2 3.8 1992 94 4.9 5.3 9.3 5.0 n.a 6.5 8.6 Source: Jackman (1998). Table I.33/34. Job Flows in Russia, Selected Transition and OECD Countries (Percent) Country (years) Job Job Job Net creation destruction reallocation employment growth OECD countries United States (1973 1988) 9.1 10.2 19.3 1.1 Canada (1979 1984) 10 10.0 20.6 0.6 France (1978 1984) 11 12.0 23.4 0.6 Germany (1988 1995) 4.6 4.1 8.7 0.4 United Kingdom (1987 1995) 5.4 5.4 10.8 0.1 Transition economies Poland (1994 1997) 3.0 3.7 6.7 0.6 Poland (1993, 1996, 1999) 8.4 9.1 17.5 0.7 Estonia (1993 1997) 9.3 8.8 18.1 0.6 Slovenia (1993 1997) 3.3 5.4 8.8 2.1 Bulgaria (1994) 1.4 5.2 6.6 3.7 Romania (1993 1997) 3.7 9.9 13.6 6.2 Hungary (1994) 1.3 6.6 7.9 5.3 Ukraine (1996) 2.5 15.3 18.0 12.0 Table I.34. Russia Faggio and Konings (1999) 1.2 4.9 6.1 3.7 Earle and Brown (2002a) (1985 1992) 0.9 3.9 4.8 3.16 (1992 1996) 2.1 11.2 13.3 9.1 (1996 2000) 3.5 8.7 12.2 5.2 Earle and Brown (2002b) (1990 1999) 2.4 9.2 11.6 6.8 Earle and Brown (2002c) (1985 1991) 1.4 4.5 5.9 3.2 (1991 1999) 2.4 10.3 12.7 8.0 Russian Economic Barometer (1996) 1.7 11.1 11.8 8.4 Sources: OECD (1997); Faggio and Konings (1999); Konings and Walsh (1999); Bilsen and Konings (1998); Davis et al. (1996); Brown and Earle (2002a, 2002b, 2002c). From different data sets; not fully comparable. Note: Job Reallocation = Job Creation + Job Destruction; Net Employment Growth = Job Creation Job Destruction. 172 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.34a. Job Creation, Job Destruction and Net Employment Growth by Ownership Type Job Job Net employment Ownership type creation destruction growth 1997/ 1999/ 1997/ 1999/ 1997/ 1998/ 1999/ 1996 1998 1996 1998 1996 1997 1998 State owned enter 17.3 21.7 37.3 26.7 19.9 15.4 5.0 prises Private owned 7.5 10.8 18.3 13.2 10.9 8.6 2.4 enterprises Foreign owned 122.6 33.4 105.8 44.4 50.8 58.5 11.0 enterprises Joint ventures 17.3 11.2 13.3 11.7 4.0 7.3 0.5 Source: Calculations from the Goskomstat Registry of Industrial Firms. ANNEX I 173 Table I.35. Growth of the New Private Sector (RLMS, 1994 2000)* 1994 1995 1996 1998 2000 Distribution of the employed by firm size Employment per firm £25 0.251 0.274 0.264 0.240 0.250 26 100 0.277 0.280 0.307 0.310 0.292 101 500 0.256 0.241 0.249 0.253 0.251 >500 0.216 0.205 0.180 0.197 0.207 Firm size is missing 0.198 0.308 0.302 0.267 0.234 Distribution of the employed by type of ownership State owned 0.754 0.683 0.663 0.647 0.605 Mixed 0.073 0.100 0.116 0.113 0.129 Domestic private 0.134 0.172 0.181 0.196 0.217 Foreign 0.040 0.045 0.039 0.044 0.049 Ownership is missing 0.181 0.155 0.148 0.140 0.127 Distribution of the employed by founding date Founded before 1980 ... 0.601 0.600 0.610 0.574 1980 1984 ... 0.062 0.060 0.045 0.042 1985 1989 ... 0.068 0.058 0.045 0.038 1990 1994 ... 0.243 0.218 0.163 0.151 after 1994 ... 0.026 0.064 0.137 0.194 Founding date is missing ... 0.510 0.488 0.446 0.422 N 4167 3781 3553 3374 3531 New private sector ... 0.229 0.249 0.300 0.327 Working individually at primary job ... 0.066 0.081 0.126 0.136 Employees at nonstate owned firms founded after 1989 ... 0.163 0.168 0.174 0.191 N ... 1758 1724 1877 2138 Employed in the private sector (Goskomstat) 22.8 23.5 27.5 28.3 *Goskomstat Annual Yearbook, 2000, p. 112. Notes: Sample includes respondents aged 15 72. Source: Calculations from RLMS. 174 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.36. Share of Public and Private Sector Employment in Transition Countries, 1996 (Percent) Sector Bulgaria Latvia Poland Romania Slovak Rep. Slovenia Private 50.3 54.0 59.5 53.7 45.8 47.4 Public 48.5 46.0 40.5 46.3 54.2 52.6 Source: OECD CEET database. Table I.37. Entry to the New Private Sector (Percent) N = 1631 Employed Employed in old sector in 2000 in new sector in 2000 Status in 1998 100.0 100.0 Employed in old sector 55.5 10.9 Employed in new sector 4.1 39.8 Employed but sector is missing 28.5 28.7 With job but not at work 5.3 3.1 Unemployed 2.1 8.2 Out of labor force 4.6 9.4 N = 1413 Employed Employed in old sector in 1998 in new sector in 1998 Status in 1996 100.0 100.0 Employed in old sector 51.3 14.4 Employed in new sector 3.9 30.4 Employed but sector is missing 32.0 30.4 With job but not at work 6.8 4.7 Unemployed 2.1 7.6 Out of labor force 3.9 11.5 Note: The new private sector includes the primary activity self employed and employees of firms with no state ownership that were founded after 1989. Source: Calculations from RLMS. ANNEX I 175 Table I.38. Characteristics of Employment in the New Private Sector 1995 1996 1998 2000 Female 0.189 0.220 0.261 0.275 Male 0.272 0.281 0.346 0.387 Age 15 24 0.409 0.436 0.527 0.486 25 34 0.358 0.393 0.422 0.439 35 44 0.178 0.232 0.294 0.333 45 54 0.153 0.142 0.193 0.213 55 72 0.118 0.077 0.095 0.125 Education Elementary 0.192 0.133 0.325 0.308 Secondary basic 0.261 0.292 0.349 0.355 Vocational 0.245 0.315 0.366 0.478 Secondary professional 0.190 0.242 0.275 0.275 University 0.245 0.227 0.248 0.276 Total 0.229 0.249 0.300 0.327 N 1758 1724 1877 2138 Note: Table shows the percentage of employed in the new private sector among all employed in a par ticular group. Source: Calculations from RLMS. I.39. Self Employment in the LFS, 1999 2000 Of which Non Members Employees Unpaid employees Employ Self of produc family ers employed tion coop workers eratives February 1999 0.917 0.083 0.008 0.053 0.020 0.001 May 1999 0.925 0.075 0.009 0.040 0.023 0.002 August 1999 0.930 0.070 0.009 0.043 0.016 0.002 November 1999 0.933 0.067 0.009 0.041 0.016 0.001 February 2000 0.937 0.063 0.008 0.038 0.016 0.001 May 2000 0.927 0.073 0.011 0.046 0.014 0.001 August 2000 0.927 0.073 0.010 0.043 0.018 0.002 November 2000 0.936 0.064 0.009 0.042 0.012 0.001 Note: Sample includes respondents aged 15 72. Source: LFS figures, reported in Goskomstat (2000c). 176 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table I.40. Self Employment in the RLMS, 1994 2000 Total population 1994 1995 1996 1998 2000 Share of the employed involved in any type of individual economic activity at any job 0.156 0.159 0.156 0.195 0.229 Of which: share of the employed reporting individual employment as their primary activity 0.085 0.102 0.098 0.143 0.171 Of which: reported not having a primary job but involved in individual economic activity 0.061 0.075 0.063 0.083 0.103 Worked individually at the primary job 0.023 0.027 0.035 0.060 0.068 Urban Population 1994 1995 1996 1998 2000 Share of the employed involved in any type of individual economic activity at any job 0.164 0.159 0.170 0.198 0.224 Of which: share of the employed reporting individual employment as their primary activity 0.088 0.101 0.100 0.141 0.161 Of which: reported not having a primary job but involved in individual economic activity 0.064 0.073 0.060 0.078 0.082 Worked individually at the primary job 0.025 0.028 0.040 0.063 0.079 Notes: Sample is restricted to the RLMS respondents aged 15 72 years old who worked at least one hour at any job in the previous month. The employed reporting individual employment as their pri mary activity consist of those reporting not having a primary job but involved in individual economic activity and those working individually at the primary job (not working at the enterprise or organiza tion with more than one employee). Source: Calculations from RLMS, reported in Earle and Sabirianova (2002b). ANNEX II WAGES Table II.1. Real Wages for Worker Groups, 1994 2000 1994 1995 1996 1998 2000 Monthly wage actually received last month 211.0 173.1 174.8 112.5 164.4 Imputed contractual wage 231.6 201.2 232.6 167.7 188.0 Notes: Sample is restricted to employees aged 15 72 with positive hours of work last month. Contrac tual wage is computed following Earle and Sabirianova (forthcoming) Source: Calculations from RLMS. Table II.2. Average Wage Level by Characteristics of Firms and Workers, 2000, Rubles Worker Mean St.Dev. Firm Mean St.dev. characteristics characteristics Total [N = 3803] 1740 2391 Rural 1142 1452 Female 1324 1326 Urban 1988 2646 Male 2207 3122 Sectors [N = 3767] Age Industry 2132 2079 15 24 1326 1393 Agriculture 762 948 25 34 1704 1806 Transportation/ 35 44 2005 3529 construction 2389 2073 45 54 1779 1809 Public services 1343 2973 55 72 1440 1445 Other services 2012 1842 Education Employment Elementary 1248 1469 per firm [N = 2886] 1728 1821 Secondary basic 1603 1741 <26 1539 1720 Vocational 1685 1617 26 100 1569 3677 Secondary professional 1591 1654 101 500 1678 1614 University 2320 3932 >500 2298 2235 Job to job Ownership [N = 3321] 1675 4771 mobility [N = 2666] State owned 1414 1481 Job stayers 1696 2713 Mixed 2240 2049 Job movers 2159 2199 Domestic private 2262 2207 Foreign 2499 1886 Note: Sample is restricted to all employees aged 15 72. Source: Calculations from RLMS. 178 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.3. Basic Wage Equations, 1992 2000 1992 1994 1996 1998 2000 1998 2000 (1) (2) (3) (4) (5) (6) (7) Women Log of actual monthly hours of work 0.381*** 0.163*** 0.241*** 0.355*** 0.349*** (16.78) (4.62) (5.62) (9.06) (8.32) Log of usual weekly hours of work 0.527*** 0.579*** (13.35) (13.07) Schooling (years) 0.038*** 0.074*** 0.056*** 0.077*** 0.076*** 0.085*** 0.090*** (7.79) (9.31) (7.00) (9.18) (8.36) (13.96) (13.37) EXP (years) 0.026*** 0.020*** 0.020*** 0.031***0.039***0.030***0.036*** (8,21) (3.84) (3.94) (5.98) (6.84) (7.50) (8.40) EXP2/100 0.052*** 0.037*** 0.046*** 0.064*** 0.085*** 0.063*** 0.072*** ( 8,16) ( 3.50) ( 4.56) ( 5.83) ( 6.82) ( 6.91) ( 7.20) Constant 3.160*** 9.875*** 10.858*** 3.167*** 3.816*** 2.260*** 2.463*** (23,87) (48.60) (44.81) (13.77) (15.52) (10.34) (9.99) N 3133 1968 1693 1664 1737 1915 1952 R2 0.303 0.267 0.331 0.307 0.294 0.419 0.420 Men Log of actual monthly hours of work 0.306*** 0.226*** 0.182*** 0.322*** 0.285*** (12.46) (6.12) (3.40) (6.47) (5.45) Log of usual weekly hours of work 0.424*** 0.203*** (7.97) (3.48) Schooling (years) 0.034*** 0.050*** 0.052*** 0.051*** 0.068*** 0.059*** 0.075*** (7,22) (6.49) (6.24) (6.00) (7.59) (9.71) (10.88) EXP (years) 0.030*** 0.020*** 0.012*** 0.023*** 0.032*** 0.021*** 0.026*** (8,46) (3.64) (1,97) (3.91) (5.49) (5.09) (6.03) EXP2/100 0.065*** 0.044*** 0.027** 0.052*** 0.063*** 0.049*** 0.061*** ( 9,57) ( 4.04) ( 2.28) ( 4.47) ( 5.39) ( 5.64) ( 6.52) Constant 3.941*** 10.315*** 11.645*** 4.172*** 4.687*** 3.645*** 5.206*** (27,67) (49.26) (39.22) (15.11) (15.74) (12.63) (16.36) N 3128 1993 1531 1466 1559 1631 1692 R2 0.374 0.342 0.344 0.339 0.377 0.472 0.461 Notes: t statistics are in parentheses; *** significant at the 1% level; ** significant at the 5% level; * sig nificant at the 10% level. Sample is restricted to employees aged 15 72. In columns (1) (5) EXP is meas ured as potential labor market experience (age minus schooling minus 6). In column (1) the dependent variable is log of after tax actual monthly wages received in the previous month. Sixteen regional dum mies are included. In columns (2) (5) the dependent variable is log of imputed contractual monthly wage. Contractual monthly wage is computed following methodology of Earle and Sabirianova (forth coming) In columns (6) (7) EXP is measured as actual labor market experience (data on actual labor market experience became available since 1998) and the dependent variable is log of usual monthly wage. Thirty eight regional dummies are included but not shown here. Source: Calculations from RLMS. ANNEX II 179 Table II.4. Extended Wage Equations with Tenure, Type of Ownership, and Founding Date, 1995 2000 Dependent Dependent variable is log variable is log of imputed contractual of usual monthly wage monthly wage 1995 1996 1998 2000 1998 2000 Log of actual monthly hours of work 0.153*** 0.180*** 0.319*** 0.284*** (5.47) (5.46) (10.43) (8.86) Log of usual monthly hours of work 0.432*** 0.368*** (13.91) (10.71) Individual characteristics Female 0.457*** 0.411*** 0.431*** 0.423*** 0.455*** 0.489*** ( 17.08) ( 15.29) ( 15.78) ( 14.95) ( 22.20) ( 22.14) Schooling (years) 0.047*** 0.050*** 0.064*** 0.072*** 0.071*** 0.080*** (8.15) (8.59) (10.48) (11.33) (16.55) (16.82) EXP (years) 0.035*** 0.011*** 0.021*** 0.029*** 0.021*** 0.026*** (8.66) (2.89) (5.18) (6.82) (6.95) (8.21) EXP2/100 0.075*** 0.032*** 0.049*** 0.063*** 0.049*** 0.059*** ( 9.44) ( 4.02) ( 5.92) ( 7.19) ( 7.56) ( 8.17) TENURE (years) 0.011** 0.017*** 0.021*** 0.018*** 0.015*** 0.014*** (2.35) (3.61) (4.42) (3.76) (4.37) (3.65) TENURE2/100 0.016 0.025* 0.038*** 0.027* 0.028*** 0.024** ( 1.11) ( 1.68) ( 2.69) ( 1.91) ( 2.58) ( 2.16) TENURE missing 0.058 0.181*** 0.055 0.045 0.015 0.033 (1.13) (3.58) (0.94) (0.70) (0.35) ( 0.69) Type of ownership (omitted: state owned) Domestic private 0.170*** 0.154*** 0.174*** 0.301*** 0.215*** 0.262*** (3.49) (3.38) (3.73) (6.65) (6.30) (7.56) Mixed 0.232*** 0.145*** 0.162*** 0.286*** 0.183*** 0.239*** (5.58) (3.59) (3.97) (6.88) (5.98) (7.46) Foreign 0.236*** 0.215*** 0.313*** 0.542*** 0.404*** 0.524*** (3.40) (2.88) (4.36) (7.83) (7.57) (9.98) Ownership is missing 0.008 0.090** 0.031 0.118*** 0.013 0.001 (0.20) ( 2.24) (0.76) (2.68) (0.42) (0.04) Founding date (omitted: old firms) New firms (founded after 1989) 0.196*** 0.285*** 0.177*** 0.205*** 0.150*** 0.126*** (4.22) (6.40) (4.10) (4.84) (4.64) (3.79) Founding date is missing 0.024 0.008 0.005 0.021 0.030 0.033 (0.81) ( 0.27) (0.16) (0.64) ( 1.30) ( 1.32) Constant 10.726*** 11.140*** 3.438*** 4.014*** 2.859*** 3.602*** (66.60) (59.98) (19.48) (21.45) (16.80) (19.07) N 3441 3224 3130 3296 3546 3644 R2 0.361 0.375 0.359 0.385 0.501 0.501 Source: Calculations from RLMS. Notes: t statistics are in parentheses; *** significant at the 1% level; ** significant at the 5% level; * significant at the 10% level. Sample is restricted to employees aged 15 72. Contractual monthly wage is computed following methodology of Earle and Sabirianova (forthcoming). In the first four columns EXP is measured as potential labor market experience (age minus schooling minus 6). In the last two columns EXP is measured as actual labor market experience (data on actual labor market experience became available since 1998). Thirty eight regional dummies are included but not shown here. 180 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY 1.11 1.08 1.11 1.21 1.22 1.06 0.85 0.93 1.30 0.95 1.24 1.17 1.00 1.11 1.17 1.76 0.94 1.20 1.42 0.80 State Nonstate/ 986.4 2160.6 2138.0 3511.0 4103.2 2241.9 3736.9 2077.6 1578.7 1656.8 1736.6 1356.9 2128.2 2307.2 2549.9 2696.8 1566.0 1137.3 2302.6 2660.5 Nonstate State 840.4 949.8 1949.3 1971.1 3165.6 3397.2 1831.8 3532.1 2451.0 1688.6 1270.3 1831.4 1097.7 2127.5 2077.1 2184.2 1532.6 1660.7 1626.7 3325.1 (Rubles) Monthly otalT 969.8 2094.4 2114.1 3441.9 4074.8 2236.5 3727.6 2130.9 1601.4 1625.9 1745.2 1348.6 2128.1 2260.0 2297.0 2164.1 1583.1 1072.3 1737.2 2923.5 1998 State 1.13 1.07 1.09 1.28 1.20 1.12 0.84 0.94 1.37 0.90 1.05 1.35 1.01 1.13 1.17 1.72 0.91 1.11 1.41 0.81 Nonstate/ 290). Ownership, p. (1999c, 8.9 7.9 16.0 16.3 24.6 31.2 15.9 28.9 15.8 12.6 12.6 12.9 10.2 15.3 16.8 18.9 18.8 10.6 15.8 18.3 Nonstate Hourly Goskomstat State/Nonstate 9.2 9.7 6.6 7.1 in State 14.1 15.2 22.5 24.4 13.2 25.8 18.7 13.4 14.4 15.1 14.9 16.2 10.9 11.7 11.2 22.7 by and reported, otalT 8.6 7.6 15.4 16.2 24.2 30.9 15.9 28.7 16.2 12.8 12.3 13.0 10.2 15.3 16.4 17.0 15.2 10.8 12.0 20.0 Survey Industry by Enterprise Costs Goskomstat catering Labor materials metals footwear and banking from paper II.5. metals and and and services Results ableT ood Economy Industry Electricity Fuels Ferrous Nonferrous Chemicals Machinery W Construction Glass extilesT Food Construction ransportT Communications radeT Restaurants Other Finance Source: ANNEX II 181 3.0 2.8 2.8 3.0 2.3 3.2 2.6 2.9 2.7 3.9 2.6 2.0 1.8 1.5 Other costs 3.3 1.5 4.2 2.4 5.9 3.0 3.9 2.8 1.4 1.0 0.5 0.3 0.4 0.2 costs Recreation 0.3 0.3 0.3 0.3 0.4 0.3 0.2 0.4 0.2 0.1 0.2 0.1 0.1 0.1 costs Training 288289). 28.3 30.2 28.5 30.7 32.4 31.5 25.1 29.9 30.5 29.1 28.2 28.8 28.1 30.2 Social pp. contributions (Percent) (1999c, 1998 4.6 2.0 6.2 3.1 3.6 2.5 costs 19.8 5.0 1.2 0.9 0.9 0.4 0.6 0.5 and Goskomstat Housing in 1995 in reported, wages Costs 60.5 63.2 58.0 60.5 55.4 59.5 48.4 59.0 64.0 65.0 67.6 68.4 69.0 67.5 Survey Cash Labor Enterprise of Goskomstat Structure from metals II.6. Results 1995 1998 1995 1998 1995 1998 1995 1998 1995 1998 1995 1998 1995 1998 ableT Economy Manufacturing Fuels Nonferrous Food radeT Services Source: 182 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.7. Incidence of Fringe Benefits by Firm Size, 2000 Fringe benefits Total Firm size <25 26 100 101 500 >500 Paid vacation 0.915 0.755 0.944 0.961 0.974 Paid sick leave 0.912 0.742 0.939 0.964 0.980 Health services 0.374 0.195 0.284 0.414 0.622 Vacation subsidies 0.438 0.190 0.375 0.506 0.680 Kindergartens 0.130 0.064 0.097 0.149 0.270 Catering 0.152 0.096 0.133 0.160 0.221 Transportation 0.142 0.083 0.094 0.215 0.194 Training 0.213 0.097 0.194 0.256 0.323 Loans 0.143 0.081 0.111 0.159 0.249 Note: The total sample size ranges from 3746 to 4102 respondents. Source: Calculations from 2000 RLMS Table II.8. Incidence and Magnitude of Wage Arrears in the RLMS, 1994 2000 Expected probability and magnitude of wage arrears 1994 1995 1996 1998 2000 Panel A: ARRDUM (dummy = 1 if worker has wage arrears) E(ARRDUM t) 0.405 0.419 0.599 0.637 0.293 (N=4716) (N=4389) (N=4166) (N=3928) (N=4151) E(ARRDUMt ½ ARRDUMt 1 = 1) 0.683 0.838 (N=1402) (N=1399) E(ARRDUMt ½ ARRDUMt 2 = 1) 0.788 0.796 0.392 (N=1213) (N=1652) (N=1798) E(ARRDUMt ½ i ARRDUMt i = 1) 0.683 0.887 0.882 0.513 (N=1402) (N=776) (N=525) (N=372) Panel B: ARRMOS (number of overdue monthly wages) E(ARRMOSt) 1.10 1.11 1.92 3.00 1.14 (N=4668) (N=4312) (N=4050) (N=3784) (N=4011) Unconditional distribution (ARRMOSt) ARRMOS < 1 month 0.603 0.594 0.415 0.379 0.731 = 1 month 0.149 0.156 0.149 0.122 0.111 2 3 months 0.164 0.170 0.250 0.219 0.085 4 6 months 0.055 0.054 0.134 0.162 0.032 > 6 months 0.029 0.026 0.053 0.119 0.041 E(ARRMOSt ½ ARRMOSt > 0) 2.75 2.73 3.27 4.82 4.24 (N=1861) (N=1760) (N=2381) (N=2358) (N=1078) Cotinued on next page ANNEX II 183 Table II.8. -- Continued 1994 1995 1996 1998 2000 E(ARRMOSt ½ ARRMOSt ). (N=3199) (N=3017) (N=2480) (N=2568) i where ARRMOSt < 1 month 0.49 1.07 1.16 0.31 I = 1 1.27 2.11 2.14 0.50 2­3 months 2.13 3.30 3.71 0.80 4­6 months 3.27 4.94 6.03 1.58 > 6 months 4.51 7.69 9.41 3.65 Notes: ARRDUMt = 1 if an employed respondent reports overdue wages on his/her primary job, 0 if no wages are overdue in year t. ARRMOSt = number of monthly wages reported overdue by an employed respondent in year t. Sample consists of all employed respondents in the respective files of the RLMS. Sample sizes are shown in parentheses for number of valid responses for ARRDUM and ARRMOS, respec tively; sample sizes vary primarily because of attrition and replacement in the RLMS panel, and second arily because of missing values for some respondents. Source: Calculations from RLMS. Table II.9. Wage Arrears: Accountants' Reports in a Firm Survey, 1991 98 Conditional mean Unconditional mean (firms with Years (all firms) wage arrears) Percentage of Amount of wage Amount of wage firms with wage arrears per arrears per arrears worker (rubles) worker (rubles) 1991 13.0 0 2 1992 14.6 3 17 1993 18.8 15 75 1994 25.0 1 913 7 041 1995 35.4 4 778 13 140 1996 46.9 5 302 11 098 1997 57.8 7 142 12 123 1998 56.3 9 321 16 532 Note: Sample is consistent across years (N = 192). A total of 66.5% of accountants (135 of 203 firms in the full sample) indicated that firms had wage arrears in 1991 98; 64.6% of accountants (124 of 192 firms in the consistent sample) indicated that firms had wage arrears in 1991 98. Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (1999). 184 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table II.10. Legal Penalties for Wage Arrears Ratio of penal Percentage Amount Number ties for wage of firms who of penalties of firms Years arrears had penalties for wage arrears with wage to the stock for wage arrears (in rubles) arrears of wage arrears 1991 0 0 0 25 1992 0 0 0 27 1993 0 0 0 34 1994 0 0 0 48 1995 0 0 0 64 1996 2.4 3997 0.003 85 1997 0.9 926 0.001 108 1998 2.8 6929 0.004 106 Note: Sample is restricted to firms with wage arrears in each year. Just 6% of firms with wage arrears (134 firms) ever had penalties for wage arrears. Source: Results from survey "Inside the Transforming Firm," reported in Biletsky et al. (1999). Table II.11. Incidence and Persistence of In Kind Substitutes for Wages in the RLMS, 1994 2000 Expected probability INKDUM (dummy) Sample of in kind substitutes 1994 1995 1996 1998 2000 Unconditional Mean Full 0.087 0.082 0.122 0.154 0.090 (INKDUMt) crosssection (N=4744)(N=4390)(N=4183) (N=3935)(N=4159) Mean (INKDUMt ½ Panel ... 0.369 0.395 ... ... INKDUMt = 1) for t, t 1 (N=306) (N=281) 1 Mean (INKDUMt ½ Panel ... 0.387 0.542 0.348 INKDUMt = 1) for t, t 1 (N=266) (N=330) (N=446) 2 Mean (INKDUMt ½ i INKDUMt = 1) Panel ... 0.369 0.565 0.846 0.679 i for t, t 1, t 2 (N=306) (N=92) (N=39) (N=28) Note: INKDUMt = 1 if an employed respondent reports in kind payments on his/her primary job, 0 if no wages are paid in kind in year t. Sample size is shown in parentheses for number of valid responses for INKDUM; sample sizes vary primarily because of attrition and replacement in the RLMS panel, and secondarily because of missing values for some respondents. Source: Calculations from RLMS. ANNEX II 185 Figure II.1. Incidence of High Pay and Low Pay in Transition Countries, 1997 &1992 ) 40 35 1997 (percent 1992 30 wage 25 median 20 >1.5 15 earning 10 5 Workers 0 FYR Yugoslavia Slovenia Poland Bulgaria Hungary Latvia Czech Romania Lithuania Moldova Belarus Russia Kyrgyz Georgia Armenia Macedonia Rep. Figure II.2. Incidence of Low Pay in Transition Countries, 1997&1992 40 (percent) 35 wage 30 25 median 1997 20 <2/3 1992 15 earning 10 5 Workers 0 Czech FYR Armenia Slovenia Bulgaria Poland Hungary Lithuania Latvia Romania Belarus Kyrgyzstan Georgia Moldova Russia Macedon Republ i i 186 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY UK otalT 0.0 0.9 12.9 13.7 11.9 10.8 50.2 49.8 100 variance Kingdom ned 8.1 4.6 3.0 United Explai variance 20.0 23.0 41.3 1999a); Russia otalT 2.7 1.5 6.5 1.0 7.5 al., 13.5 32.7 67.3 100 variance & ned Explai rianceaV 1.1 5.5 17.4 17.3 38.2 20.6 (Lehman Georgia Russia otalT 0.4 1.9 6.1 6.1 7.2 13.4 35.0 65.0 100 variance 1999b); ned 0.8 1.1 23.7 25.4 37.7 10.9 Bank, Explai variance orld Armenia (W Inequality otalT 0.3 0.4 7.8 8.4 3.6 12.4 32.8 67.2 100 variance Georgia ) ned 7.7 0.0 2.5 Earnings Explai variance 35.3 37.1 22.9 33.1 1999);, (1996 emtsov (Log) Macedonia otalT 8.9 9.4 5.8 2.0 0.0 8.4 0.6 25.7 74.3 100 (Y variance to Armenia ) ned 5.0 9.1 0.0 Explai 61.7 42.7 11.3 13.1 variance Factors 1999); Hungary (1996 otalT 2.1 3.8 0.0 4.8 5.5 25.9 17.9 42.1 57.9 100 variance Selected (Rutkowski, of ned 34.9 27.4 11.6 20.5 1.9 8.6 26.3 Explai variance Poland (1995) Macedonia otalT 9.1 3.8 6.8 11.5 0.6 8.7 2.9 33.1 66.9 100 and variance Contribution Hungary 1999b). al., II.12. Poland,: et tertiary explained experience ableT o/w Education Job Gender Ownership Sector Location otalT Unexplained otalT Sources: (Lehman ANNEX II 187 Table II.13. Distribution of Workers with Particular Characteristics by Wage Quintiles,2000 1998 wage quintiles Worker and firm character istics 1 (low) 2 3 4 5 (high) Female 0.245 0.279 0.193 0.172 0.111 Male 0.144 0.168 0.161 0.283 0.245 Age 15 24 0.341 0.220 0.136 0.220 0.083 25 34 0.245 0.221 0.165 0.218 0.151 35 44 0.176 0.227 0.181 0.213 0.204 45 54 0.177 0.213 0.184 0.247 0.179 55 72 0.175 0.298 0.211 0.186 0.131 Education Elementary 0.329 0.273 0.137 0.153 0.108 Secondary basic 0.269 0.216 0.152 0.217 0.147 Vocational 0.213 0.221 0.176 0.259 0.131 Secondary professional 0.195 0.249 0.212 0.203 0.142 University 0.076 0.207 0.189 0.251 0.278 Job to job mobility Job stayers 0.198 0.236 0.183 0.215 0.168 Job movers 0.213 0.189 0.155 0.259 0.184 Rural 0.378 0.273 0.139 0.139 0.070 Urban 0.129 0.212 0.195 0.254 0.211 Sectors Industry 0.124 0.177 0.220 0.261 0.218 Agriculture 0.532 0.258 0.116 0.082 0.013 Transportation/construction 0.075 0.179 0.155 0.309 0.283 Public services 0.242 0.284 0.173 0.182 0.119 Other services 0.141 0.220 0.188 0.243 0.208 Employment per firm <26 0.274 0.240 0.161 0.184 0.142 26 100 0.252 0.273 0.157 0.184 0.134 101 500 0.188 0.222 0.177 0.243 0.170 >500 0.103 0.152 0.213 0.273 0.260 Ownership State owned 0.237 0.250 0.177 0.207 0.129 Mixed 0.113 0.188 0.204 0.249 0.246 Domestic private 0.112 0.231 0.196 0.208 0.254 Foreign 0.087 0.111 0.175 0.381 0.246 Total [N = 2474] Notes: Sample is restricted to all employees aged 17 72. Characteristics of firms and workers are taken from 2000. The sum of shares does not add up to one because of missing values Source: Calculations from RLMS. 188 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Figure II.3. Cumulative Change in the Distribution of Real Wages, 1998 2000: RLMS Data stinuevitarap moC Pace of change Figure II.4. Growth in Wages by Quintile, 1998 2000 30,00 25,00 (percent) 20,00 wages 15,00 mean in 10,00 Change 5,00 0,00 1 2 3 4 5 Qunitles ANNEX II 189 Table II.14. Poverty Rates: Households With and Without Children Poverty Rates Round 8 of RLMS* (% of Heads of Households7) Non poor2 Poor3 Total Pensioner 66.1 33,9 100 Employed without wage 65.1 34.9 100 arrears Employed with wage arrears 48.6 51.4 100 status Unemployed6 and not receiving unemployment 39.1 60.9 100 benefits Unemployed6 and receiving 20.0 80.0 100 unemployment benefits Employment Not in the labor force 58.3 41.7 100 (not pensioner) of Not unemployed6 58.7 41.3 100 Not unemployed6 41.2 58.8 100 ment Unemployed6 Duration unemploy 36.1 63.9 100 for a year or more Total 56.0 44.0 100 Poverty rates (percent of house Round 8 of RLMS* hold heads7 with children8) Non poor2 Poor3 Total Pensioner 45.1 54.9 100 Employed without wage 59.6 40.4 100 arrears Employed with wage status 39.9 60.1 100 arrears Unemployed6 and not receiv 32.7 67.3 100 ing unemployment benefits Unemployed6 and receiving 28.6 71.4 100 unemployment benefits Employment Not in the labor force 60.0 40.0 100 (not pensioner) Not unemployed6 48.4 51.6 100 of Unemployed6 for less than 31.3 68.8 100 a year ment Unemployed6 for a year or Duration unemploy 33.6 66.4 100 more Total 45.8 54.2 100 Cotinued on next page 190 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Notes to the Table II.14: * Round 8 of the RLMS survey was conducted in Russia in October 1998 January 1999. 1 Adults those who 16 are years or older. 2 Poor households with total expenditures (see explanation in # 4) below official regionally differentiated (see explanation in # 5) subsistence minimum adjusted for economies of scale in the household (MLSD). 3 Non poor households with total expenditures (see explanation in # 4) above or equal to official regionally differentiated (see explanation in # 5) subsistence minimum adjusted for economies of scale in the household (MLSD). 4 Total expenditures total household monetary food and non food expenditures excluding big purchases, purchases of luxury goods, bonds/stocks, and savings plus value of home produced food evaluated at prevailing market prices. 5 Regionally differentiated subsistence minimum 8 regional poverty lines com puted as population weighted average across 78 official regional subsistence minima to match survey sample division of Russia into 8 regions 6 Unemployed those who do not report any work, receive neither pension nor disability benefit, and would like to work. 7 Household head was determined as follows: the oldest prime aged male (male aged 18 59), if there was no prime aged male in the household then the oldest prime aged female (female aged 18 54), if there was no prime aged female in the household then the oldest male aged 60 and over, if there was no male aged 60 and over then the oldest female aged 55 and over, if there no adults (18 or over) in the household, then the oldest person in the household was chosen as a head. 8 Children those below 16 years of age ANNEX III LABOR MARKET REGULATION Table III.1. Employment by Type of Labor Contract, 2000 (000s) f which f which Permanent Temporary job full time part time job full time part time Total 54 836 54 061 775 1 954 1 733 231 Male 27 849 27 577 272 1 204 1 086 118 Female 26 988 26 484 504 750 637 112 Source: Average LFS figures for February 2000 -- November 2000, reported in Goskomstat (2000c). Table III.2. Legal Arrangements for Fixed Term Contracts and Tempo rary Agency Work, Four OECD Countries Fixed term contracts Temporary agency work Germany · Widely possible without justifi · Generally approved except for cation construction · Maximum number of 4 con tracts/24 months (no limits in justified cases) Japan · < 1 year duration without · Restricted to specific occupa restriction tions · up to 3 years for particular types of workers Spain · Permitted for various reasons · Legal for justifiable cases (for example, specific projects; temporary replacements; train ing contracts; production eventualities; special categories of workers; long term unem ployed) United States · No restrictions · No restrictions Source: OECD (1999a) and country documents. 192 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table III.3. Legal Arrangements for Termination, Four OECD Countries Justifiable reasons Severance require Advance notice for eonomic dsmissal ments required Germany · Compelling business · No legal entitlement · Progressive increase or operational needs but often in collective based on years of serv agreement ice (from 2 weeks notice in trial period to 7 months for >20 years of service) · 1 month delay required after public notice for mass layoffs Japan · Rational restructuring · No legal entitlement · 30 days' notice reason or unavoidable but most large enter · Notification also to redundancy (court prises have voluntary Public Employment precedence, not law) plan Security Office in mass · Reasonable selection layoff (>30 workers) criteria Spain · Economic redundancy · 20 days' wages for · 30 days' notice each year of service · for mass layoffs, con (up to 12 years) sultation required for 30/15 days in firms with 50+/<50 employ ees United States · No restriction (except · No legal requirement · No regulation for indi in public sector) but voluntary or nego vidual dismissal tiated policies exist · 60 days' notice for mass layoffs Source: OECD (1999a) and country documents. ANNEX III 193 Table III.4. Ratio of Real Wage to Official Wage, Employee Reports by Sector, 1999 Real wage divided De novo State Privatized by official wage private Percent of employees Less than one 0.9 1.6 1.9 One (equal) 88.4 87.5 59.8 Up to two 9.1 7.4 13.8 Three 0.9 1.6 8.0 Four 0.4 0.8 2.7 Five 0.2 0.8 1.9 Six or more 0.2 0.2 10.2 No regular ratio 0.0 0.1 1.6 Total 100.0 100.0 100.0 N 558 771 884 Source: Tchetvernina (2000). Table III.5. Union Membership by Type of Enterprise, October 1997 De novo Budgetary State sector Privatized private Average entities firms Membership in trade unions 69 62 65 10 56 (percent of employees) Entities with trade unions 85 79 75 10 68 (percent of organizations) Source: Goskomstat survey in Kemerovo and Komi supplemented by questions designed by ISITO. 194 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table III.6. Union Membership Trends by Region, Mid 1980s to Mid 1990s Percentage point change in union density, mid 1980s to mid 1990s Decline of more Decline of up to Gain of up Gain of more than 10 points 10 points to 10 points than 10 points Africa Kenya ( 25) Egypt ( 4) Zimbabwe (+2) South Africa Mauritius ( 9) (+27) Uganda ( 4) Zambia ( 6) Latin America Argentina ( 29) Colombia ( 4) Chile (+4) Costa Rica ( 13) Dominican El Salvador (+2) Mexico ( 17) Rep. ( 2) Venezuela, Guatemala ( 4) R.B. de ( 13) Uruguay ( 8) Asia India ( 11) Bangladesh ( 8) Korea (+0.4) Pakistan ( 1) Philippines (+6) Thailand ( 0.1) Eastern Europe Azerbaijan( 33) Bulgaria ( 4) Turkey (+4) Malta (+17) and Central Asia CzechRep.( 34) Cyprus ( 9) Estonia( 46) Romania ( 10) Hungary( 20) Poland( 25) SlovakRep.( 15) Industrialized Australia ( 20) Canada ( 7) Hong Kong (+4) Israel ( 77) United States ( Belgium (+3) New 4) Finland (+10) Zealand ( 22) Japan ( 6) Spain (+7) Austria ( 13) Singapore ( 4) Sweden (+7) Greece ( 12) Denmark ( 3) Ireland ( 14) France ( 4) United Germany ( 10) Kingdom ( 13) Italy ( 4) Luxembourg ( 6) Netherlands ( 5) Norway ( 1) Switzerland ( 4) Total number [%]Developing 19 [32.8] 26 [44.8] 11 [19.0] 2 [3.4] countriesDevel 12 [35.3] 14 [41.2] 6 [17.6] 2 [5.9] oped countries 7 [29.2] 12 [50.0] 5 [20.8] 0 [0.0] Source: Betcherman, Luinstra, and Ogawa (2001), based on data from ILO (2000a). ANNEX III 195 Table III.7. Activities of the Federal Labor Inspectorate, 1994 98 1994 1995 1996 1997 1998 Inspections 35 051 117 457 211 178 267 895 263 977 Violations registered 391 227 1 414 732 2 915 509 2 237 524 2 098 350 Instructions ordered to 59 924 113 271 169 137 187 624 201 057 eliminate violations Inspections on wage 203 7 520 53 796 45 945 41 192 issues Wage arrears elimi nated as a result 41.2 302.8 8 325.3 7 733.4 10 275.5 of inspections (million rbl.) Workers who were unlawfully fired 74 1 579 2 930 3 770 2 814 returned to jobs Employers penalized for violation of labor 3612 12 554 32 260 34 029 32 963 legislation Employees of Labor 2 601 4 135 4 647 4 812 4 720 Inspectorate Source: Federal Labor Inspectorate, MLSD. 196 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table III.8. Means of Resolving Labor Conflicts, Evidence of Employers and Employees, 1999 Employers Employees Percent citing means of resolution for conflict Labor management committee 3.9 2.7 Trade union committee 4.9 7.0 Court 9.8 3.7 Dismissal 17.7 0.5 Agreement reached with 18.6 35.8 employer (informal) Conflict expired 22.5 4.8 Conflict not resolved 15.7 44.3 N 123 226 Source: Tchetvernina (2000). ANNEX IV SOCIAL SAFETY NETS Box AIV. 1. Different Systems of Unemployment Protection Aside from unemployment insurance programs, three other types of unemployment compensation programs are used worldwide means tested or flat unemployment benefits, severance pay (discussed in the previous chapter) and ISAs. All programs cover formal sector workers, but differ in their risk pooling arrangements, source of fund ing, and their eligibility and benefit conditions. Unemployment Insurance and Unemployment Assistance involve risk pooling at the economy level. Severance Pay provides firm level risk pooling, while ISAs provide risk pooling at only the individual level. Public works programs are also used to provide income support, and are available to both formal and informal sector workers. With respect to financing, unemployment assistance is mainly financed from general revenues; unemployment insurance is financed through individual, government, or employee payroll tax revenues; ISAs directly from own contributions of workers; and severance pay from own or employer contributions. Eligibility and benefit conditions as well as financing also differ across programs: Unemployment assistance programs generally provide means tested benefits to households with income below a particular threshold (as a primary benefit or once unemployment insurance benefits have been exhausted). However, some transition countries have unemployment assistance programs where the benefit is flat or is some proportion of average wage. Unemployment insurance systems and severance pay are defined benefit programs, where benefits are linked to past wages and years of service. ISAs provide benefits that are defined contribution, that is, benefits are based on investment returns on worker contributions; and as such are not predetermined. Most programs restrict benefits to laid off workers to avoid moral hazard problems, although some countries provide restrictive bene fits to new entrants and special provisions for unemployed nearing retirement. The incidence of these programs varies worldwide. In many countries, both developing and developed coun tries have some form of severance agreement. The existence of unemployment insurance depends a great deal on the level of income and region. Most OECD and transition countries (including Russia until 2001) have these sys tems; in East Asia, only China, Korea, and Japan have it. Unemployment assistance exists in many European coun tries and transition countries as a supplemental program for long term unemployed who have exhausted their benefits. It is a primary system of unemployment assistance in Australia. ISAs exist in some Latin American coun tries, where they have evolved from severance pay funds. Public works are prevalent in most countries, and along with severance pay, are the main protection systems in developing countries. Source: Betcherman (2000). 198 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY in to , Lux and 8%. latter. or state arise. the In rates very are con how to cases adminis 3 financed the Italy the that The from employees the rates few Spain, Japan, also of financing employees; contribute for of contributions a Employee: subsidies. covers Contribution of both the and Although and employers state contribute Iceland, ypically deficits States, only UI. countries, significantly contribution Most have range programs higher Italy state the countries. of fund, only States; to the UI where or are any costs. 3%. Source employers provides the vary have in contributions UI United both Most by employers cases and the equal There where employees (Employer: United embourg).T covers In state the Italy trative atypical, tributes rates between majority ever, below remainder rates of of Countries work. Ice to search of 4 is of require of keeping refusal non capable, period undergo or job and reporting for be however, 1 Finland, Disqualifica office to willing to Degree are offer, with number a benefits programs and Spain. all failure unjustified job in determines Regularly Groups between and recipient for by Conditions employment Almost the available, Exceptions land, tion training, suitable compliance requirements. offense disqualification; usually months. to required countries. Ice and at or 10 in 24 coun Resi in mis in months employ 6 last All office. Iceland quit, offer Range: weeks in conditions in denied Programs, stoppage, 52 year. days registration Portugal. minimum last work suitable countries. in Benefits voluntary (UI) requirement: past in 540 required of all to require of Eligibility employment the General ment in weeks land months tries the dency France. cases conduct, refusal almost of ben 36 period dura related recent length and age. 8 duration limit Belgium: most indefinite. is and/or the entitlement sometimes between entitlement of contributions, is UI also of Benefit countries entitlement. duration is length Most UI efit Maximum usually weeks. tion to period employment, , rates the per How tran UnemploymentInsurance maxi of Sweden and (range: for coun Bel Excep limit period: inde some children countries and of 75% In Additional Germany assistance (90%). voluntary or Kingdom, some aiting misconduct, additional benefit include France, earnings Most In recipients replacement and wages W of days. for extended or of earnings. benefits, 7 example, UI 40 side on provisions combination, is months). Belgium, Netherlands, United in cases spouses (for initial, Denmark Ireland, and in 6 average France). high or 3 Level average in with Kingdom dismissal period to benefits of on and Flatrate Kingdom). States, unemployment Canada, Features between ceilings or example, longterm benefits weeks into Generally vary recent tions (80%) ever, mumbenefit range. pendent offered United between countries, quit waiting 6 flatrate centage workers (for United tax gium, United Denmark, tries, sit all or in and, and work United UI. Countries UI.The cover irre based Stylized for industry Many or (volun special, excluded Finland, exclude casual Ireland, offer of Germany exist Spain, .1a. OECD coverage apprentices Few voluntary IV programs individuals type selfemployed, groups, Canada to government of of graduates. the generally conditions.Public and/or Austria, employees and example, Denmark, Portugal, countries provisions ableT other (for Sweden: Coverage Most majority employed spective sector. Luxembourg: extended training exclude whether occupation on sector Austria tary provincial employees). domestic ers Japan, .States). and ANNEX IV 199 ) Nine or for pro State Repub require UI employee excep (Slove Repub (Slove Employee. generally common. special financing state needed) is Only Slovak of UI 0.06% contributions rates 0.06% require (Slovak (Albania). finances Continued( countries contributions. entirely 1% 6% (when all finances 21) financing state groups. of Estonia, between and Sources Employer between and state Almost employer (out contributions. tion: financed contribution vary nia) lic). vary nia) subsidies deficit Latvia: special lic: grams. to if Countries recipi claim. of keeping the willing reduced, comply require training, for and are terminated not fraudulent benefits require or search, able does half be Benefits (job files Groups labormarket or to Conditions by About ent work. postponed, recipient with ments etc.) , 4 to in in Ser Roma be In few and for house below a eligible to from Russia) In workers Commonly employ months be not due offer. months must wage. Bulgaria, conditions employment 6 12 Registration offices Latvia, countries. Bulgaria Programs, ranges to in example, must dismissal all In if past require years. income. (Armenia, by (for Belarus, suitable (Bulgaria). year 2 level Ukraine Montenegro, Moldova), workers, of (UI) between minimum income Eligibility last last and and eligible years employment unemployment Minimum requirement months 4 countries ment the the at required Income nia, below bia hold stipulated countries Armenia, Georgia, not misconduct. Hungary if refusal is on age Slo and pro con maxi High In Croatia, age. the duration usually those years. entitlement and/or depending duration Slovenia, for weeks). 2, University countries Bulgaria, entitlement (26 varies employment, period, retirement countries, Russia, graduates Some of Benefit Countries) entitlement countries, Hungary shorter Republic). extensions early most months In mum 6 end: some duration length tribution (Azerbaijan, Poland, vak training have periods. vide near for and can cause flat the rates and wage pro wage] the of UnemploymentInsurance regional, on (for Estonia, or Uzbek level (Transition 50 by double level to unem min. yrgyzK supple present). wage) of or (usually local, average entrants graduated benefit min. new, benefit Asia limited countries or Croatia, the of wage, be Benefits benefits benefits Ukraine, percentage or (ceiling replacement between the dependent addition Azerbaijan, benefits in can dependent a range avg. be Some wage vary or (2001). wage). Albania, Earningsrelated ypicallyT Russia, wage Level (usually of flatrate income (usually loss. labormarket Level Albania, usually provide Features Central time. dependent national min. flatrate benefits Raju job Initial generally 75%. floors ceilings or the sometimes of vide ¦(minimum instead earningsrelated example, Georgia). rate over ployed receive wage. Republic, istan: ments; minimum each and and 16 or cover for casual Milan, Stylized Europe (citizen personnel eligible. required). usually 1654 Romania: and .1a. University East age: programs and workers military excluded. IV UI. by of Croatia, residency for graduates domestic are odopivec,V: men or ableT for Coverage Majority employed ship Coverage 59 women. discharged eligible training Usually workers Source 200 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY ) 0.75 2%; pay of of of (state 0.5%; sales 12.5%) UI finances taxes, cost. 0.7%; 0.1% contribute. contribute funds. both (rate provisions); finances China, between state total Republic business allocations financing vary (employees, contributions a govt. state: employers: percentage of employers, including 0.61% employers 3%; aiwan,T employers: and insurance countries employees employerfinanced earmarked local Islamic rates Continued( Korea: 5/7 and Ecuador 1%), higher 15%; employers: Uruguay: through on 0.5%. 0.2%; administration, social or security of wages, Sources In deficits).Chile: various for 0.65%. employers: South 2%. costBrazil: subsidies. cost other equal (N=4). social mainly employers: of Contribution and employers Except employers, an roll (employees, for finances total through but tax Bangladesh, China, dependent state: Iran, deficit. employees: employees: state: employee from and will Countries benefits Iran, and enezuela,V able China: of be of and keeping must aiwan,T available, for Republic Chile, and work. recipients capable, Groups to Islamic Korea, be work. de, Argentina, to by Conditions In R.B. willing China, South must ing if of to also Brazil: stipu registra Brazil, months; Islamic refusal employed ineligible Argentina: means months; 6 South In. or 6 Registration been some employment. unemploy required. require offices. recipients benefits. requirement: years. in be Iran: Korea: conditions other 2 Programs, have of unemployment offer. recent Chile applicants lack household. China: office involuntary misconduct of misconduct. South or China: must, months and cannot be Iran, for for must of suitable 12 self year; (UI) Eligibility employment employment Republic be to period Uruguay: socialsecurity 1 aiwan,T at 6 aiwan,T must accept employment Generally for lated Argentina tion Chile, dismissal applicants other claimant support Insured Islamic China: and Korea, ment Republic cannot to at sta days, length on entitle employ aiwan,T on Repub entitlement 12 on and (benefits based marital to special 3 Brazil: entitlement 90240 based in cases). Islamic and duration claimant months, maximum (range: dependent Brazil: increased. Korea: of employment special months, length. length Benefit year Argentina, are age in 650 1 South period length. on 316 Iran: countries, years. previous of All period months). ment ment cases/circumstances, durations 2 based of extended China: employment lic employment tus. of 4 (60 + of per to of benefits 20% UnemploymentInsurance de earnings. length below 90 Korea up varies flatrate Uruguay: benefit aiting Brazil replace benefit W wages on 50% average service rate maximum: of functions reemploy claimant rate of 10% A if South Uruguay: duration days. benefit and Brazil, days), initial average basic flat lumpsum based Korea: 14 of 60% (3 wage, Bolivariana of of (minimum: Iran: the Caribbean graduated dependent day). maximum wage. days). period: offered and China: China: half supplement. 55% 50% per South per is Level and (30 is benefits (permanent/casual) wage. wage" the 50 Chile: earnings Ecuador: Argentina, proportional on aiting on Barbados minimum W W are Republica aiwan,T rate sum daily bonus before minimum nature of Features aiwan,China,T and Incomereplacement between earnings. benefits. (based length). minimum limits the dependent period: days), enezuelaV Iran, ment Bangladesh: lump and employment. minimum supplement dependents. "basic cent 30,000 ment leaves period. and to . the the and per only in some less Chile, Korea: America and, outside sector Republic under Stylized Republica (excludes insured coverage excludes and with de Exceptions: significantly China:. excludes Brazil, workers), workers South workers public aiwan,T selfemployed). Islamic Iran: covered firms .1a. Uruguay Latin provide enezuelaV of Bangladesh: China: casual the Asia differs voluntarily industry workers. commerce), Korea, and IV China, those enterprises, Barbados, de contract Bolivariana and and and Mexico, (excludes (excludes enterprises. and and South Republic and aiwan,T provisions. workers. 5 ableT employed Iran, firms. Coverage Argentina, Ecuador, Republica enezuela.MostV all Uruguay industry Bolivariana domestics Barbados employees Bangladesh, of China.Coverage Islamic selfemployed, persons, other commerce manent publicsector collective all selfemployed than ANNEX IV 201 ) 2%, Africa, 1%. employ South 1.5%; cost. employees: deficit. employers: total Continued( financing 1%; of Egypt, employees: state: finances 2.5%. Sources Algeria, ers: state: employees: unisia,T Countries willing Africa: and of keeping South for and benefits available, Groups unisia,T able, be Conditions work. by Egypt, must to 3 South to par employ 6 weeks. that quit job or current sources Africa). prior 52 requires Unemploy be Egypt: quarters; last require other applicants covered also search suitable Africa) (South Conditions Programs, must 12 in of no voluntary misconduct of unisiaT unisia:T for South strike Algeria active years weeks have dependents. in 3 unisia:T refusal 13 and of unisia),T (UI) Eligibility employer contributions. have cannot (Egypt, income. Algeria: ment; with months; Africa: Algeria applicants of months application. must ment (Egypt, (Egypt), offer ticipation on 16 entitle based (1236 contribu between months. on 3 duration varies maximum weeks. varies based unisia:T 26 employment Egypt: Benefit duration of weeks duration Africa: 28 length. Algeria: length months). ment and tion South a and Alge with UnemploymentInsurance indus 45%. gradu South of of allowances sectors. earnings benefit Africa, wage earnings latter; days period: 7 of the spousal South average of aiting Egypt: Level minimum commerce of minimum 75% 60%; and and of benefits; Features mean Egypt: unisia:T trial ria: national floor ated provided.W Africa eco pub agri and Africa: Egypt, Coverage Algeria: Stylized from highly of excludes casual South and excludes unisia.T .1a. Africa countries. workers (>76,752 Republic and Egypt: IV workers. unisia:T employees, domestics Arab workers. Africa, between salaried sector. employees ableT Coverage Algeria, South differs laidoff nomic licsector agricultural excludes paid Rand/year). cultural 202 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY through Financing of revenues. tax Sources Governmentfinanced general Countries the of seeking Keeping unem available (periodic for require throughout and fully conditions actively receipt Benefits be of Groups to and capable satisfied conducted). programs be by Conditions work; period Many claimant ployed, for work.Eligibility must the checks 5 (5 or test; or as Gen his in pre or is 8 years In ineli of Nether coun then UI to assets such year France 10 only Ger Austria). limited the meanstest contribu for dispute, offer, and benefits). of the in some and Programs, conditions in In job contribution employment applicants unemployment); employment labor extended satisfy A example, example assistance A) irrespective or of conditions U is available unemployment to income housing exceptions: months unemployment). for (for (for if reduced (U must, state years (6 employment UI months). only due, suitable are and unemployment); of period offered (4 preceding of countries, A Eligibility Some. employment for 6, U Australia: ypicallyT (household excludes family erally employment tory lands years Portugal tory ceding years preceding some contribution applicable gible many tries: exhaustees In voluntary refusal benefits waiting weeks. if or (1 5 and the 30 depend condi age for days; Portugal, as Exceptions with on In duration exhaustee months; (150 long Netherlands (6 UI as those a duration basis). fulfilled. is the for depends Benefit are Spain Sweden (longer Indefinite, tions include year); months ents); dayweek duration claimant not latter). Ire ypiT mari rate spe, peri UnemploymentAssistance some meet on flat number have and/or older not younger minimum to special of or are above Usually the to waiting period example, days). families. 5 benefit rates for unemployed depends levels, reduced no, (for of of their countries applicants Benefits are exist granted UI level presence Sometimes, waiting to meanstested, uniform and eliminated. Sweden, needs Some also Level at and income from Generally benefits days; Features basic benefit, time. applied 3 status provisions However, dependents. Generally income the individuals cally tal of over threshold which completely cial unemployed. provisions persons. ods. times transiting land, UI all for occu mem and Stylized countries irrespec of or, selfstand half .1b. Australia vailableA OECD (complement have workers, industry IV about Only in systems. sector, Zealand A ableT countries U of Coverage Present ber systems). New ing unemployed tive pation. ANNEX IV 203 ) through through through financing of Continued( revenues. revenues. revenues. tax tax tax Sources Governmentfinanced general Governmentfinanced general Governmentfinanced general bene Countries willing claimant or in condi of the and keeping throughout results for training require of receipt. Eligibility benefits available, offer satisfied AU job be of Refusal Groups Conditions programs capable, must period by be work. cancellation. Most to to acceptable fit tions the sat and Must. Regu assets) work. employ must, to unem (and income. of history contributions as work. of of of dependents, Programs, conditions willing unemployed. labor/employment Generally involuntarily income source as and be have A) the registered capable quarters irrespective contribution to 12 (U and must other Eligibility or required. fund, no registered visits household Capable the Provided ment be lar office isfy test. unisia:T orker to ployed, W unemployed, have or 12 3 eligi Enti Excep grad if as (18 for 3 days), Russia is unem has workers. of long recent fulfilled. (180 limited entitlement so months). as Romania Bulgaria, entitlement social extended income duration are possible), unemployed; (6 by if retirement, Estonia the such or level. years), for dependents Poland, sometimes example, duration near Benefit countries, indefinite, (2 renewal for Slovenia In is decided is months. Countries) is groups 3 conditions include (for children, poverty some months Estonia, In length bility tions Hungary months, (6 months ployed), tlement certain uates months). duration In claimant more below unisia:T to and level not cer any mini a in UnemploymentAssistance number size, receipt fully (for pres regis Hun be com (Transition assistance does of of benefit, the, are income above can benefits and unisia:T and status, benefit However, or days benefits Asia of household Usually full Republic, status 30 minimum assistance level. amount, AU benefits flatrate industry Generally marital after in Level on social either from Czech kind. marital meanstested in dependents. rates. children. benefit Poland: on unemployment. wage Features Central of stipulated or of flat dependents, other Meanstested at depends of age of affect earnings, tain deducted example, gary). cash Meanstested, based ence Mauritius: households tered mum merce. and sys and Russia, nona covered Stylized Estonia, coun Special all Europe system most AU graduates Security officers. Romania, in unisia: workers .1b. East Republic, Slovenia.Dual an Ua workers. Social recent Asia Africa IV Coverage has China and for military salaried Czech Poland,, Republic, UI salaried Mauritius.T of Kong, National ableT Estonia .All Bulgaria, Hungary Slovak tems tries. only provisions discharged Hong unisia,T gricultural under Fund. 204 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY a , of con first old and percent contri work admin further system, half in the a 8 insurance individual necessary social/pri the covers allowances, one if sickness months; employees social percent an to dual a that fnancing 6 , earnings: 3 to to is Peru: employers a of contribute family of Colombia: by payroll death, America each goes less (this goes system , rate; Uruguay of paid percent government, benefits, Sources salary In pesos fee, Employers deficits Ecuador, are 15 balance, Latin ISAs. new the percent the insurance disability unemployment). Brazil, contribution monthly butions ers' tribute 7.47 and istrative account. 12.5 and finances vate age, maternity and as rea pro Brazil, Programs, the educa of Some reasons and other Exception: conditions for health Accounts (regardless dismissed. is access example, Eligibility separation). worker Savings separation allow of if (for expenditures). Upon son only grams well tion and (plus Insurance interest individual employer enezuela,V previous plus the percent Panama of de in 40 In benefit . dismissed, of (deposits if multiple additional penalty as Level Brazil, Bolivariana accumulated an as set account Unemployment pay of Amount savings earned).In must interest) Republica penalty wages. in de empoT Features de Ecuador, workers. established Bolivariana Garantia Stylized de FGTS, Colombia, Coverage Republica formalsector .1c. All IV (Fundo Chile, Peru, Servicio ableT Brazil In do 1967), Panama, enezuela,V ANNEX IV 205 not f if f MLS MLS (MLS) of f of maximum, and standard percent percent percent Countries percent living wage) 140% 140% 150180 150180 150180 300 (minimum if MLS levels minimum minimum mini MLS of wage wage of of of of i European before) employed benefit 1990 percent before percent percent percent not percent minimum minimum c 70 in 70 before if of of since 70 70 but but wage but (but employed Eastern percent percent percent employed not percent percent percent Unemployment expressed 100 90 85 none, employed none, mum before none, not none if 60 60 80 and b case 10 gross in months months months months months months then 5 6 retraining 6 retraining 6 retraining 6 12 monthly wage of of of months Central months months months 6 months last next 6 6 6 case case case percent course) 6 in individual's month, earnings of for d first following to d first following in first following in first following in (60 first following following first less minimum for to percent percent percent percent percent percent percent percent percent percent percent percent rate rate percent percent Relation months retraining percent percent percent 100 wage percent Equal 60 60 60 50 (70 course) 60 50 (70 course) 60 50 (70 course) 50 40 6 of Flat Flat 70 60 45 systems of con indi g a on basis) duration benefit benefits months months months months months months months months months months months months Max. 6 12 12 12 12 6 6 6 6 6 (3 extensions sidered vidual 24 min. record days days months Required employment months months months months months months months 6 6 6 9 12 12 12 180 180 18 unemployment of e e e e period Reference months months months months months months years years years years bold) 12 12 12 12 3 3 3 12 12 3 features in Date Main 1989 1991 1992 1998 1991 1992 1996 1998 1991 1995 1989 .1d. legislation IV R. ableT (latest Bulgaria Czech Estonia Hungary 206 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY I II wage phase phase minimum f minimum during maximum, during of of t p and pension wage percent fiorints Countries fiorints percent percent percent percent percent percent percen percent percent wage) 200 18000 1500 180 oldage 140 average None None 200 210 None 150180 150 150 320 300 (minimum old levels minimum wage of European benefit minimum percent minimum i i of i in of p Eastern fiorints percent percent percent pension percent percent percent percent percentl percent Unemployment expressed 100 none 8600 90 age 70 100 None None 7580 7692 None 45 None None 80 100 and 60 to gross I IIh wage at entitle period entitle period months average months months months months 3 months 9 of of entrants) 6 6 6 9 3 months Central half half 9 months phase phase reduced months months paid 9 months retraining months retraining months months months 3 half half new percent 3 9 for 6 6 3 3 3 in individual's minimum national for earnings 50 first following to first second first second during during h of for later, first following after of first following during first following during first following first following first following amount and o o period period zl percent percent percent percent percent percent percent percent percent percent percent percent rate percent percent percent percent percent percent percent percent percent percent percent percent Relation percent percent percent percent percent percent systems 70 ment 50 70 ment 50 70 50 65 90 (70 70 percent 70 60 45 36 wage flat 378,2 5060 5060 65 60 70 65 60 70 60 50 60 50 70 60 months 70 60 months of excep k in duration cases benefit benefits days days months months months months months months months months months years months months months Max. 360 360 6 6 None 12 2 tional 18 9 9 12 6 12 12 24 24 min. j record days days months months days days months months months Required employment year months year year 360 90 90 None 180 1 6 1 1 12 12 12 912 912 unemployment of (Continued) period j Reference months months months years years years year year year years years years bold) 4 4 4 None 1 18 1 1 3 3 3 18 18 features in Date Main 1991 1992 1993 1997 1993 1993 1989 199294 1997 1996 1998 1991 1992 1995 1997 1996 1998 .1d. legislation IV R. ableT (latest Latvia Lithuania Poland Romania Slovak Slovenia ANNEX IV 207 of level the and tested, means is program This months. 18 wage um) age w min (maxim gross the minimum of the additional and UI an percent for full 80 wage to to average course. entitled PROGRAM member the months. ANCE remain W training six wage. family a earnings. of between per of ALLO month period average per previous a of income difference to SUPPORT for completion wage a the of equal for receive paid percent average is upon set 60 the is minimum and at qualify initially percent raise the paid 20 awarded to is of benefit they leavers course. years. plus half the months) income. two 9 months benefit assistance. to school member wage percent training than then 15 a (after and pension in more wage family wage. remaining extended minimum not each then additional course. enrolled was exhaustees unemployment minimum for (2000). unemployment the if it minimum unemployed minimum of to an work of the Raju percent, training when than benefit level time and equal wage; a percent casual than 70 months in less 1990. the first at supplement 180 1990 from of to a lower guaranteed were if three months' paid the orgotter, IV.Id insurance only of W 1990. six enrolled January if receives earning September percent receive wage unemployment ableT last until earnings in 60 months can of the is applies the additional October of to an required recipient year monthly three percentage odopivec,V a Plus Unemployment Since verageA previous Not Net Notes a b c d e Thef One Unemployed g h Ifi Introducedj Some k allowance Minimuml First Recipients As m n o p Source 208 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.2. Spending1 on Passive and Active Labor Market Policies in Selected OECD and EU Accession Countries2,3 Passive policies Active policies Unemp. rate Spending Spending Percent Percent per per of GDP of GDP unempl. unempl. OECD Austria (1999) 3.7 1.22 0.32 0.52 0.14 Belgium (1998) 9.5 2.51 0.26 1.34 0.14 Denmark (1999) 5.2 3.12 0.60 1.77 0.34 Finland (1999) 10.3 2.33 0.23 1.22 0.12 France (1999) 11.3 1.85 0.16 1.33 0.12 Germany (1999) 8.7 2.12 0.24 1.30 0.15 Greece (1997) 9.8 0.50 0.05 0.35 0.04 Ireland (1996) 11.7 2.42 0.21 1.66 0.14 Italy (1999) 11.4 0.64 0.06 1.10 0.10 Netherlands (1999) 3.3 2.81 0.85 1.80 0.55 Norway (1999) 3.3 0.47 0.14 0.82 0.25 Portugal (1996/98) 5.2 0.83 0.16 0.87 0.12 Spain (1999) 15.9 1.41 0.09 0.81 0.05 Sweden (1999) 7.2 1.70 0.24 1.84 0.26 Switzerland (1997/98) 4.2 1.03 0.25 0.41 0.10 United Kingdom (1997/98) 7.0 0.82 0.12 0.37 0.05 Canada (1998/99) 8.3 0.99 0.12 0.51 0.06 United States (1998/99) 4.5 0.25 0.06 0.18 0.04 Japan (1998/99) 4.1 0.52 0.13 0.09 0.02 Australia (1998/99) 8.0 1.06 0.13 0.52 0.07 New Zealand (1998/99) 7.4 1.57 0.21 0.62 0.08 EU average4 1.73 0.26 1.16 0.16 OECD average5 1.43 0.23 0.92 0.14 CEEC Czech Republic (1999) 8.8 0.31 0.04 0.19 0.02 Estonia (1998) 9.9 0.08 0.01 0.08 0.01 Hungary (1997) 8.7 0.56 0.06 0.40 0.04 Poland (1996) 14.3 1.71 0.12 0.49 0.03 Slovak Republic (1996) 11.1 0.54 0.05 0.56 0.05 Slovenia (1998) 7.9 0.89 0.11 0.83 0.11 CEEC average 0.68 0.06 0.42 0.04 Sources Riboud et al. (2001) Data for 1996, 1997 and 1999. in Riboud, Sanchez Paramo, and Silva Jauregui (2001). 1/ Data from different years (in parentheses). 2/ Spending Measure 1: Ratio of GDP spending on UI to unemployment rate (both in percentage terms). 3/ Spending Measure 2: Spending per unemployed individual as a percentage of GDP per labor force participant. 4/ Does not include Luxemburg. 5/ Average for all OECD countries in the table ANNEX IV 209 rate (Labor Force Survey) 0.1% 0.1% 0.3% 0.4% 0.0% 0.1% 0.2% 0.0% 0.1% 0.0% 0.4% 0.0% 0.5% 0.1% 0.4% 0.4% 0.4% 0.0% 0.0% 0.5% 0.0% 0.2% 0.0% 0.1% 0.0% 0.1% Unemploy ment aver wage rate age 16.73% 18.79% 13.86% 18.10% 13.13% 14.66% 7.88% 18.84% 19.43% 8.56% 14.43% 22.96% 20.76% 16.02% 25.72% 28.50% 22.46% 7.74% 20.17% 18.62% 20.97% 15.19% Replacement Percent fund 9.8% 0.5% 5.7% 2.7% 0.0% 0.8% 0.4% 0.5% 0.0% 0.0% 0.3% Payment arrears 12.3% 27.6% 36.2% 22.7% 12.0% 36.9% 47.7% 8.2% 40.2% 38.4% 37.2% 47.3% 15.6% 13.9% 8.2% employment total expenditures arrears 18.8% 25.9% 16.3% 11.4% 71.0% 0.6% 1.0% 50.6% 3.9% 0.2% 7.4% 7.3% 20.1% 9.7% 0.7% 8.0% 8.9% 39.1% 13.1% 9.7% 0.0% 23.8% 21.4% 25.8% 36.5% 9.7% Contribution Percent expenses 0.04% 0.02% 0.01% 0.04% 0.03% 0.02% 0.02% 0.06% 0.03% 0.04% 0.02% 0.02% 0.01% 0.04% 0.04% 0.04% 0.01% 0.09% 0.06% 0.09% 0.03% 0.10% Administra tive 1999 GDP of ALMPs 0.05% 0.09% 0.05% 0.03% 0.06% 0.06% 0.05% 0.05% 0.04% 0.05% 0.09% 0.09% 0.08% 0.15% 0.05% 0.09% 0.08% 0.06% 0.08% 0.07% 0.05% 0.08% Percent Parameters, benefits Unemploy 0.16% 0.39% 0.24% 0.20% 0.33% 0.42% 0.20% 0.15% 0.27% 0.21% 0.36% 0.34% 0.40% 0.46% 0.14% 0.22% 0.30% 0.08% 0.20% 0.14% 0.18% 0.06% ment Program area (oblast) Unemployment region .3. region autonomous Karelia Komi IV Republic of of estern region region region region region region region region region region region region region region Nenets region region ableT Region Federal North Arkhangelsk Incl. ologradV Murmansk Republic Republic NorthW St.Peterburg Leningrad Novgorod Pskov Central Bryansk Vladimir Ivanovo verT Kaluga Kostroma Moscow Moscow Orel Ryazan Smolensk Central 210 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY rate (Labor Force Survey) 0.1% 0.1% 0.3% 0.4% 0.0% 0.1% 0.2% 0.0% 0.1% 0.0% 0.4% 0.0% 0.5% 0.1% 0.4% 0.4% 0.4% 0.0% 0.0% 0.5% 0.0% 0.2% 0.0% 0.1% 0.0% 0.1% 0.0% 0.0% Unemploy ment aver wage rate age 16.73% 18.79% 13.86% 18.10% 13.13% 14.66% 7.88% 18.84% 19.43% 8.56% 14.43% 22.96% 20.76% 16.02% 25.72% 28.50% 22.46% 7.74% 20.17% 18.62% 20.97% 15.19% 25.84% 19.47% Replacement Percent 0.5% 5.7% 2.7% 0.0% 0.8% 0.4% 0.5% 0.0% 0.0% 0.3% Payment arrears 9.8% 12.3% 27.6% 36.2% 22.7% 12.0% 36.9% 47.7% 8.2% 40.2% 38.4% 37.2% 47.3% 15.6% 13.9% 8.2% 1.1% 2.1% employment total expenditures fund arrears Percent 18.8% 25.9% 16.3% 11.4% 71.0% 0.6% 1.0% 50.6% 3.9% 0.2% 7.4% 7.3% 20.1% 9.7% 0.7% 8.0% 8.9% 39.1% 13.1% 9.7% 0.0% 23.8% 21.4% 25.8% 36.5% 9.7% 19.1% 29.5% Contribution expenses 0.04% 0.02% 0.01% 0.04% 0.03% 0.02% 0.02% 0.06% 0.03% 0.04% 0.02% 0.02% 0.01% 0.04% 0.04% 0.04% 0.01% 0.09% 0.06% 0.09% 0.03% 0.10% 0.02% 0.01% Administra tive GDP of ALMPs 0.05% 0.09% 0.05% 0.03% 0.06% 0.06% 0.05% 0.05% 0.04% 0.05% 0.09% 0.09% 0.08% 0.15% 0.05% 0.09% 0.08% 0.06% 0.08% 0.07% 0.05% 0.08% 0.05% 0.06% Percent benefits Unemploy 0.16% 0.39% 0.24% 0.20% 0.33% 0.42% 0.20% 0.15% 0.27% 0.21% 0.36% 0.34% 0.40% 0.46% 0.14% 0.22% 0.30% 0.08% 0.20% 0.14% 0.18% 0.06% 0.26% 0.21% ment area (oblast) region region autonomous Karelia Komi region Republic of of estern region region region region region region region region region region region region region region Nenets region region region Region Federal North Arkhangelsk Incl. ologradV Murmansk Republic Republic NorthW St.Peterburg Leningrad Novgorod Pskov Central Bryansk Vladimir Ivanovo verT Kaluga Kostroma Moscow Moscow Orel Ryazan Smolensk Central lauT aroslavlY ANNEX IV 211 0.1% 0.0% 0.2% 0.1% 0.2% 0.1% 0.1% 0.0% 0.1% 0.0% 0.0% 0.4% 0.1% 0.1% 0.0% 0.0% 0.3% 0.0% 0.0% 0.4% 0.1% 0.0% 0.0% 0.0% 0.1% 0.0% 0.2% 0.2% 0.0% 0.5% 25.36% 16.14% 0.00% 23.12% 20.83% 34.81% 16.85% 22.66% 23.19% 18.42% 20.11% 28.48% 13.60% 13.03% 17.34% 19.48% 23.15% 0.00% 24.72% 16.83% 21.80% 26.37% 17.90% 25.25% 16.05% 8.43% 32.22% 10.0% 0.5% 21.7% 12.5% 21.8% 12.1% 8.5% 0.4% 6.4% 0.0% 0.1% 37.0% 6.0% 8.9% 0.1% 2.3% 26.3% 1.3% 2.9% 42.5% 6.2% 0.5% 0.6% 0.3% 9.6% 0.0% 23.5% 16.2% 2.5% 45.0% 33.2% 44.6% 15.0% 27.0% 22.7% 36.5% 27.9% 12.3% 39.4% 40.7% 30.8% 18.9% 14.2% 0.0% 7.9% 9.7% 17.5% 16.0% 62.9% 15.3% 21.0% 33.5% 0.4% 24.8% 26.3% 10.0% 1.5% 24.3% 86.8% 1.8% 0.03% 0.04% 0.05% 0.01% 0.03% 0.12% 0.05% 0.02% 0.05% 0.02% 0.01% 0.06% 0.02% 0.08% 0.03% 0.03% 0.03% 0.00% 0.03% 0.05% 0.02% 0.01% 0.08% 0.08% 0.01% 0.12% 0.34% 0.05% 0.08% 0.09% 0.07% 0.07% 0.04% 0.05% 0.03% 0.05% 0.09% 0.09% 0.05% 0.02% 0.07% 0.06% 0.05% 0.23% 0.00% 0.04% 0.09% 0.08% 0.12% 0.05% 0.12% 0.06% 0.06% 0.15% 0.22% 0.31% 0.27% 0.28% 0.31% 0.14% 0.19% 0.17% 0.09% 0.30% 0.38% 0.13% 0.27% 0.33% 0.19% 0.17% 0.77% 0.00% 0.12% 0.23% 0.11% 0.17% 0.12% 0.58% 0.21% 0.17% 0.58% region region region El OssetiaAlania region region Mariy Mordovia Chuvash region Kalmykia atarstanT region region Caucasus Adygeya territory KarachaevCircassian Dagestan KabardianBalkar North Ingushetia Chechen of of of region region region territory Novgorod region region of of of of of of of of of region region region region region olgaVyatkaV Nizhny Kirov Republic Republic Republic centralChernozyem Belgorod oronezhV Kursk Lipetsk ambovT Povolzhye Astrakhan olgogradV Samara Penza Saratov Ulyanovsk Republic Republic Northern Krasnodar Republic Stavropol Republic RostovonDon Republic Republic Republic Republic Republic 212 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY rate (Labor Force Survey) 0.3% 0.0% 0.0% 0.1% 0.0% 0.0% 0.3% 0.2% 0.0% 0.1% 0.0% 0.2% 0.5% 0.3% 0.1% 0.0% 0.2% 0.2% 0.2% 0.1% 0.0% 0.0% 0.2% 0.6% 0.0% Unemploy ment aver wage rate age 15.37% 24.32% 22.28% 17.83% 22.07% 20.15% 0.00% 10.08% 11.74% 28.35% 22.79% 25.94% 20.77% 11.59% 11.39% 16.89% 0.00% 20.72% 31.62% 38.93% 0.00% 15.68% 56.26% 46.59% 35.79% 28.77% Replacement Percent 6.4% 0.0% 0.1% 0.0% 2.0% 0.6% 1.1% 5.8% 3.7% 5.5% 1.2% 3.3% 0.0% Payment arrears 27.0% 13.6% 26.3% 12.8% 19.6% 10.6% 12.8% 21.9% 46.7% 34.8% 11.3% 16.3% 19.2% 20.0% 20.5% 55.4% employment total expenditures fund arrears 20.8% 18.3% 17.8% 28.0% 7.9% 21.3% 28.0% 5.3% 27.5% 28.1% 31.1% 30.4% 27.5% 15.2% 12.6% 74.3% 42.1% 5.5% 69.7% 14.1% 14.4% 20.2% 6.5% 9.0% 2.6% 99.6% 36.9% 8.7% 23.3% Percent Contribution expenses 0.06% 0.04% 0.04% 0.01% 0.03% 0.06% 0.00% 0.06% 0.93% 0.00% 0.07% 0.03% 0.04% 0.02% 0.05% 0.04% 0.06% 0.03% 0.07% 0.04% 0.38% 0.07% 0.05% 0.02% 0.12% 0.04% Administra tive GDP of ALMPs 0.09% 0.03% 0.04% 0.19% 0.04% 0.05% 0.00% 0.08% 1.10% 0.01% 0.07% 0.06% 0.09% 0.06% 0.08% 0.02% 0.04% 0.05% 0.07% 0.41% 0.57% 0.05% 0.11% 0.07% 0.34% 0.07% Percent benefits Unemploy 0.32% 0.05% 0.14% 0.56% 0.22% 0.19% 0.00% 0.35% 4.15% 0.03% 0.24% 0.19% 0.16% 0.24% 0.27% 0.16% 0.16% 0.23% 0.29% 1.49% 0.42% 0.21% 0.15% 0.25% 0.42% 0.27% ment area area area area area autonomous autonomous region area area region autonomous region autonomous Buryat autonomous region region territory Buryat region region Bashkortostan Udmurtia Altai Khakassia Buryatia of of Siberian region of region Siberian region Siberian of of region region region region region autonomous autonomous region KomiPermyatsky territory UstOrdyn Aginski estern estern Region Ural Kurgan Orenburg Perm incl. Sverdlovsk Chelyabinsk Republic Republic W Republic Altai Kemerovo W Novosibirsk Omsk omskT yumenT KhantyMansi maloNenetzaY Eastern Krasnoyarsk Republic imyraT Evenki Irkutsk Incl. Chita Incl. Republic ANNEX IV 213 0.1% 0.3% 0.2% 0.0% 0.1% 0.2% 0.0% 0.1% 0.2% 0.1% 0.0% 0.2% Rubles 2000 n/a 0.19% 0.54% 0.00% 0.00% 0.27% 4.76% 0.07% mln Rubles 13.79% 25.26% 18.70% 39.08% 28.63% 22.83% 16.28% 22.78% 13.28% 13.53% 29.75% 30.62% 1999 0.32% 0.21% 0.56% 0.00% 0.00% 0.31% 4.71% 0.08% mln 11.4% 15.2% 26.0% 20.1% 1.5% 12.6% 16.8% 0.0% 15.2% 10.7% 21.1% 15.0% 1.0% 15.6% Rubles 1998 0.33% 0.25% 0.76% 0.00% 0.01% 0.32% 7.02% 0.20% mln 15.2% 29.6% 15.2% 22.4% 17.5% 65.9% 29.7% 83.2% 29.6% 0.0% 84.4% 15.7% 42.7% 4.9% 19982000, 0.04% 0.03% 0.04% 0.03% 0.21% 0.02% 0.00% 0.03% 0.02% 0.02% 0.12% 0.08% GDP of 0.19% 0.09% 0.07% 0.16% 0.09% 0.12% 0.16% 0.11% 0.16% 0.09% 0.04% 0.06% state Percentage Fund for as planned as leave, budget budgets) sick, federal 0.30% 0.31% 0.29% 0.32% 0.20% 0.40% 0.39% 0.35% 0.39% 0.32% 0.14% 0.32% Employment expenditures the regional etc.), by expenditures IT SIF Expenditures and maternity etc. from to payment included) state investment (2000). not Social (federal expenses, benefit, financed including, capital pension area SIF oblast SIF + area service by compensated total pensions akutia) Goskomstat program by Semipalatinsk (Y allowances uvaT and Selected programs region benefits territory administrative funeral region autonomous .4. benefits benefits military of territory Sakha (planned Mayak, budget region region IV child financed autonomous region autonomous region of region MLSD: protection East Koryak East federal Republic Far rimorskyP Khabarovsk ewishJ Amur Kamchatka Incl. Far Magadan Chukchi Sakhalin Republic Kaliningrad Sources ableT Social Employment Monthly (excluding Benefits birthgrants, Chernobyl Chernobyl Recreational Pensions pensions Chrnobyl, by 214 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.5. Unemployment Benefit Replacement Rate, Select CEE Countries, 1992 99 1991 1992 1993 1994 1995 1996 1997 1998 1999 Bulgaria 0.68 0.29 0.34 0.31 0.30 0.38 0.30 ­ ­ Czech Republic 0.42 0.30 0.28 0.27 0.25 0.24 0.24 0.20 0.20 Estonia ­ ­ 0.17 0.10 0.08 0.07 0.07 0.09 0.09 Poland ­ ­ ­ ­ 0.31 0.31 0.30 0.27 0.25 Slovenia 0.32 0.29 0.33 0.34 0.33 0.31 0.36 0.37 0.37 Slovak Republic 0.49 0.32 0.30 0.27 0.25 0.23 0.30 ­ ­ Source: Vodopivec, M., A. Worgotter, and D. Raju (2000). Table IV.6. Survey Unemployment Rate, CEE Countries 1991 1992 1993 1994 1995 1996 1997 1998 Bulgaria ­ ­ 21.4 20.5 14.7 13.7 15.0 16.0 Czech Republic ­ ­ 3.9 3.8 4.1 3.9 4.8 6.5 Estonia 1.5 3.7 6.5 7.6 9.7 10.0 9.7 9.6 Hungary ­ 9.3 11.9 10.7 10.2 9.9 8.7 7.8 Latvia ­ ­ ­ ­ 18.9 18.3 14.4 13.8 Lithuania ­ ­ ­ 17.4 17.1 16.4 14.1 13.5 Poland ­ 13.7 14.9 16.5 15.2 14.3 11.5 10.6 Romania ­ ­ ­ 8.2 8.0 6.7 6.0 6.3 Slovak Republic ­ ­ 122 13.7 13.1 11.1 11.6 11.9 Slovenia 7.3 8.3 9.1 9.0 7.4 7.3 7.4 7.9 Source: Vodopivec, M., A. Worgotter and D. Raju (2000). Table IV.7. Registered Unemployment Rate 1991 1992 1993 1994 1995 1996 1997 1998 Bulgaria 11.1 15.3 16.4 12.8 11.1 12.5 13.7 12 Czech Republic 4.1 2.6 3.5 3.2 2.9 3.5 5.2 7.5 Estonia n.a. n.a. 5.0 5.1 5.0 5.5 4.6 2.0 Hungary 7.4 12.3 12.1 10.4 10.4 10.5 10.4 9.1 Latvia 0.6 3.9 8.7 16.7 18.1 19.4 14.8 13.8 Lithuania 0.3 1.3 4.4 3.8 6.2 7 5.9 6.4 Poland 11.8 13.6 16.4 16.0 14.9 13.2 10.5 10.4 Romania 3.0 8.2 10.4 10.9 9.5 6.6 8.8 10.3 Slovakia n.a. n.a. 12.2 13.7 13.1 11.1 11.6 11.9 Slovenia 8.2 11.5 14.4 14.4 13.9 13.9 14.4 14.5 Source: Vodopivec, M., A. Worgotter and D. Raju (2000). ANNEX IV 215 Figure IV.1. Replacement Rate of Unemployment Insurance Payments, Transition Economies, Early and Late 1990s* (Percent) * Average replacement rate in the first six months of benefit eligibility. For Estonia, the benefit is flat, so the rate is calculated as the level of the benefit divided by the average wage. Source: Vodopivec, M., A. Worgotter, and D. Raju (2000). Figure IV.2. Maximum Potential Duration of Unemployment Insurance Payments, Transition Economies, Early and Late 1990s (in Months) Source: Vodopivec, M., A. Worgotter and D. Raju (2000). 216 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.8. Determinants of Regional Registration Rates SUMMARY OUTPUT Registered Unemployment Rate Regression Multiple 0.831 R Square 0.691 Adjusted R Square 0.615 Standard Error 0.009 Observations 77 ANOVA df SS MS Regression 15 0.01002 0.00067 Residual 61 0.00447 0.00007 Total 76 0.01450 Coefficients Standard Error t Stat Intercept 0.00225 0.00562 0.40 Unemployment 0.00248 0.00029 8.43 GDP/capita 0.00000 0.00000 1.34 UIBenefit/Average Income 0.03874 0.01510 2.57 Unemployed per Employee 0.00005 0.0001 4.02 Mono 0.00002 0.00021 0.08 North 0.01142 0.00446 2.56 NW 0.00119 0.00499 0.24 V/B 0.00857 0.00456 1.88 C C 0.00026 0.00462 0.06 Pvolz 0.00041 0.00395 0.10 N. Cauc 0.01195 0.00531 2.25 Ural 0.00426 0.00462 0.92 W. Sib 0.00095 0.00441 0.22 E. Sib 0.01102 0.00529 2.08 Far E. 0.00046 0.00421 0.11 ANNEX IV 217 Table IV. 9. Poverty Rate and Poverty Gap for Households with Chil dren* in Russia (calculations are based on the data from Round 8 of RLMS**) Average number Poverty gap1 Type of household with Number of of children* per Poverty rate1, (for households children* households yousehold percent in group), per cent Including child 260 1.26 55.00 20.83 (may well) allowance child other as Excluding child Household 260 1.26 60.38 26.75 receives receive allowance allowance benefits Including unem 2 ployment bene 7 1.43 85.71 26.76 other bene unem fits head fits Excluding Household than unemployment 7 1.43 100.00 41.07 member receives ployment benefits Including head's UI unemployment 5 1.60 60.00 22.16 receives benefit benefit receive too) head may Excluding head's benefits unemployment 5 1.60 60.00 37.43 benefit Household unemployment (other's Including unem ben ployment bene 12 1.50 75.00 24.84 fits receives household efits Excluding unemployment 12 1.50 83.33 39.84 Any member benefits unemployment Including other 485 1.19 55.88 23.37 benefits other (may list well) 5 benefits the as Excluding other Household 485 1.19 67.42 36.33 receives from benefits benefits receive above * Children are those below 16 years of age. ** Round 8 of RLMS survey was conducted in Russia in October 1998 January 1999. 1 Poor -- households with total expenditures below official regionally differentiated subsistence min imum adjusted for economies of scale in the household (MLSD). Non poor -- households with total expenditures above or equal to official regionally differentiated subsistence minimum adjusted for economies of scale in the household (MLSD). 2 Household head was determined as follows: the oldest prime aged male (male aged 18 59), if there was no prime aged male in the household then the oldest prime aged female (female aged 18 54), if there was no prime aged female in the household then the oldest male aged 60 and over, if there was no male aged 60 and over then the oldest female aged 55 and over, if there no adults (18 or over) in the household, then the the oldest person in the household was chosen as a head. 218 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY 3 Total expenditures -- total household monetary food and nonfood expenditures excluding big purchases, purchases of luxury goods, bonds/stocks and savings plus value of home produced food evaluated at prevailing market prices. 4 Regionally differentiated subsistence minimum -- 8 regional poverty lines computed as pop ulation weighted average across 78 official regional subsistence minima to match survey sample divi sion of Russia into 8 regions. 5 Other benefits included different types of pensions and subsidies and benefits from apartment renting as well as subsidies for fuel. Table IV.10. Poverty Impact of Unemployment Programs in Select Transition Economies, Mid 1990s Bulgaria Estonia Hungary Latvia Poland Slovak Slovenia Rep. Poverty reduction165 1.1 0.5 14.8 2.2 16.7 2.7 6.8 Coverage166 3.8 3.8 7.5 2.5 5.6 0.6 11.5 Targeting167 17.4 31.1 4.9 12.4 6.8 0.5 16.0 Adequacy168 13.0 15.2 25.4 29.8 34.1 7.3 21.2 Source: Vodopivec and Raju (2001). 165 Change in poverty headcount brought about by unemployment benefit receipt, in percent. Poor are defined as individuals with consumption less than 50% of median. 166 The share of poor who were unemployment benefit recipients, in percent. 167 The share of unemployment benefit received by the poor in percent. 168 Average share of unemployment benefit in total household income of recipients, in percent. ANNEX IV 219 6.91 0.14 33.29 12.15 48.90 Richest are quintiles 2 bonds/stocks 4th 8.00 0.83 41.10 15.03 58.90 goods, (calculations luxury expenditure 3d 9.53 0.97 of 37.29 12.71 56.08 Russia purchases in percapita 2nd 9.10 0.83 By 38.76 13.79 56.83 size. purchases, Transfer big household of 5.66 0.97 33.84 14.78 52.62 Poorest the ypeT by excluding by divided Rural 6.39 0.52 41.36 12.25 53.72 1999. prices expenditures Transfers Urban January market 5.66 0.80 34.50 13.84 52.63 nonfood settlement of 1 1998 and prevailing food Receiving type at October By 23.40 0.99 39.41 16.26 68.23 in RLMS*) of Metropolies monetary evaluated Russia 8 in food Petersburg. Households otalT 14.03 7.84 0.75 36.86 13.69 54.67 St. household of Round and conducted total -- from was from Moscow homeproduced subsidies of survey data Percentage benefits value benefits included the children RLMS expenditures .11. renting, plus Transfer and of IV on for pension 8 of pensions fuel savings, ableT based ypeT Subsidies apartment for Benefits Unemployment Oldage Other otalT Round* Metropolies Percapita 1 2 and 220 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY 3.26 3.52 17.21* 15.75 11.06 13.97 Richest quintiles 2 bonds/stocks Russia 4th 4.79 7.18 8.75* 28.19 21.29 27.10 in goods, luxury expenditure 3d 8.07 8.10 of 19.32* 41.92 29.47 36.25 Transfers purchases percapita 2nd 6.67 By 13.39 22.13* 51.35 36.70 47.28 size. Receiving purchases, big household 15.64 22.05 42.81* 95.50 62.84 81.51 Poorest the by excluding Households for divided Rural 13.58 11.57 24.16* 32.16 23.77 32.02 1999. RLMS*) expenditures prices of 8 Urban January market 4.96 8.27 17.93 33.36 22.60 29.54 nonfood settlement Consumption of 1 1998 and Round prevailing food otalT type at October in from By 5.02 3.84 14.60* 26.65 20.85 21.62 in Metropolies monetary evaluated Data Russia in food Petersburg. Transfer the otalT 5.33 7.06 18.45 32.05 22.55 28.68 St. household of on and conducted total -- Based 10. was from < Moscow homeproduced subsidies of are cell) survey Percentage benefits benefits the value included 12.. children of RLMS expenditures renting, plus transfer and of IV for pension (size 8 of pensions n fuel savings, ableT (Calculations ypeT Metropolies Percapita Subsidies apartment for Benefits Unemployment Oldage Other otalT Round Note: * 1 2 and ANNEX IV 221 6.88 7.80 0.62 80.51 27.05 Richest 100.00 quintiles 2 bonds/stocks 4th 4.66 7.74 0.89 (Calculations 98.22 32.87 100.00 goods, luxury Russia expenditure 3d 5.80 7.20 1.51 of in 86.64 24.40 100.00 2 purchases Percapita 2nd 2.99 8.15 1.04 By 86.42 23.92 100.00 size. purchases, Expenditures big household 4.52 4.47 1.40 66.06 23.55 Poorest 100.00 the by excluding PerCapita divided Rural 2.08 5.92 0.88 74.03 17.09 100.00 and 1999. prices expenditures Urban January market 5.56 4.94 0.83 settlement 66.11 22.55 100.00 nonfood of 1 1998 and Urban/Rural prevailing by RLMS*) ypeT food at October of By 7.82 8.97 1.07 59.31 22.83 100.00 in 8 Metropolies monetary evaluated Russia in Transfers food Round Petersburg. of otalT 4.98 5.69 0.88 67.24 21.21 100.00 St. household and from conducted total -- was Data from Moscow homeproduced subsidies of survey the benefits Distributions value on benefits included children RLMS expenditures .13. renting, plus transfer and of IV for pension 8 of pensions Based fuel savings, ableT are ypeT Subsidies apartment for Benefits Unemployment Oldage Other otalT Round* Metropolies Percapita 1 2 and 222 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY the on bonds/stocks Based goods, are luxury of purchases (Calculations taloT 100.00 100.00 100.00 100.00 100.00 100.00 size. purchases, Russia big household in 1 27.66 22.05 11.43 19.27 20.52 20.04 Richest the by excluding divided 4th 22.05 21.90 16.35 23.51 24.94 23.58 quintiles 1999. prices Expenditures expenditures 1 January 3d market 23.80 20.38 27.62 20.73 18.51 20.46 nonfood 1998 and PerCapita expenditure prevailing food at by October 2nd 11.89 23.04 19.03 20.68 18.15 19.83 in monetary evaluated percapita Russia in Transfers By food of 14.60 12.63 25.58 15.81 17.87 16.09 Poorest household RLMS*) conducted total of -- was 8 from homeproduced of survey benefits Distributions Round value benefits fuel children RLMS expenditures .14. renting, plus transfer and for of IV for pension 8 from of pensions savings, ableT Data ypeT Subsidies apartment subsidies Benefits Unemployment Oldage Other otalT Round* Percapita 1 and ANNEX IV 223 65.85 83.32 Richest 567.00 288.71 222.07 279.2 quintiles Computed 2 bonds/stocks 4th 42.87 64.90 81.12 253.51 188.16 244.24 goods, Russia in luxury expenditure 3d of 47.41 47.96 109.60 248.35 172.25 214.67 purchases Quintiles Percapita 2 2nd By 27.06 52.41 85.82 199.00 145.13 184.30 size. purchases, big household 5.00 42.00 84.59 Expenditure Poorest 192.41 124.24 163.62 the by excluding divided Rural 75.87 61.37 127.50 223.57 130.40 208.09 1999. PerCapita RLMS*) prices expenditures of and 8 Urban January market Transfer settlement 39.09 66.33 85.43 236.67 176.20 216.86 nonfood of of 1 1998 and Round ypeT prevailing ypeT food at October Urban/Rural from By 44.57 38.96 137.33 238.84 193.01 205.67 in by Metropolies monetary evaluated Data Russia in Particular food Petersburg. a the otalT 42.40 56.65 100.22 233.08 166.29 212.98 St. household on Transfers and conducted total -- Receive was Based from Moscow homeproduced subsidies of Who are survey benefits PerCapita value benefits included children RLMS expenditures .15. renting, Those plus transfer and of IV for pension 8 of pensions fuel savings, ableT Among (Calculations ypeT Subsidies apartment for Benefits Unemployment Oldage Other otalT Round* Metropolies Percapita 1 2 and 224 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY . thus neutral privi society some or concen (Peru). increases in income (in labormarket pay workers, rich already to on the of countries) redistribution. inequalities redistribution participants severance Effects progressive on evidence. among formalsector contributing advantage Mildly developed effects Little Program trated By dualism, the leged increasing per devel well who in: higher work fairly generous. is (Peru). consump Programs transition mildly transition less pay the in: similar . for smoothing unemployed Adequacy claimants most benefits employed of In benefits reduction poverty those severance are economies, evidence.Consumption level of otherwise who Consumption oped tion preserved. countries, Poverty economies, reduce Little head received than ers avail de to UnemploymentBenefit wide agricul coun or legal formal of groups. of workers formalsector of always access of spite available not in economies, developing subset Effects Coverage not a (selfemployed, household to segments workers, disadvantaged and mostly to provided by developed In coverage tural excluded).In tries able workers. vailableA sector facto entitlement.Hinders jobs Distributive of Summary pay .16. IV ableT Unemployment insurance/assistance Severance ANNEX IV 225 on intro its small design. concen (Colom of be by obtained rich to redistributive the effects likely results form, States). participants are progressive. eliminated among pure United its In effects Program trated bia). Redistributive duction (simulation the Strongly reduc poverty evidence. on effects Inconclusive Strong tion. 1 (par digits; all formal develop they below often to of double kept some jobs, in subset a available reach economies, reality to rates typically In workers. been countries principle, transition vailableA sector In ticipation ing in have percent). rationed. (2001). Raju insurance and accounts works odopivecV: Unemployment savings Public Source 226 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Table IV.17. Distribution of Benefits and Beneficiaries of Unemploy ment Support Programs, Mid 1990s* 2nd 2nd Poorest Richest poorest Middle richest quintile quintile quintile quintile Unemployment insurance** Average 15.4 22.3 22.5 20.0 18.9 Brazil 10.6 24.6 19.1 25.1 13.6 Bulgaria 17.8 14.9 32 13 22,4 Estonia 31.1 17.7 19.6 18 13.6 Hungary 7.8 20.4 28.2 24.6 19.1 Latvia 15.7 13.8 18 26 26.5 Poland 14.8 24.1 22.9 21.6 16.6 Slovak Rep. 3.1 33.2 20.8 18.8 24.1 Slovenia 22.5 30 19 13.1 15.4 Unemployment insurance savings accounts Colombia 0.0 4.3 n/a 19.1 76.6 Severance pay Peru 4.7 9.5 28.6 33.3 23.8 Public works Argentina 78.6 15.3 3.5 2.1 0.4 Training Mexico 69.9 15.5 8.1 5.0 1.5 Notes: *Share of benefits received by individual quintile, for transition economies, and share of benefici aries in population group, for Latin American countries. **Unemployment insurance benefits include both payments of unemployment insurance and unem ployment assistance. Source: Vodopivec and Raju (2001) IV 227 stabi per GDP pre page unem growth down 15 the automatic to milder next and as output growth insurance. reduces 10 about under on but UI during on on as by Theoretical acting Output By macroeconomic lizer, losses turns cent. dictions effects inconclusive.The effects insignificant. Similar, effects ployment Continued of vs. UI under insur jobs evidence precari enter as Brazil, to regular increase into In effects informal entry jobs. Encouragement taking Inconclusive on ous payments probability selfemployment. Similar unemployment ance. for of supply of workers. members family Programs supply labor job. members spouses disincentive a family other Labor the take Reduces of unemployed Strong other to not strong job unem restruc enter overall of increases; increase layoffs of milder and States hindered, on insurance. equilibrium Because is under adjustment (Blanchard but as adjustment turing Enhancing prises United temporary Attractiveness restructuring in evidence of (partial results). creation overall assisted 1997). Similar, effects ployment UnemploymentBenefit of of in labor reduc miss prima rate.For unem force increases adjust and (Euro positive but as milder outcomes insurance. increase [[word equilibrium labor up under inactivity down shocks but Effects persistence as groups on in to unemployment persistent unemployment). unemployment show slow Equilibrium market and benefit unemployment.Bene A increases ing?]]the unemployment some effect participation, tions rily in fits ment make more pean Similar, effects ployment evi unem unem Efficiency effort prob improve earners. of wages disincen matching disincen the particularly leaving job leaving hazard on postunemploy ment for of wages). postunemploy for lowwage Jobsearch and Significant tives ployment (moral lem).Inconclusive dence ment (via ment Significant tives ployment, for Summary .18. IV ableT Unemployment insurance Unemployment assistance 228 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY unem growth growth milder on and under insurance. but researched. as effects well Output Similar, effects ployment The not ) of vs. under insur jobs as regular effects Continued( informal evidence. Encouragement taking Similar unemployment ance. No for of members Family Programs Supply job. Members disincentive a family Other Labor take evidence. Strong other to No into job unem on is Restruc Enter Overall so milder of "sclerosis" and under insurance. inflows effects but but as Adjustment reallocation turing Enhancing prises Similar, effects ployment Negative labor economy's increased: unemployment reduced, creation. UnemploymentBenefit of of Labor unem particu workers. milder Outcomes under insurance. unemploy but reduces participation Effects Persistence as young on inconclusive. of Unemployment selfemployment. Equilibrium market and Similar, effects ployment Strongly employment, larly Increases in Effects ment prob ages unem (Large Efficiency Effort from W earners. the of disincen particularly hazard incentives leaving jobsearch costs PostUnem unemploy over separation.) but for increased. of lowwage moral with (2001). JobSearch and ployment enter Significant tives ployment, for No lem effort, to ment litigation disputes cause Raju Summary and .18. pay IV odopivecV: ableT Unemployment assistance Severance Source ANNEX IV 229 Table IV. 19. Expenditures on ALMPs as a Percentage of Total ALMP Expenditures, 1995/96 Micro Employ Total as Public Job sub Country Training enter ment percent works sidies prises services of GDP Australia 33.7 26.5 3.6 7.2 28.9 0.84 Belgium 35.7 40.7 0.0 7.9 15.7 1.41 Canada 48.2 5.4 7.1 3.6 35.7 0.56 Denmark 77.0 12.8 3.5 1.3 5.3 2.26 France 55.8 17.1 3.1 12.4 11.6 1.30 Germany 55.2 21.0 2.1 4.9 16.8 1.43 Ireland 32.0 38.3 1.1 14.3 14.3 1.75 Netherlands 54.7 9.5 0.0 9.5 26.3 1.37 Sweden 59.1 19.1 3.1 7.6 11.1 2.25 UK 53.2 2.1 2.1 0.0 42.6 0.46 USA 57.9 5.3 0.0 0.0 36.8 0.19 Czech Rep. 14.3 7.1 0.0 7.1 71.4 0.14 Hungary 30.2 25.6 0.0 14.0 30.2 0.43 Poland 40.6 21.9 6.3 25.0 6.3 0.32 Source: OECD (1997). REFERENCES Aidt. T., and Z. Tzannatos. 2000. Collective Bargaining and Country Economic Perfor mance Review of the Empirical Literature. Unpublished paper. World Bank, Social Protection Unit, Washington, D.C. Akerlof, George A., and Lawrence F. Katz. 1989. "Workers' Trust Funds and the Logic of Wage Profiles." Quarterly Journal of Economics 104 (August):525 36. Alfandari, Gilles, and Mark E. Schaffer. 1996. "Arrears in the Russian Enterprise Sector." In Simon Commander, Qimaio Fan, and Mark E. Schaffer, eds., Enterprise Restructuring and Economic Policy in Russia. Washington, D.C.: World Bank. Asaad, R. 1999. "Matching Severance Payments with Worker Losses in the Egyptian Public Sector." The World Bank Economic Review 13:117 54. Aslund, Anders. 1997. "Social Problems and Policy in Postcommunist Russia." In Ethan.B. Kapstein and M. Mandelbaum, eds., Sustaining the Transition: The Social Safety Net in Postcommunist Europe. New York, NY: Council on Foreign Relations. Aslund, A. 1999. "Winner Takes All." Carnegie Endowment for International Peace. Processed. Atkinson, A. B., and J. Micklewright. 1992. "Economic Transformation in Eastern Europe and the Distribution of Income." Cambridge University Press. Bertola, G., T. Boeri, and S. Cazes. 2000. "Employment Protection in Industrialized Countries: The Case for New Indicators." International Labour Review 139 (1). Betcherman, G. 2000. An Overview of Labor Markets World Wide: Key Trends and Major Policy Issues. World Bank, HDNSP. _____. 2001., An Overview of Labor Markets Worldwide: Key Trends and Major Policy Issues. World Bank, HDSPN. Betcherman, G., A. Luinstra, and M. Ogawa. 2001. "Labor Market Regulation: Interna tional Experience in Promoting Employment and Social Protection." Unpub lished paper. World Bank, Social Protection Unit. Biletsky, Serhiy, J. David Brown, John S. Earle, Julia Khaleeva, IvanKomarova, Olga Lazareva, and Klara Z. Sabirianova. 2002. "Inside theTransforming Firm." Work ing Paper, Budapest: CEU Labor Project. Bilsen, V. and J. Konings (1998), "Job Creation, Job Destruction, and Growth of Newly Established, Privatized, and State owned Enterprises in Transition Economies: Survey Evidence from Bulgaria, Hungary and Romania", Journal of Comparative Economics, Vol.26, 1998. Blanchard, Olivier. 1997. The Economics of Post Communist Transition. Clarendon Lectures in Economics. Oxford: Oxford University Press. REFERENCES 231 Blanchflower, David G., and Andrew J. Oswald. 1998. "What Makes an Entrepreneur?" Journal of Labor Economics 16(1):26 60. Blanchflower, David.G. 2000. "Self Employment in OECD Countries." Labour Eco nomics. Boeri, Tito, and Katherine Terrell. 2002. "Institutional Determinants of Labor Reallo cation in Transition." Journal of Economic Perspectives 16 (1): 51 76. Brainerd, Elizabeth. 1998. "Winners and Losers in Russia's Economic Transition." American Economic Review 88(5):1094 1116. Broadman, H., ed. (1998). Case by Case Privatization in the Russian Federation: Lessons from International Experience, The World Bank. Broadman, H., ed. (1999). Russian Enterprise Reform: Policies to Further the Transition, The World Bank. Broadman, H., and F. Recanatini. 1999. "Corruption and Market Institutions in Transi tion Countries," Washington, D.C.: World Bank Broadman, H., and F. Recanatini. 2001. "Is Russia Restructuring: New Evidence on Job Creation and Destruction," World Bank Policy Research Paper No. 2641. Broadman, H., and F. Recanatini. (forthcoming). "Seeds of Corruption," Economics of Planning. Brown, J.D, and J.S. Earle, 2002a, Job Reallocation and Productivity Growth: A Com parative Study of Russian and the Ukraine. Draft. Heriot Watt University (j.d.brown@hw.ac.uk) and Upjohn Institute for Employment Research and Cen tral European University (earle@upjohn.org) Brown, J.D, and J.S. Earle, 2002b, The Reallocation of Workers and Jobs in Russian Industry: New Evidence on Measures and Determinants. Draft. Heriot Watt Uni versity (j.d.brown@hw.ac.uk) and Upjohn Institute for Employment Research and Central European University (earle@upjohn.org) Brown, J.D, and J.S. Earle. 2002c. Gross Job Flows in Russian Industry Before and After Reforms: Has Destruction Become More Creative? SITE, Stockholm School of Economics, and CEU, Budapest. Forthcoming in the Journal of Comparative Economics. Bureau of Labor Statistics (BLS). 2000. Employment and Earnings. Washington, DC. Burgess, Simon. 1999. "Reallocation of Labour: An International Comparison Using Job Tenure Data." Processed. Card, D., and A.B. Krueger. 1995. Myth and Measurement: The New Economics of the Minimum Wage. Princeton, NJ: Princeton University Press. _____.2000. "Minimum Wages and Employment: Case Study of the Fast Food Industry in New Jersey and Pennsylvania: Reply." American Economic Review 90(5) Chase, Robert. 1998. "Markets for Communist Human Capital: Returns to Education and Experience in the Czech Republic and Slovakia." Industrial and Labor Rela tions Review 51(3): 401 23. _____., 1999. The Formation of a Labour Market in Russia. Cheltenham: Edward Elgar, Clark, Jon. 1999. "Adversarial and Investigative Approaches to the Arbitral Resolution of Dismissal Disputes: A Comparison of South Africa and the UK." Industrial Law Journal 26(4) 232 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Clark K.B., and L.H. Summers. 1982. Unemployment Insurance and Labor Market Transitions." In M.N. Bailey, ed., Workers, Jobs, and Inflation. Washington, D.C.: The Brookings Institution. Clarke, Simon. 1998b. "The Financial System and Demonetization of the Economy." Processed. Clarke, Simon. 1999. New Forms of Employment and Household Survival Strategies in Russia. ISITO and Centre for Comparative Labour Studies. Coventry, U.K., and Moscow. Clarke, S., and V. Borisov. 1999. "New Forms of Labour Contracts and Labour Flexibil ity in Russia." Economics of Transition 7(3):593 614. Clarke, S., and V. Kabalina. 1999. "Employment in the New Private Sector in Russia." Post Communist Economies 11(4):421 44. _____., 2000. "The New Private Sector in the Russian Labour Market." Europe Asia Studies 52(1):7 32 Commander, S., and R. Jackman. 1993. "Providing Social Benefits in Russia," Policy Research Working Paper No. 1184. EDI World Bank. Commander, Simon, John McHale, and Ruslan Yemtsov. 1995. "Russia." In Simon Commander and Fabrizio Coricelli, eds., Unemployment, Restructuring, and the Labor Market in Eastern Europe and Russia. Washington, D.C.: EDI World Bank. Commander, Tolstopiatenko, and Yemtsov, 1999, "Channels of Re Distribution: Inequality and Poverty in Russia's Transition," Economics of Transition 7(2):411 47. Dar, A., and Z. Tzannatos. 1999. "Active Labor Market Programs: A Review of the Evi dence from Evaluations." Human Development Network, Social Protection Department, World Bank, Washington, DC. Daveri, F, and G. Tabellini. 2000. "Unemployment, Growth and Taxation in Industrial Countries", Economic Policy: A European Forum (30): 49 104. Davis, S., J. Haltiwanger and S. Schuh (1996), "Small Business and Job Creation: Dis secting the Myth and Reassessing the Facts", Small Business Economics, Vol.8, 1996. Davis, S., J. Haltiwanger and S. Schuh (1996), Job Creation and Destruction, MIT Press. De Ferranti, D., G.E. Perry, I.S. Gill, and L. Serven. 2000. Securing Our Future in a Global Economy. World Bank Latin and Caribbean Studies, Washington, D.C.: World Bank. De Melo, Denizer, and Gelb. 1996. From Plan to Market: Patterns of Transition. World Bank Policy Research Paper No. 1564. Denisova, I., G. Friebel, and E. Sadovnikova. 1998a. TheRegulatory Framework of the Russian Labour Market: Current State and Reform Tendencies. Working Paper Series No. 5. Russian European Centre for Economic Policy. _____. 1998b. Internal Labour Markets and Human Resources Policies: An Account from the Existing Surveys. Working Paper Series No. 6. Russian European Centre for Economic Policy. Desai, Padma, and Todd Idson. 2000. "Work without Wages. Russia's Nonpayment Cri sis." Massachusetts Institute of Technology Development Practice. 1993. "Argentina's Privatization Program: Experience, Issues, and Lessons Learned." World Bank. REFERENCES 233 Di Tella, R., and R. MacCulloch. 1999. "The Consequences of Labor Market Flexibility: Panel Evidence Based on Survey Data." www.people.hbs.edu/rditella/research.html. Earle, John S., and Klara Z. Sabirianova. 2000. "Equilibrium Wage Arrears: A Theoreti cal and Empirical Analysis of Institutional Lock In." IZA Discussion Paper No. 11964. _____. (forthcoming). "How Late to Pay. Understanding Wage Arrears in Russia." Jour nal of Labor Economics. Earle, John S., and Sabirianova, Klara Z. 2002b. Structural Change, Mismatch, and Job Mobility: Evidence from Russia." Unpublished manuscript. Earle, John S., and Zuzana Sakova. 2000a. "Business Start Ups or Disguised Unem ployment? Evidence on the Character of Self Employment From Transition Economies." Labour Economics 7(5):575 601. Easterly, W., and P. Viera Da Cunha. 1994. "Financing the Storm: Macroeconomic Cri sis in Russia." Economics of Transition 2 (December):443 65. Edwards and Manning. 2000. "The Economics of Employment Protection and Unem ployment Insurance Schemes: Policy Options for Indonesia, Malaysia, Philip pines and Thailand." In G. Betcherman and R. Islam, eds., East Asian Labor Mar kets and Economic Crisis: Impact, Response and Lessons. Washington, D.C.: World Bank. Elmeskov, J., J. Martin, and S. Scarpetta. 1999. "Key Lessons for Labour Market Reform: Evidence from OECD Country Experiences." Swedish Economic Policy Review European Bank for Reconstruction and Development (EBRD). 2000. Transition Report Update May 2000. London: EBRD. Faggio, G. and J. Konings (1999) "Gross Job Flows and Firm Growth in the Transition Countries: Evidence Using Firm Level Data on Five Countries", CEPR Discussion Paper, No. 2261. Feldstein, M. 1978. "The Effect of Unemployment Insurance on Temporary Layoff Unemployment." American Economic Review 68: 834 46. FIAS (2001) "Administrative Barriers to Investment within Subjects of the Russian Federation", World Bank, Washington, DC. Field, Mark.G. 2000. "The Health and Demographic Crisis in Post Soviet Russia: A Two Phase Development." In Mark.G. Field and Judyth.L. Twigg, eds., Russia's Torn Safety Nets. New York: St. Martin's Press. Fischer, Stanley, and Ratna Sahay. 2000. "The Transition Economies After Ten Years." International Monetary Fund Working Paper. Flanagan, R. 1995. "Wage Structures in the Transition of the Czech Economy." Inter national Monetary Fund Staff Papers 42(4):836 54. Flinn, Christopher. 1986. "Wages and Job Mobility of Young Workers." Journal of Polit ical Economy 94(3):S88 110. Forteza, A., and M. Rama. 2000. "Labor Market 'Rigidity' and the Success of Economic Reforms Across More than One Hundred Countries." World Bank. Processed. Freeman, R.B. 1993. "Labor Market Institutions and Policies: Help or Hindrance to Economic Adjustment?" Proceedings of the World Bank Annual Conference on 234 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Development Economics, 1992. Supplement to the World Bank Economic Review and the World Bank Research Observer. Fretwell, David (2000), "Note on Safety Nets for Economic Restructuring", ECSHD, World Bank. Fretwell, David, and Wilson. 1999. Public Service Employment: A Review of Programs in Selected OECD Countries and Transition Economies. World Bank Discussion Paper No. 9913 Fretwell, David, T. Beck, and E. Johannsson. 1995. "Privatization and Restructuring: Issues Related to Divestiture of Labor and Social Assets." World Bank Working Paper. Fretwell, D., J. Benus, and C.J. O'Leary. 1999. "Evaluating the Impact of Active Labor Programs: Results of Cross Country Studies in Europe and Central Asia." Social Protection Discussion Paper No. 9915. The World Bank. Friebel, G., and S. Guriev. 2000. "Should I Stay or Can I Go?" Draft paper. Gacs, Janos, Robert Holzmann, and George Winckler, eds. 1995. Output Decline in Eastern Europe -- Prospects for Recovery? Kluwer Academic Press. Gaddy, C., and B.W. Ickes. 1998. "Underneath the Formal Economy: Why are Russian Enterprises Not Restructuring?" Transition: The Newsletter about Reforming Economies. World Bank, Development Research Group. Garsia Iser, M., Smirnov, S., Kashepov, A., Babushkina, T and E. Sindyashkina (1998). Rynok truda Rossii. Moscow, Fast Print. Gerber, T. 2000. Regional Migration Dynamics in Russia Since the Collapse of Commu nism. University of Arizona. Tucson. Gimpelson, Vladimir. 1998. "Politics of Labor Market Adjustment (the Case of Rus sia)." Processed. Gimpelson, Vladimir, and Douglas Lippoldt. 1997. "Labour Turnover in the Russian Economy." In OECD, Labour Market Dynamics in the Russian Federation. Paris: OECD. _____. 1999. "Private Sector Employment in Russia: Scale, Composition and Perfor mance." Economics of Transition 7, no. 2(1999): 505 33. _____. 2001. The Russian Labour Market: Between Transition and Turmoil. Lanham: Rowman and Littlefield, 2001. Goskomstat. 1993. Russian Federation in 1992: Statistical Yearbook. Moscow: Goskomstat of Russia. _____. 1996. Strikes in Enterprises and Organizations of the Russian Federation in 1990 1995. _____. 1997. Russian Statistical Yearbook: 1997. Moscow: Goskomstat of Russia. _____. 1998a. Social Economic Situation of Russia 1997. Moscow: Goskomstat of Rus sia. . _____. 1998b. Instructions for the Survey of Population on Problems of Employment, dated December 28, 1998. Moscow: Goskomstat of Russia. _____. 1998c. Russian Statistical Yearbook: 1998. Moscow: Goskomstat of Russia. _____. 1999a. Survey of Population on Problems of Employment: November 1999. Moscow: Goskomstat of Russia. REFERENCES 235 _____. 1999b. Regional Yearbook of Russia. Moscow: Goskomstat of Russia. _____. 1999c. Labor 1999. Moscow: Goskomstat of Russia. _____. 1999d. Russian Statistical Yearbook: 1999. Moscow: Goskomstat of Russia _____. 2000a. Population and Migration in the Russian Federation in 1999. Moscow: Goskomstat of Russia _____. 2000b. Russian Statistical Yearbook: 2000. Moscow: Goskomstat of Russia. _____. 2000c. Survey of Population on Problems of Employment: November 2000. Moscow: Goskomstat of Russia. _____. 2000d. Prices in Russia: 2000. Moscow: Goskomstat of Russia. _____. 2001a. Social and Economic Situation in Russia: January 2001. Moscow: Goskomstat of Russia. _____. 2001b. Short Term Economic Indicators: March 2001. Moscow: Goskomstat of Russia. Graeser, Paul. 1998. "Human Capital in a Centrally Planned Economy: Evidence." Kyk los 41(1): 75 98. Granick, David. 1987. Job Rights in the Soviet Union. Cambridge University Press. Oxford. Gregory, Paul, and Janet Kohlhase. 1988. "The Earnings of Soviet Workers: Evidence from the Soviet Interview Project." Review of Economics and Statistics 70:23 35. Gruber, J. 1997. "Consumption Smoothing Effects of Unemployment Insurance." American Economic Review 87:192 205. Guasch, Luis. 1996. "Estimating the Employment Benefits of Labor Reform in Argentina." World Bank, Sector Study. Haltiwanger J. C., and M. Singh (1999). "Cross Country Evidence on Public Sector Retrenchment." The World Bank Economic Review 13: 23 66 Hamermesh, D.S., and D.T. Slesnick. 1995. "Unemployment Insurance and Household Welfare: Microeconomic Evidence 1980 93." NBER Working Paper No. 5315. Hammer, J.D.G., and C. Ville. 1998. "The Role of Labour Inspection in Transition Economies." ILO Government and Labour Law and Administration, Document No. 48. Available at www.ilo.org./public/english/govlab/papers/1998/labins/index.htm. Heckman, J.J., and C. Pages. 2000. The Cost of Job Security Regulation: Evidence from Latin American Labor Markets. National Bureau of Economic Research, Working Paper No. 7773. Cambridge, MA. Heleniak, T. 1997. "Internal Migration in Russia during Economic Transition." World Bank, International Economics Department. Heleniak, T. (2000), Russia's Modest Migration Gains Unlikely to Stop Population Decline; Population Today, Population Reference Bureau. Hendley, Kathryn;Ickes, Barry; Murrell, Peter; and Ryterman, Randi "Observations on the Use of Law by Russian Enterprises." Post Soviet Affairs, 13, no.1 (January March 1997): 19 41. Heron, R., and C. Vandenabeele. 1999. "Labour Dispute Resolution: An Introductory Guide." ILO Regional Office for Asia and the Pacific, ILO East Asia Multidiscipli nary Advisory Team, Bangkok. Hoeven, Robert, and Gyorgy Sziraczki. 1997. "Lessons from Privatization, Labor Issues in Developing and Transitional Economies." International Labour Organisation, Geneva. 236 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Hoffer, F. 1997. "Traditional Trade Unions During Transition and Economic Reform in the Russian Federation." International Labour Organisation, Enterprise and Cooperative Development Department, Geneva. _____. 2000. "Good Governance and Social Dialogue in Russia." Unpublished paper, ILO. IMC Consulting Ltd. 1999. "Mono Company Towns: Final Report" IMF World Economic Outlook. 1999. Institute for Comparative Labour Relations Research (ISITO). 1998. "Situation with Wage Arrears in the Samara Oblast." ISITO Reports on the Non Payment of Wages in Russia, Moscow. Institute for the Economy in Transition. 1994. "Russian Economy in 1993: Trends and Prospects." Moscow: Institute for the Economy in Transition. International Labour Office (ILO). 1998. 1998 Yearbook of Labour Statistics. Geneva: International Labour Office. International Labour Organisation (ILO).1999. "Social and Labour Issues: Overcom ing Adverse Consequences of the Transition Period in the Russian Federation." Moscow _____. 2000a. "Organization, Bargaining and Dialogue for Development in a Global izing World." Paper prepared for the Working Party on the Social Dimensions of Globalization, ILO. GB.279/WP/SDG/2 _____. 2000b. "Good Practices in Settlement of Individual Labour Disputes: the Case of Chile." Available at www.ilo.org. Ivanova, Nadezhda, and Charles Wyplosz. 1998. "Arrears: The Tide that is Drowning Russia." Paper presented at the RECEP Conference on Economic and Social Reform in Russia, September 1998. Jackman, R. 1998 "Unemployment and Restructuring in Central and Eastern Europe." In P. Boone, S. Gomulka, and R. Layard, eds., Emerging From Communism: Lessons from Russia China and Eastern Europe. Cambridge, MA: MIT Press. Javeline, Debra. 1999. Protest and Passivity: How Russians Respond to Not Getting Paid. Manuscript, Harvard University. Jeon, D., and J. Laffont. 1999. "The Efficient Mechanism for Public Sector Downsizing." The World Bank Economic Review 13:67 88. Johnson, Simon, Daniel Kaufmann, and Andrei Shleifer. 1997. "The Unofficial Econ omy in Transition." Brookings Papers on Economic Activity 2(1):159 239. Kapeliushnikov, Rostislav. 1997. "Job and Labour Turnover in the Russian Industry." Russian Economic Barometer VI(1) _____. 1997. "Job Turnover in a Transitional Economy: The Behaviour and Expecta tions of Russian Industrial Enterprises." In OECD, Labour Market Dynamics in the Russian Federation. Paris: OECD. _____. 1998. "Overemployment in Russian Industry: Roots of the Problem and Pro posed Solutions." Studies on Russian Economic Development 9(6):596 609. Kikeri, Sunita. 1998. "Privatization and Labor, What Happens to Workers When Gov ernments Divest." World Bank Technical Paper No. 396. REFERENCES 237 Konings and Walsh (1999), "Employment Dynamics of Newly Established and Tradi tional Firms: A Comparison of Russia and Ukraine", LICOS Discussion Paper, No. 81/1999. Konings, J., H. Lehmann and M. Schaffer (1996), Job Creation and Job Destruction in a Transition Economy: Ownership, Firm Size and Gross Job Flows in Polish Man ufacturing, 1988 91", Centre for Economic Performance, Discussion Paper, No.282. Kornai, Janos. 1992. The Socialist System: The Political Economy of Communism. Princeton: Princeton University Press. Larroulet, Cristian.1992. "Impact of Privatization: The Chilean Case 1985 1989". Working Paper. Institute of The Americas, San Diego. Layard, Richard, and Andrea Richter. 1995."How Much Unemployment is Needed for Restructuring: The Russian Experience." Economics of Transition, 3(1):35 58. Lazear, Edward P. 1979 "Why is there Mandatory Retirement?" Journal of Political Economy 87(6):1261 84. _____. " (1981) Agency, Earnings Profiles, Productivity, and Hours Restrictions." American Economic Review 71no.4: 606 20. Lazear, E.P. 1990. "Job Security Provisions and Employment." Quarterly Journal of Eco nomics 105(3):699 726. Lehmann, Hartmut H., and Jonathan Wadsworth. 1997. "Job Insecurity and Wage Arrears in the Russian Federation." In OECD, Labour Market Dynamics in the Russian Federation. Paris: OECD. Lehmann, Hartmut H., Jonathan Wadsworth, and Alessandro Acquisti. 1999. "Grime and Punishment: Job Insecurity and Wage Arrears in the Russian Federation." Journal of Comparative Economics 27(4):595 617. Lippoldt, Douglas, and Alex Grey. 1997. "Labour Market Dynamics in the Russian Fed eration: An Introduction and Overview." In OECD, Labour Market Dynamics in the Russian Federation. Paris: OECD. Lokshin , Michael, and Ravallion, Martin (2000). "Welfare impact of the 1998 financial crisis in Rusisan and the response of the Safety Net, Economics of Transition, Vol ume 8 (2). Maleva, T., P. Kudyukin, S. Misikhina, S. Surkov, (2001), "How Much Does the Labor Code Cost?" Working Papers, Issue 3, Carnegie Moscow Center, Moscow. Malle, Silvana. 1990. Employment Planning in the Soviet Union: Continuity and Change. MacMillan, New York. McKinsey Global Institute. 1999. Unlocking Economic Growth in Russia; http://mgi.mckinsey.com/mgi/russian.asp Mitchneck, Beth, and David A. Plane. 1995. "Migration and the Quasi Labor Market in Russia." International Regional Science Review 18(3):267 88. Neal, Derek. 1999. "The Complexity of Job Mobility among Young Men." Journal of Labor Economics 17(2): 237 61. Nesterova, Daria, and Klara Z. Sabirianova. 1998. "Investment in Human Capital under Economic Transformation in Russia." EERC Working Paper No. 99/04. Neumark, D., and W. Wascher. 2000. "Minimum Wages and Employment: Case Study 238 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY of the Fast Food Industry in New Jersey and Pennsylvania: Comment." American Economic Review 90(5) Nickell, S., and R. Layard. 1997. "The Labour Market Consequences of Technical and Struc tural Change." Centre for Economic Performance, Discussion Paper Series, No. 23. North, Douglass C. 1990. Institutions, Institutional Change, and Economic Perfor mance. New York: Cambridge University Press. Organisation for Economic Co operation and Development (OECD). 1995. OECD Economic Surveys: The Russian Federation. Paris: OECD. _____. 1996. Employment Outlook. Paris. _____. 1997. Employment Outlook. Paris. _____. 1997. OECD Economic Surveys 1997 1998: The Russian Federation. Paris: OECD _____. 1998. "Early Identification of Jobseekers at Risk of Long Term Unemployment: The Role of Profiling." Paris _____. 1998. Employment Outlook. Paris. _____. 1999. OECD in Figures. Paris: OECD _____. 1999a. Employment Outlook. Paris. _____. 1999b. Implementing the OECD Jobs Strategy: Assessing Performance and Pol icy. Paris. _____. 2000. Employment Outlook. Paris _____. 2000. "Review Report of Labor Market and Social Policies in the Russian Fed eration: Main Issues for Discussion." Directorate for Education, Employment, Labour and Social Affairs, Employment, Labour and Social Affairs Committee . _____. 2000. OECD Economic Outlook, No. 68, December 2000. Paris: OECD Ofer, Gur, and Aaron Vinokur. 1981. "Earnings Differentials by Sex in the Soviet Union: A First Look." In S. Rosefilde, ed., Economic Welfare and the Economics of Soviet Socialism: Essays in Honor of Abram Berson. Cambridge, MA: Harvard Uni versity Press. Oi, Walter. 1962. "Labor as a Quasi Fixed Factor." Journal of Political Economy 70:538 55. Ovtcharova, L. 2000. "Children at Risk of Poverty and Social Protection Programs." Background Paper for the Child Welfare Report. Oxenstierna, Suzanne. 1989. From Labour Shortage to Unemployment? The Soviet Labour Market in the 1980s. Swedish Institute for Social Research.. Pencavel, John H. 1972. "Wages, Specific Training, and Labor Turnover in U.S. Manu facturing Industries." International Economic Review 13: 53 64. Perova, I., and L. Khakhulina. 1997. "Informal Secondary Employment." VTsIOM Bulletin 6. Pinto, B., V. Debrentsov, and A Alexandrova, (2001). The Russian Federation after the 1998 Crisis: Win Win Strategies for Growth and Social Protection, World Bank. Prokopov, F.T., and T.M. Maleva. 1999. "Anti Unemployment Policy." Bureau of Eco nomic Analysis Foundation, Moscow. Rama, M. 1999. "Public Sector Downsizing: An Introduction." World Bank Economic Review 13(1):1 22. REFERENCES 239 Rashid, M, and J. Rutkowski. 2001. "Labor Market Strategy for Transition Countries: Recent Developments and Future Options," Social Protection Policy Paper, World Bank. Rashid, M. and R. Townsend, 1994. "Targeting Credit and Insurance" Efficency, Mech anism Design, and Program Evaluation, ESP Discussion Paper, World Bank. Rein, Martin, Barry Friedman, and Andreas Woergoetter, eds. 1997. Enterprise and Social Benefits After Communism. Cambridge: Cambridge University Press. Riboud, M, C. Sanchez Paramo, and C. Silva Jauregui. 2001. Does Euroschlerosis Mat ter? Institutional Reform and Labor Market Performance in Central and Eastern European Countries in the 1990s. World Bank. Richter, R. 2000. "Government Cash Transfers, Household Consumption and Poverty Alleviation, the Case of Russia." Discussion Paper No. 2422. Center for Economic Policy Research, London. Rimashevskaya, Natalia, and Valeriy Patsiorkovski. 1992. "Social Economic Research of Welfare and Living Standards of Urban Population." Moscow: ISEPN Russian Economic Trends 4, No. 1 (1995) Russian Economic Trends, October 2000. Russian Expert Institute. 2000. Mono profile Towns and Town forming Enterprises of the Russian Federation. Review Report. Moscow Russian Federation. 1999. Ministry of Labor and Social Development Report on Strikes in the Russian Federation in 1999 Sabirianova, Klara. 2002. "The Great Human Capital Reallocation: A Study of Occupa tional Mobility in Transitional Russia." Journal of Comparative Economics 30(1): 191 217. Sah, Raaj K. 1991. "Social Osmosis and Patterns of Crime," Journal of Political Economy XCIX:1272 95 Salop, Joanne, Steven and Salop. 1976. " Self Selection and Turnover in the Labor Mar ket. Quarterly Journal of Economics 90: 619 2 Schwab, Stewart. 2000. Comments on the Russian Labor Code. Processed Skoufias, E. 2001. "Consumption Smoothing in Russia: Evidence from the RLMS. IFPRI Discussion Paper. Washington, D.C Standing, G. 1999. Global Labour Flexibility: Seeking Distributive Justice. New York: St. Martin's Press State University Higher School of Economics, Institute for Macroeconomic Research of RF Ministry of Economy. 2000. "Development of a Concept and Drafting New Leg islation on Employment of the Population of the Russian Federation.. Report com missioned by the World Bank and Russian Foundation for Social Reforms. Moscow. Strumilin, Stanislav. 1924. "The Economic Significance of National Education." National Planning. Suvorov, A.V. 2000 Macroeconomic Analysis of Poverty Feminization in Russia. Report prepared for the World Bank (translated from Russian Svejnar, Jan. 1999. "Labor Markets in the Transitional Central and East European Economies." In O. Ashenfelter and D. Card, eds., Handbook of Labor Economics, Vol. 3b. Amsterdam: Elsevier Science. 240 THE RUSSIAN LABOR MARKET: MOVING FROM CRISIS TO RECOVERY Tchetvernina, Tatyana, (2000) Employment Services and Social Protection: World Bank Summary Report. _____. 2001. "Unemployment Benefit Programs in Russia. Draft chapters. Unpub lished. Tchetvernina, T.Y., I.V. Soboleva, E.N. Sobolev, P.G. Smirnov, A.A. Moskovskaya, and S.V. Lomonosova. 2001. Employment Situation in Russian Enterprises. Unpublished interim report prepared for the World Bank. Thomason, G. 1993. "Preventing and Resolving Industrial Conflict, Final Report." Labour Market and Social Policy Occasional Papers No. 11. OECD. Topel, R. 1983. "On Layoffs and Unemployment Insurance." American Economic Review 73:541 59. Topel, Robert H., and Michael P. Ward. 1992. "Job Mobility and the Careers of Young Men." Quarterly Journal of Economics 107(2 ):439 79. Treisman, Daniel, and Vladimir Gimpelson. (forthcoming). "Political Business Cycles and Russian Elections, Or the Machinations of Chudar." British Journal of Politi cal Science. Tzannatos, Z., and S. Roddis. 1998. "Unemployment Benefits" World Bank, Human Development Network. Vodopivec, Milan, and Dhushyanth, Raju. 2001. "Income Support Systems for the Unemployed: Issues and Options." Social Protection Department, Human Devel opment Network. Vodopivec, M., A. Worgotter, and D. Raju. 2000. "Unemployment Benefits in Transition Economies: Lessons Learned and Policy Options." Washington D.C. Von Richthofen, W. 1999. "New Prevention Strategies for Labour Inspection, Introduc tion." ILO Government and Labour Law and Administration, Document No. 56. Available at ww.ilo.org./public/english/govlab/papers/1999/labins/index.htm. Vroman, W. (2001). "Unemployment Insurance and Unemployment Assistance: A Comparison."The Urban Institute, processed (prepared for the World Bank). Wesolowsky, T. 2000. "Russia: Trade Unions Face Skeptical Public." Radio Free Europe. www.rferl.org/nca/features/2000/01/F.RU.000112155117.html. World Bank, 1993 "Argentina's Privatization Program, Experience, Issues, and Lessons Learned." Washington, DC. World Bank, 1995. Workers in an Integrating World. World Bank Development Report. Washington, D.C. World Bank, 1997. Poverty Assessment: Poland. Washington, DC World Bank, 1998. Poverty and Social Policy in Romania, Washington DC. World Bank, 2000. Implementation Completion Report, Employment and Social Protection Project, ECSHD. World Bank, 2000a. Balancing Protection and Opportunity: Social Protection Strategy for Transition Economies. Washington, D.C. World Bank, 2000b. "Russia: Poverty in 1998." Draft: not for citation. Human Devel opment and Poverty Reduction and Economic Management. REFERENCES 241 World Bank 2001 "The Russian Federation after the 1998 Crisis: Towards "Win Win" Strategies for Growth and Social Protection." Unpublished paper. Poverty Reduc tion and Economic Management Sector Unit, Europe and Central Asia Region. World Bank. 2001a. Slovak Republic, Living Standards, Employment and Labor Mar ket Study. ECSPE, Washington, D.C. World Bank. 2001b. Poland: Labor Market Study: The Challenge of Moving from Old to New Jobs. ECSPE, Washington, D.C. Zack, A. 1997. "Can Alternative Dispute Resolution Help Resolve Employment Dis putes?" International Labour Review 136(1) « » : 101831, , , , 9 .: (095) 923 68 39, 923 85 68 : (095) 925 42 69 e mail: vesmirorder@vesmirbooks.ru http://www.vesmirbooks.ru : ( ) : .. : .. : .. : .., .. 03510 15.12.2000 (« ») 04.06.2003. . 70´100 1/16 .. . 17 « » 101831, , ., 9 «» 127247, , , 107