o C w- cr * -n f. (\ * E4 Xa r-~~~~~- o QZ CL (74080 1 RECENT WORLD BANK TECHNICAL PAPERS No. 264 Miller and Jones, Orgles1ic andi Coinpostt-Based Groo'ing' Mclia tor Trcc Sccdllin,' Nurscries No. 265 Viswariath, Buildinx Partniershipi;s tor Povertiy Reduction: 1/ic Particilatort, Project PlFnncijug Approach of tilt Wooien'zs 1 EnterprAise l7nageinieciit Traininlg OntreRnch Programn (WEMTb P) No. 266 Hill and Bender, Dcvcloping thic Regui n toril Enzion n eet fct-or Cone petitivc Atgrici ltitiraI Markets No. 267 Valdes and Schaeffer, Surv,cillance 0of Agriciui/tiral Pr-ices na t Trdec: A Handilook for the Dan7eitn1171 Repiublic No. 268 Valdes and Schaeffer, Suorecillance of Agriciultiural Prices a Mid Tra7d1: A Hanilidoak for Co/ovib7i17 No. 269 Scheierling, O,ecrcotnimg Agricnilteiral Polliitioin of IVatcr: The Cllge of rating Agrice ut ii r/ai Environncntal Policies in the Eu lape ni Uizioni No. 270 Banerjee, Rehabilitation of Dcgradcd Forests ini Asia No. 271 Ahnmed, Tcci/iologicail Devclop ciictit ,titi Pollittiou Alatetnticc t: A St inid of Hoav En tcrpriscs Alc Fcii ing Altemrlatiocs to Cilorofluorocirbaoiis No. 272 Greanev and Kellagha n E. iiitel Issl iii Pib/lic Exaetitiitioa s iSl Dc'clojia/u Counlitries No. 273 Grirmshawxv and Helfer, editors, Ve icrr Gross for Soil aiidi Watcr Coiiscr 'ation. Lal id Rciha/ilita time, andl Eiln,a ki'nent StaniliZCation: A Collect ion of Papcrs a lid Ncws/ctterrs Coaiiilce bi1 the Vectivcr Netwvork No. 274 Goxvindaraj, Murray, and Chellaraj, Hcalth Expenditurcs ll I atini Atnicrica No. 275 Heggie, Manai'gcllcit illil Finniliig ot Roails: An A gcila tor RctOrll No. 276 Johnson, Qua? itil R,icw S/uchic's atr A editors: Tlicir Potenttial fr Slib-S1/ia r7 i Afr-ica No. 277 Convery, Applh,ing Eliroln eeeiital Econollonics ill A fr-icc No. 278 Wijetilleke and Karuna rat ne, Air Qiia/li tileia,cL'lit: IuEs idcrptlei.s lr Deeap0il'e Coilintries No. 279 Andersonand Almei, rh case tor Solar E rIlnX!/vy ecto4rnits No. 2S0 Row at, Malik, and Dakohias, In? icial Rctorele ill Lattii ll rica nIrl tlIn' Caribealc e Procecliiegs ofta Worldt Banik Colifi'rectic No. 281 Shen and Con treras-Hermo>i l la, Enz vir 11Iiita? tel 1l Ecoi01 ic tsss ic iil Forcst rue: Selectcd Case St idic< iti Asic No. 282 Kim and Benton, Cost-Benetit Anialisis af tihe Oilchoccrciansis Conitrol1 Pr(ogrelle (O)CP) No. 283 Jacobsen, Scobie and Duncaln, StatiltoriY ltiterelintioel iele A 'griiltin i lrketicu: A Nco Zcn/neatl Pcrspcctioe No. 284 Valdes and Schaeffer in collaboration with Roldos and c hiara .Sirz'eillaiice ofAgriceiltieral Price tilil Trade Policies: A Handilook for LItreg,'nall No. 285 Brehm and Castro, T-heC Market for IVn'tcr Rights iln Chi/c: MAjor Issiics No. 286 Tavoulareas and Charpentier, ClCai Coil7 TclopolonVes for DC L)tIM'lg ColeItrhi's No. 287 Gillhant, Bell, Arin, Matthews, Rumeuer, and Hearin, Cottole Protlictiotn Prospects for flt' Next Dcaidc No. 288 Biggs, Shaw, and Srivastiva, Tcchiilologic) l Capablbilities oh? LCeteriiieig icl Africant Enterprises No. 289 Dinar, Seidl, Olem, Jorden, Duda, and Johnson, Restorinsg oih? Protectieg tite World's Lakes and Reserv1oirs No. 290 Weijenberg, Dagg, Kampen Kalunda, Mailu, Ketema, Na\ arro, and AbLdi Noor, Strewngthieninig N7tionial Agricillteinl Researchi Systcniis iii Eastern 1iiid Cciitral Atrica: A Fralnczi,ork tor Actioci No. 291 Valdes and Schaeffer in collaboration with ErrazuLriz and Francisco, Seirveillacicc of Agriciilterel Price u lid Trade Policies: A Hi d/book for Cl/ile No. 292 Gorriz, Subramanian, and Simas, Irrigietionl Maliagencuecit Trerisfi'r iii Mclxico: Process antd Progress No. 293 Preker and Feachem, Markct Mecicsceist; ancid tlee' Healthi Sector ice Ceoltral iiizd Eistern Eeortpe No. 294 Valdes and Schaeffer in collaboration witlh Sturzenegger and Bebczuk, Slerveillieice of Agriciltiteriel Price acid Trade Policies: A Htntdlitokfior Argt'litiena No. 295 Plohli, Jedrzejczak, and Andersoll, Crtetillg Ci7jeital Markitets ice Ceclttrtfl acel Eistercl Ellrope No. 296 Stassen, Sniall-Sctle Bioniass Geesifiers for Heat ncid Powt'r: A Globi7l Rezietw No. 297 Bulatao, Key Ilcdicators fior Falliily Plaieing Pro/ects No. 298 Odaga and Henexeld, Girls alitL Schools icn Setle-Saiernil Africa: Froni Acnalelsis to Actiotn No. 299 Tamale, Jones, and Pswarayi-Riddihough, Technologies Related to Ptrticipntorui Forestry ice Tropical aced See/ctropical Coiljitrics (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NO. 339 Social Challenges of Transition Series The Financing of Pension Systems in Central and Eastern Europe An Overview of Major Trends and their Determinants, 1990-1993 Emily S. Andrews Mansoora Rashid The World Bank Washington, D. C. Copyright © 1996 The Intemational Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing October 1996 Technical Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Puiblications, which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from the Distribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'1ena, 75116 Paris, France. Enmily S. Andrews and Mansoora Rashid are human resources economists in the Human Resources Operations Division of the World Bank's Europe and Central Asia Department. ISBN 0-8213-3749-1 ISSN: 0253-7494 Cover montage by Joyce C. Petruzzelli. Photos by Nelson Ryland. Library of Congress Cataloging-in-Publication Data Andrews, Emily S. The financing of pension systems in Central and Eastern Europe an overview of major trends and their determinants, 1990-1993 / Emily S. Andrews and Mansoora Rashid. p. cm. - (World Bank technical paper, ISSN 0253-7494; no. 339. Social challenges of transition series) ISBN 0-8213-3749-1 1. Pensions-Europe, Eastem-Finance-Statistics. I. Rashid, Mansoora, 1958- . 11. Title. III. Series: World Bank technical paper ; no. 339. IV. Series: World Bank technical paper. Social challenges of transition series. HD7164.7.A83 1996 96-34970 331.25'2'0947-dc2O CIP -iii- Table of Contents LIST OF FIGURES ............................................... iv FOREWORD ................................................v ABSTRACT ............................................... vii ACKNOWLEDGMENTS ............................................... ix SECTION 1. INTRODUCTION ...............................................1I SECTION 2. TRENDS IN PENSION FINANCING, 1990-1993 ................................................;.3 SECTION 3. SOURCES OF FINANCIAL PROBLEMS IN THE PENSION SYSTEMS, 1993 ........... 12 SECTION 4. CONCLUSIONS ...................................................................... 22 STATISTICAL ANNEX ...................................................................... 23 Annex 1: Data ...................................................................... 23 Annex 2: The Model ...................................................................... 24 Annex 3: Data for Figures ...................................................................... 31 -iv- List of Figures Figure 1 Total Pension Expenditures as a Percentage of GDP, 1990-1993 ...................................3 Figure 2 Trends in Total Pension Expenditures as a Percentage of GDP Relative to 1990 ...........4 Figure 3 Pension Expenditures as a Percentage of Government Expenditure, 1989-1993 ............4 Figure 4 Trends in the Ratio of Contributor per Pensioner Relative to 1990 .............. ..................5 Figure 5a Trends in Number of Pensioners to Population Age 60 and Over Relative to 1990 .......6 Figure 5b Trends in Number of New Pensioners Relative to 1989 ..................................................6 Figure 6 Trends in the Ratio of Contributors to Working Age Population Relative to 1990 ........7 Figure 7 Trends in the Ratio of Population Age 60 and Over to Working Age Population Relative to 1990 ...........................................................7 Figure 8a Replacement Rates for Countries with a Net Increase (Net Wages), 1990-1993 ............9 Figure 8b Replacement Rates for Countries with a Net Decrease (Net Wages), 1990-1993 ...........9 Figure 9a Trends in Wage Share/Tax Base Relative to 1990 for Countries with a Net Decrease. 10 Figure 9b Trends in Wage Share/Tax Base Relative to 1990 for Countries with a Net Increase.. 10 Figure 10 Pension Revenue as a Percentage of GDP, 1990-1993 .................................................. 11 Figure 11 Pension Deficit as a Percentage of GDP (Total Pension Expenditures), 1990-1993 ..... 11 Figure 12 Pension Deficit and Government Deficit as a Percentage of GDP, 1993 ...................... 12 Figure 13 Pension Revenues and Total Pension Expenditures as a Percentage of GDP, 1993 ..... 13 Figure 14 Pension Deficit as a Percentage of GDP (Total Pension Expenditures), 1993 .............. 13 Figure 15 Payroll and Effective Tax Rates, 1993 .......................................................... 14 Figure 16 Actual Deficit vs. Potential Deficit as a Percentage of GDP (Total Pension Expenditures), 1993 .14 Figure 17 Pensioners as a Percentage of Contributors (System Dependency Rates), 1993 ........... 15 Figure 18 Percentage of Persons Age 60 and Over in Total Population, 1993 .............................. 16 Figure 19 Ratio of Working Age Population to Persons Age 60 and Over, 1993 ........... .............. 16 Figure 20 Percentage of Population Receiving Pensions (Primary System), 1993 ........................ 17 Figure 21 Percentage of Old-Age Pensioners in First Year of Retirement Under Age 60 to Total Number of Pensioners, 1993 .................................... ................ 18 Figure 22 Percentage of Pensioners Receiving Disability Benefits, 1993 ..................................... 18 Figure 23 Percentage of Contributors to Working Age Population, 1993 ..................................... 19 Figure 24 Percentage of Employed and Contributors to Working Age Population ....................... 20 Figure 25 Replacement Rates (Net Wages), 1993 .................................................... 20 Figure 26 Pension Benefit Expenditures and Total Pension Expenditures as a Percentage of GDP, 1993 .21 Foreword Transition from central planning in Central and Eastern Europe has led to dramatic and rapid changes in the lives of millions of people. Democratic liberalization has empowered citizens through the provision of basic rights and freedoms and the development of civil society, while economic liberalization has brought the promise of improved living standards, and granted individuals increased choice in consumption, education, health and employment. Despite remarkable expansion of opportunity, transformation has not come without great pain. Poverty has risen throughout the region, real incomes have declined, and crude death rates have climbed in many countries. Equality among households has suffered as a result of increasing income disparity and the erosion of access to basic social services. While the widening income gap is an expected side effect of transition, accompanying the liberalization of prices and wages, heightened social dissatisfaction threatens political stability and the success of economic reforms. On the most basic level, economic recovery and growth in transition depend upon the capacity of individuals to respond and adapt to the changing requirements of market society. The socialist economies embarked upon reform with strong legacies of state commitment to the development of human capital. However, rigidities of central planning and bureaucratic control limited the quality and efficiency of social services. Transition has exacerbated these trends, as unprecedented collapses in output and consequent fiscal crises have undermined the abilities of governments to sustain pre-transition levels of access to social programs and services. At the same time, the introduction of market forces has opened up an array of opportunities for innovation in the provision and financing of health, education and social insurance, and for addressing increases in poverty and unemployment. This paper is one in a series of reports based upon the Social Challenges of Transition (SCT) database. Developed by the Human Resources Operations Division of the Central and Southern Europe Departments of the World Bank, SCT was initiated in order to document empirically the evolving effects of transition on individuals and families. The project examines the social risks facing people and the policy responses taken by governments since 1989, by monitoring indicators of health, earnings, education, labor markets, pensions, social assistance and poverty. The findings contained in this initial set of thematic cross- country papers make a valuable contribution to our understanding of social developments in Central and Eastern Europe five years into the transition. They provide a basis for further improvements in the content and quality of our support to the countries in the region. Kenneth G. Lay Jean-Michel Severino Director Director Country Department I Country Department II Europe and Central Asia Region Europe and Central Asia Region -vii- Abstract The economic transformation in Central and Eastern Europe increased unemployment and pushed workers into early retirement. At the same time, the insolvency of enterprises and the growth of the informal sector reduced tax compliance and the number of contributors to the system. Contrary to common perception, the resultant increase in the ratio of pensioners to contributors did not translate into an increase in pension costs relative to GDP for all countries in the region. Many countries reduced the generosity of their pension systems (pension benefits/wages) to mitigate or completely overcome increases in system dependency rates, and realized fairly stable (Romania and Czech Republic) or declining pension costs (Albania and Croatia) over the first few years of the transition. Countries that realized high and increasing costs (Slovak Republic, Bulgaria, FYR Macedonia and Poland) are therefore largely those that have either raised pension payments relative to wages, or have not been able to reduce pension generosity sufficiently to counteract increases in system dependency rates. The paper concludes that decreasing the level and rate of growth of pension costs is the key to reducing pension fund deficits in evidence in nearly all countries in the region. High payroll tax rates make further increases in revenues (through even higher taxes) an unrealistic scenario. The paper shows that improvements in tax compliance should help improve pension fund revenues. But even under full compliance, payroll tax rates would remain high, constraining employment and dampening economic growth. The paper finds that the source of financial problems in each pension system (aging populations, lax eligibility conditions, high benefit levels, low contributors, large informal sector, high non-benefit expenditures, diminished tax compliance) vary considerably across the region. It concludes that policies to improve pension fund financing cannot be generalized and should be designed to address the specific cause of high pension costs in each country. -ix- Acknowledgments The material reported in this paper was prepared as a part of a larger, multi-sectoral study on the Social Challenges of the Transition in the CEE countries, undertaken by the World Bank's Human Resources Operations Division, Central and Southern Europe Departments (EC I /2HR) under the direction of Ralph W. Harbison, Division Chief. We would like to thank the participants in the SCT workshop, colleagues in ECI/2HR, and Estelle James (peer reviewer) for valuable comments and suggestions. We are also indebted to the officials at the statistical offices in the CEE countries, who provided the data and descriptions on which this paper is based. Brandon Cline provided excellent technical and editorial support. 1. Introduction In Central and Eastern Europe, growing unemployment and enterprise restructuring has pushed workers into ever early retirement, while the insolvency of enterprises and growth of the private sector has reduced tax compliance. Economic restructuring and growing unemployment increased the ratio of pensioners to contributors (the system dependency rate) in all countries, putting upward pressure on pension costs relative to GDP. Contrary to common perception, the paper finds that increased system dependency rates did not increase pension expenditures relative to GDP in all countries in the region. Most countries reduced the generosity of their pension system (measured as the ratio of average pensions to average wages) to dampen the impact of increased system dependency rates on pension costs. Thus, Albania and Croatia realized declines in pension expenditures relative to GDP, while Romania and the Czech Republic maintained fairly constant pension spending over the transition. The countries with sharp growths in pension expenditures were FYR Macedonia, Bulgaria, Poland and Slovakia. In Bulgaria and Slovakia, pension generosity was curtailed but could not compensate for the increase in system dependency rates. In Poland and FYR Macedonia, the Government increased pension payments relative to wages, compounding the financing problems of the pension system. However, payroll tax revenues were not sufficient to cover pension costs. Diminished tax compliance and a fall in the number of contributors reduced pension fund revenues in most countries in the region. As a result, with the exception of Romania , all countries, even those with declining or stable pension costs, realized pension fund deficits in 1993. These findings illustrate that the pay-as-you-go systems that were an integral part of the cradle-to-grave protection afforded by socialist regimes are no longer financially viable and, in keeping with a market economy, the Government role in providing old age security will need to be reduced; individuals will need to take greater responsibility for obtaining income security from private insurance and savings schemes. One way to improve pension fund finances and fiscal discipline and reduce the public role in the provision of pensions is to reduce the costs of public pension schemes. This can be achieved by taking measures to make existing public systems more efficient and equitable. High payroll tax rates make further increases in revenues (through higher taxes) an unrealistic scenario. The absence of full tax compliance is a major factor contributing to low pension revenues, therefore, improving tax compliance will help improve pension fund financing in most countries. However, even with full compliance, high tax rates may continue to have adverse effects on employment and growth. The large share of pension costs in government budgets means reductions in pension expenditures should have a favorable impact on fiscal discipline and macroeconomic stability in the region. Efforts to contain pension costs will need to address the underlying cause of high pension costs in each country. The paper identifies these factors as: (i) aging populations in Hungary and Bulgaria; (ii) extensive coverage of pension systems in Albania, Hungary, Bulgaria, Poland and Slovakia; (iii) too many new pensioners in Albania and FYR Macedonia; (iv) a large number of disability pensions in Poland, and Hungary, and fast increasing number of disability benefits in Romania; (v) generous pension benefits in Poland, FYR Macedonia and Croatia; and (vi) high non-benefit expenditures in Bulgaria, Hungary and FYR Macedonia. The paper also identifies tax compliance as a particular problem for the collection of pension revenues in all countries in the region. These findings raise several questions that are beyond the scope of this report and should be the subject of future research: (i) the reasons (economic or political) for divergent financing paths taken by countries in the region; (ii) whether the different factors responsible for high pension costs reflect optimal 1 Romania realized a deficit in the pension fund in 1995. -2- choices for pension systems for each country; (iii) the implication of these findings for alternate pension financing schemes; and (iv) the impact of pension financing developments on the welfare of pensioners. The paper is organized as follows. The second section examines the trends in pension fund financing and their detenninants: demographic conditions, changes in the eligibility and benefit generosity of pension schemes, and economic developments impacting the number of contributors and tax compliance. The reasons for the disparities in the level of pension expenditures across the region in 1993, based on an analysis of similar determinants, are detailed in the third section of the paper. The fourth section presents the conclusions of the study. Annex 1 provides a description of the data. The model and summary tables of results are given in Annex 2. -3- 2. Trends in Pension Financing, 1990-93 This section evaluates trends in pension financing between 1990 and 1993, the first four years of the transition. It finds system dependency rates increased over the transition for all countries in the region, but, contrary to common perception, they did not always translate into higher pension expenditures relative to GDP. Many countries reduced pension fund generosity to counteract increases in the number of pensioners in the system. As a result, some countries charted stable or even declining pension expenditures relative to GDP over the transition. However, revenues did not keep pace with pension fund expenditures. Thus, pension deficits increased as a share of GDP in almost all countries in the region, even in most of those that realized low or stable pension costs. Trends in pension benefit expenditures, the major component of total pension benefit expenditures, are explained by changes in the system dependency rate, the ratio of pensioners to contributors, and the transfer ratio, the ratio of average pensions to average income, a measure that captures the generosity of the pension system (see Annex 2 for details). Revenue trends are analyzed by changes in the tax base and the tax rate. The deficit of the pension fund is defined as the difference between revenues and total pension expenditures (benefits, administrative and other expenses of the fund). Pension expenditures have not increased uniformly in the region. Transition economies can be grouped into three categories based on how pension expenditures changed relative to GDP: 0 High pension expenditure growth countries: FYR Macedonia, Poland, Slovakia and Bulgaria. Bulgaria, FYR Macedonia, Poland, and Slovakia realized the sharpest increases in total pension expenditures (benefits and administration) relative to GDP between 1990 and 1993 (Figures 1, 2). Pension expenditures to GDP increased 6 percent for Poland, 3 percent in Bulgaria and 2 percent in FYR Macedonia and Slovakia over this period. As a result, in 1993, Bulgaria, Poland and FYR Macedonia supported the costliest pension systems in the region. Expenditures on pensions cost Bulgaria 14 percent of GDP, while FYR Macedonia and Poland spent 13 percent of economic output on pensions. Pensions also represented a large share of total government outlays in these countries. Poland and FYR Macedonia spent nearly a quarter of all government outlays on pensions, while Bulgaria spent more than 20 percent of the government budget on pensions (Figure 3).2 14% 12% 10% 16% 019 14%- _ __ Macedonia Poland Bulgaria Croatia Albania Hungary Slovak Czech Romania Figure 1 Total Pension Expenditures as a Percentage of GDP, 1990-1993 2 Pension spending as a share of total govemment expenditure increased over the transition for almost all countries in the region (Figure 3). 4- FIGURE 2 Trends in Total Pension Expenditures as a Percentage of GDP Relative to 1990 1.80 1.60 - ---/Albania -.4---Bulgaria -* Czech 1.,oo 2- - ---- - - )_- Hungary -0- Macedonia 10.80 -- - -- - - - _ Poland l-eO Romania 0.60 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - -- +-4 - Slovak 0 .4 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0 J.2 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0.00 1990 1991 1992 1993 FIGURE 3 Pension Expenditures as a Percentage of Government Expenditure, 1989-1993 30% 25% 20% _ 1989 15% n19 I 10% d l I i *199 5% Poland Macedonia Bulgana Slovak Czech Romania Albania Hungary -5- O Low pension expenditure growth countries: the Czech Republic and Romania. Not all countries realized an increase in pension spending relative to economic output. The Czech Republic and Romania maintained pension spending fairly constant relative to GDP between 1989-93. O Countries with declining pension expenditures. Albania and Croatia. Albania and Croatia reduced pension expenditures relative to GDP over the transition. In 1993, these countries had the least expensive pension systems in the region. Pension expenditures constituted 5 percent of GDP in Albania and nearly 6 percent of GDP in Croatia. What explains these divergent trends in pension expenditures? Pension benefit expenditures comprise the bulk of pension costs. Therefore, we can explain trends in pension expenditures between 1990-93 by evaluating the reasons for changes in pension benefit expenditures over this period. As noted above in equation la, changes in pension benefit expenditures relative to GDP can be explained by changes in the system dependency rate and the transfer ratio. The system dependency rate. The system dependency rate increased in all countries. The ratio of contributors to pensioners decreased (or the system dependency rate, the inverse of this ratio, increased) for all countries in the region (Figure 4), putting upward pressure on pension costs. The largest decline in the ratio of contributors to pensioners occurred in Albania, Bulgaria and Poland. It was least marked in FYR Macedonia and Czech Republic. The reasons for changes in system dependency rates varied across countries. As shown in Annex 2, changes in system dependency rates can be influenced by changes in: (i) the coverage of pension systems--reflected in the ratio of pensioners to population over 60 years of age; (ii) economic conditions--indicated by the contributor to working age population; and (iii) changes in the demographic burden--measured by changes in the proportion of aged in total population. O The number of pensioners increased. The shedding of redundant labor in state-owned enterprises increased the ratio of pensioners to the population over 60 years of age in all transition economies (Figure 5a). Workers were often granted special early retirement options (now more restricted) or disability entitlement to induce them to leave their jobs without joining the ranks of the unemployed. Eligibility conditions were relaxed the most in Albania and Croatia and were most contained in the Czech Republic and Bulgaria. The increase in pensioners occurred mainly in 1991, the year that the greatest retrenchment of labor occurred in the region. By 1993, the number of new pensioners had 1.00 0.80 - - - - - - - - - - - - - - - - - E3~~~~~~~-9- Albania 7*-+ Bulgria -4Croatia 0.70 -- - - - - - - - - - - - - - - - - - - - - - - - -----Czech I& Hungary 0.60 -- - - - - - - - - - - - - -~ - -_ M acedonia: a Poland 0.50 -_ __ __ - - __ - _. - - - Romnia . \ Slovak 0 .4 0 - - - - - - - -_ - - - - - - - - - - - - - - - - - - - 0.30 1990 1991 1992 1993 Figure 4 Trends in the Ratio of Contributor per Pensioner Relative to 1990 -6- FIGURE 5a Trends in Number of Pensioners to Population 60 and Over Relative to 1990 1.35- _ 2- -Albania 1.25 -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 40- B ulgaria IU-- Croatia L z u -- - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - - - - - - - - - ~ ~~C zech orL_Mja~~~~~~~~~~~~~~~~~~~~~~~~ugr 1.15 ------------------------ ------- ~ ~ ~ ~ ~ ~--- -*-Macedonia -II-Poland -0- Romania 1.10 -------------- ----- ~ ~~~~ ~~~~~~~-40--Slovak 1.00T 1990 1991 1992 1993 FIGURE 5b Trends in Number of New Pensioners Relative to 1989 4.50 4.00--_------------------- ~-- --------- 3_B Albania 3___Bulgaria 3.00 ----------------------------U1-- Croatia Czech 2.50- ---Macedonia 1.00- 0 .5 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0.00 I I 1989 1990 1991 1992 1993 'Croatia: 1991 data absent -7- 1.00 * 0 Albania -4- Bulgaria 08 Croatia ---Czech 0.70 -- - - ---- - Hungary \ i Macedonia 0.60 -- - - - Poland \ | oRomania 0.50 ~ ; +Slovak 0.40 I 1990 1991 1992 1993 Figure 6 Trends in the Ratio of Contributors to Working Age Population Relative to 1990 NOTE: Hungary and Macedonia: Employment proxied for Contributors. subsided to 1989 levels in all countries except Croatia, FYR Macedonia and Albania. In fact, new pensioners continued to strain pension finances in FYR Macedonia and Albania in 1993 (Figure Sb). 0 The number of contributors to the primary system declined. The increase in unemployment, reduction in the labor force due to lay offs, and the increase in unregistered informal sector workers sharply reduced the tax base for pension funds. The ratio of contributors to working age population declined in all countries in the region (Figure 6). The decline was the greatest in Albania, Bulgaria, Slovakia and Hungary. It was the least steep in FYR Macedonia and Poland. 0 The aging of the population did not contribute greatly to financing trends between 1990-93. Not surprisingly, the aging of the population, reflected by the ratio of population over age 60 to working age population, did not change much over the short period under study (Figure 7). The demographic burden increased only 5 percent for FYR Macedonia and Romania, while it fell by nearly 5 percent in Slovakia and 10 percent in Croatia -- perhaps a reflection of worsening mortality rates among the aged in these countries. 1 .0 8 - - - - - - - - - - - - - - - - - 1.06 -- - - - - - ----- = =_--.Albania 1.04 ---------j Bulgaria 1.02 = z =~~~~~~~~~~~~~~~~~~~~~~~~~~Croatia 1.00 1 xHungary 0~~~~~~~~~~~~~~~~~~~~~~~~~~~~.98~-\ Macedonia 0.98 ---- -------- - -- Poland 0.96 . - - e Romania 0.94 \ + - ~.~ Slovak 0.92 - - 0.90 1990 1991 1992 1993 Figure 7 Trends in the Ratio of Population Age 60 and Over to the Working Age Population Relative to 1990 -8- The transfer ratio. The transfer ratio declined in most countries across the region. The transfer ratio is a product of two factors: (i) the replacement rate, the ratio of pensions to wages and (ii) the wage share--the ratio of wages to total income (see Annex 2 for details). The decline in the transfer ratio reflects a declines in both components of the transfer ratio--the share of wages in total income and replacement rates--declined across the region. Thus, many countries controlled the upward pressure on pension costs created by the rise in system dependency rates by reducing the generosity of the pension system. O Replacement rates declined in most countries in the region. Albania, Bulgaria, Croatia, the Czech Republic, Hungary, Romania and Slovakia all reduced their pension to wage replacement rates over the transition as a means of controlling expenditure growth (Figures 8a, b). Declines in the replacement rate were the highest in Romania (16%), Albania (11%) and Croatia (11%) and were the least marked in the Czech Republic (4%) and Slovakia (2%). O Replacement rates increased in Poland and FYR Macedonia. Poland and to a lesser degree, FYR Macedonia, increased the generosity of their pension systems by allowing average pensions to rise relative to wages. Replacement rates increased by nearly 18 percent in Poland and by 3 percent in FYR Macedonia. O The wage share declined or remained unchanged in most countries in the region, complementing the declines in replacement rates (Figures 9a, b). The only exceptions were Romania and FYR Macedonia where the wage share increased between 1990 and 1993. In FYR Macedonia, the rise in the wage share compounded the increase in replacement rate, so that the transfer ratio, increased. In Romania, the decline in replacement rate overwhelmed the increase in the wage share, so that the transfer ratio declined. What were the most important factors explaining changes in pension expenditures in each country? A summary of the main factors contributing to pension cost trends in each country are summarized below.3 O High expenditure growth in FYR Macedonia, Poland, Slovakia and Bulgaria reflect diverse pressures on pension funds. * In Poland, expenditures increased relative to GDP because of an increase in the number of pensioners, a decline in the number of contributors, and an increase in the replacement rate. * In FYR Macedonia, an increase in the number of pensioners and the replacement rate were most important factors in explaining an increase in pension costs. The number of contributors declined slightly and did not contribute greatly to the growth in pension expenditures. * High growth rates of pension expenditures relative to GDP in Bulgaria and Slovakia were mainly due to a decline in contributors. In both these countries, the increase in pensioners was marginal and replacement rates (and the transfer ratio) fell over the transition. 0 The low rate of growth of pension expenditures in Romania and the Czech Republic are explained by reductions in pension fund generosity. Both countries because reduced pension fund generosity, by reducing replacement rates, mitigating increases in the system dependency rate. 0 Pension expenditures declined relative to output in Albania and Croatia due to reductions in pension fund generosity. Reductions in pension fund generosity completely overwhelmed the large increases in system dependency rates in both countries. In both countries, the transfer ratio declined because of declines in the replacement rate and the wage share. 3These results are based on an analysis of the decompositions of equations 1. 2 and 3 and are presented in the Annex, Tables 1, 2, and 3 respectively. -9- FIGURE 8a Replacement Rates for Countries with a Net Increase (Net Wages), 1990-1993 85% 75% 65% Macedonia 1 55% --- .-- - Poland 45% . 35% 25% 1990 1991 1992 1993 FIGURE 8b 80% Replacement Rates for Countries with a Net Decrease (Net Wages), 1990-1993 70% - --- - - - - --- -A-ba- -a 0-4-- Bulgaria 50% -----------------------~~ ~~~~~~~-U- I - Croatia 50%- g =Czech - Hungaiy 40% ---------- - Romania \ +-Slovak 3 0 % - - - - - - - - - - - - - - - - - - - - - - - - - 20% 0% 1990 1991 1992 1993 -10- FIGURE 9a Trends in Wage Share/Tax Base Relative to 1990 for Countries with a Net Decrease 1.30 1.20 1.10 -. o e -6--Albania 0.90 -f _-Hungary -U-Croatia 0.80 0.70 0.60 0.50 - - 0.40 0.30 1990 1991 1992 1993 FIGURE 9b Trends in Wage Share/Tax Base Relative to 1990 For Countries 1.30 with a Net Increase 1.30 -_ .000------ -- - -0- Macedonia 0.90 - - - - - - - - - - - - - - - - - - - - - - - - - --- - Poland -4--Bulgaria 0.80 - - - - - - - - - --.- - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- - - -- - 9 LR om ania 0 .7 0 - - - - - - - - - - - - - - - - - - - - - -_-- - - - - 0 .6 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0 .4 0 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 0.30 1 1990 1991 1992 1993 14%/n - 12% 10% a 1990 8% ~~~~~~~~~~~~~~~~~~~~~~~~~~~~01991 6% 01992- U 1993 4% ~ 2% 0% --- Poland Bulgaria Macedonia Hujgary Romania Slovak Czech Croatia Albania Figure 10 Pension Revenue as a Percentage of GDP. 1990-1993 Revenues fell over the transition in most transition economies. The changes in pension expenditures described above coincided with a decline or tapering off of pension fund revenues (Figure 10). The change in the ratio of revenues to GDP can be described by changes in the tax base relative to GDP (the number of contributors times the average wage expressed as a share of GDP) and the tax rate (see Annex 2). Perhaps surprisingly, sharp declines in revenue are rare and have occurred only in Croatia and Albania (1991/93).4 In these countries, the decline in revenues is explained largely by a decline in the tax base (Figures 9a, b). In Romania (1990-92), Bulgaria (l991-93)5, FYR Macedonia (1990-93), and the Czech Republic and Slovakia (1992-93). revenues increased as a share of GDP. In all these countries, the increase in revenues approximate trends in the tax base (Figures 9a, b). Since contributors have declined in all countries, the increase in tax base is indicative of an increase in the ratio of wage to GDP. A greater share of average wage in economic output reflects a larger revenue base relative to GDP. In Poland, revenues increased between 1990 and 1993, while the wage share declined. Therefore, increases in the payroll tax rate most likely explain increases in pension fund revenues in Poland. Pension fund deficits grew over the transition. Pension fund revenues did not keep pace with pension fund expenditures. Therefore, in most countries, pension fund deficits increased or fund surpluses were reduced. The deficit increased the most in FYR Macedonia (3.5%), followed by Bulgaria (3%) and Poland (2%) (Figure 11). Despite declines in pension fund expenditures, pension fund 2% I1% 0% n 1990 -1%-019 -2% 01992 -3% * 1993 -5% -6% Macedonia Albania Bulgaria Slovak Hungarv Poland Croatia Czech Romania Figure 11 Pension Deficit as a Percentage of GDP (Total Pension Expenditures), 1990-1993 4Time series data for gross wages and payroll tax rates are not complete. Therefore we have used net wages to approximate changes in the tax base. Note that the wage share, the ratio of wage/GDP/contributor is identical to the ratio of the wage bill/GDP. where the wage bill is defined as wage*contributors. The data show that trends in gross and net wages are similar for the period under study. 5The slight decline in revenues/gdp in Bulgaria between 1992/93. despite an increase in the tax base. likely reflects a decline in contributors to the pension system. -12- 6% . . ..... .. .. 4% 2% 0%- -2% 3_ -4% -6% _ _ 0 -13Pension -8% * Government -10% -- -12% -14% 4 - ____ __. _ -16% _-_-- - ._ _ .. Bulgaria Macedonia Slovak Hungary Croatia Poland Albania Romania Czech Figure 12 Pension Deficit and Government Deficit as a Percentage of GDP, 1993 deficits increased in Albania and Croatia. In Romania, prudent pension policies helped the pension fund increase its surplus between 1990 and 1992, although it has started to decline since then. Slovakia, Czech Republic and Hungary were the only countries in which the pension fund deficit declined since 1992. However, in these countries, revenue data are only available for 1992 and 1993 so strict comparison to other countries is not possible. 3. Sources of Financial Problems in Pension Systems, 1993 The trends described above resulted in pension fund deficits for all countries (with the exception of Romania) in 1993. This section evaluates differences in pension fund finances across countries in 1993, the most recent year for which we have data. There are two obvious ways to reduce deficits: increase revenues and reduce expenditures. This section shows that increases in pension fund revenues are not possible in the short run. The high payroll tax rates prevailing in many countries mean that further increases in payroll tax rates are not feasible. There does appear to be some scope for increasing compliance, particularly for low pension cost countries. But, even with full compliance, payroll tax rates will remain high and are likely to distort labor markets and constrain economic growth. Thus, reducing the level of pension expenditures will be critical for reducing pension fund deficits. This section reveals that reducing pension costs will imply different strategies because the factors responsible for raising pension costs differ across the region. Almost all countries realized deficits in their pension funds in 1993. Pension deficits put pressure on the budget with potentially adverse impacts on fiscal deficits and macroeconomic stability. The pension deficit contributed to the overall budget deficit for all countries (except the Czech Republic and Romania) in the region (Figure 12). This reflects the large share of pension expenditures in government budgets in transition economies and indicates the critical importance of pension reform to restoring fiscal discipline and economic growth in the region. This need for reform is the most critical in countries which have the highest deficits in their pension funds relative to economic output. O High Deficit Countries (>2% GDP). FYR Macedonia, Albania, Bulgaria and Slovakia had the largest pension deficits relative to GDP in 1993 (Figures 13, 14, next page). With the exception of Albania, these countries also had the highest expenditures relative to GDP (Figure 1). The deficit was the largest in FYR Macedonia (almost 6% of GDP), followed by Albania, Bulgaria and Slovakia. 0 Low Deficit Countries (1-2% GDP). The Czech Republic and Croatia had the smallest pension deficit (1.2% of GDP), followed by Poland.6 Pension fund deficits in Hungary was higher -- close to 2 percent of GDP. 6 Poland was the only high expenditure country to realize low deficits in the pension fund. But, Polish data excludes the farmers fund and so underestimates the pension fund deficit. -13- FIGURE 13 Pension Revenues and Total Pension Expenditures as a Percentage of GDP, 1993 14% - 412%1 l- l -ORevenues/GDP 10% ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ _____ETota] Pension Deficit as a______ _ Percentage ofGP(TtlPnso xeniue),1 Expenditure/GDP 8% 6% 4% 2% 0% Macedonia Bulgaria Poland Hungary Slovak Czech Albania Romania Croatia FIGURE 14 Pension Deficit as a Percentage of GDP (Total Pension Expenditures), 1993 1% 0% -2% -3% -4% -5% __________________ -6%- Macedonia Albania Bulgaria Slovak Hungary Poland Croatia Czech Romania -14- 40% 35% 30% 25% 1 1 i3fi z Payroll Tax Rate 20% g Effective Tax Rate 15% 10% 5% 0% LL Croatia Slovak Macedonia Czech Albania Romania Hungary Bulgaria Poland Figure 15 Payroll and Effective Tax Rates, 1993 0 Surplus Countries Romania was the only country with a surplus in its pension fund in 1993. The surplus has however declined since that year and the pension fund realized a deficit in 1995. How can countries reduce pension fund deficits? The obvious answer is to raise pension revenues and decrease pension expenditures. One way to raise revenues is to improve tax compliance. High payroll tax rates make further increases in these tax rates untenable. Figure 15 shows that tax compliance is a major issue in most countries. Effective tax rates, the ratio of actual to potential revenues, are much lower than the official payroll tax rates in the region. If countries could achieve full tax compliance under existing tax rates, revenue levels would increase sufficiently to reduce pension deficits in all countries in the region. This can be seen by comparing potential deficits, those that would obtain under full tax compliance (assuming existing tax rates) to actual deficits in Figure 16. In 1993, under full tax compliance, pension funds would have realized surpluses in all countries, except FYR Macedonia, Albania and Bulgaria. 3 % 2% _ __-___ 1%. 0% -2% -3%. . * - Potential 4% _ Actual -4% .- __- _ -6% . _ _ _ _ _ _ _ _ _ _ _ _ _ Macedonia Albania Bulgaria Slovak Hungary Poland Croatia Czech Romania Figure 16 Actual Deficit vs. Potential Deficit as a Percentage of GDP (Total Pension Expenditures), 1993 -1 5- Even if countries achieve full tax compliance, financing problems remain. First, payroll tax rates remain high and impose high costs on current workers. with adverse implications for employment and growth. This problem is all the more acute in countries that are supporting expensive pensions. Second, tax compliance may not be independent of high payroll tax rates. That is, compliance may increase only if tax rates are reduced and this may be possible only when economic growth resumes. There is an endogeneity problem: high tax rates may themselves distort labor markets and constrain economic growth. Thus, reduction of pension costs is the most effective way of reducing deficits There is scope for low expenditure countries, such as Croatia, to focus on strengthening tax compliance. But, reducing pension costs is critical for FYR Macedonia, Bulgaria. Slovakia and Poland -- countries with the highest pension expenditures relative to GDP in the region.7 How can countries reduce pension costs? This section shows that controlling costs on pension benefits will vary across the region depending on the reason for high pension costs in particular countries. Countries can control pension costs by reducing (i) system dependency rates (ii) generous benefit levels and (iii) high non-benefit expenditures in the region. (i) System dependency rates vary considerably across the region (Figure 17). Albania, Hungary and Bulgaria had the highest system dependency rates while FYR Macedonia had the lowest ratio of pensioners to contributors in the region. As noted above in equation 3a, variations in system dependency rates can be explained by differences in: (i) aging populations as measured by: the ratio of population age 60 to total population and the ratio of population over 60 to working age population; (ii) extensive coverage of the pension system, as measured by the ratio of pensioners to population over 60 years of age; and (iii) adverse economic conditions, as measured by the ratio of contributors to working age population. Aging Populations. The current demographics of the region indicate an aging population. Population aging increases pension costs even in well-designed pay-as-you-go systems. As noted in equation 3a, the demographic burden of a pension system can be identified by a high ratio of pensioners in total population and/or a high ratio of pensioners to working age population. 900% 80% 70%- MM% iSQSit 60% 50% 200/ 10% 0% Albania Bulgaria Hungary Slovak Poland Czech Romania Croatia Macedonia Figure 17 Pensioners as a Percentage of Contributors (System Dependency Rates), 1993 7Although Poland has no deficit. it has one of the most expensive pension systems in the region, and should consider ways to reduce its pension costs As noted above, countries with no deficits but with high pension costs relative to output are also experiencing financial stress because they impose a considerable tax burden on current employees -16- _5 O I ,- . . 2(% N iFaM 12 S1 ~~~~~* i1; E..7 11l1 Bulgaria H.' =arv Czech Romnania, Croatia Slovenia Slovak Poland Macedonia Albania Figure 18 I'ercentage o0 '-rsons Age 60 ancd Over in Total Population, 1993 H k,.ii Prop'riioni (if agedl in toial population. Figure 18 shows that the burden of an aging 'iop.uatioii is most critical in Bulgaria with 21 percent of the population age 60 and above, toilowed by lwingary (19%) and the Czech Republic (18%). FYR Macedonia and Poland have more favorable demographics with 12 percent and 13 percent, respectively, of aged in the total population. Albaniia has the most favorable demographic situation with only 8 percent of the popuwation age 60 and over. * [!,.gh vroportiwi of aged to wiorking age population. The viability of a pension system is also related to the size of the working age population (15-59) available to support pensioners. Two countries withi the same number of aged (60 and above) may have different proportions of working age individuals to support the aged. The country with a greater pool of working age iiidividuals will be better able to support the same number of aged in a pay-as-you-go pension system. The Eatio of the working age population to persons aged 60 and above is given in Figure i9 'he tigure shows that Bulgaria and Hungary have the highest demographic burden while Albania. FYR Macedonlla. and Poland have the most favorable demographic situation. Eaensive (Coverage. FIxtensive coverage may overshadow both adverse and favorable demographic factors. Pensioni systems that favor early retirement and systems that have pensioned-off redundant workers will raise the pensioner-to-working-age population ratio and possibly result in higher pension costs. 8.0 70 6.0 5.0 2.0 .0 0).0 Albania Macedonia Poland Slovak Slovenia Croatia Romania Czech Hungar-y Bulgaria Figure 19 R atio of Wvorking Age Population to Persons Age 60 and Over, 1993 -17- 30% 10% 0% Bulgaria Hungary Czech Slovak Poland Croatia Romania Albania Macedonia Figure 20 Percentage of Population Receiving Pensions (Primary System). 1993 * Too AMany Pensioners. Almost all countries have a higher number of pensioners than persons age 60 and above in total population (compare Figure 20 to Figure 18). One reason for this finding is that even in well-run pension systems, pensions are paid to young survivors and disabled workers in addition to old-age pensioners. Another reason is that old-age pensions are paid to individuals under 60 years of age (typically women, and special categories of workers). In FYR Macedonia, Romania and to a lesser degree in Croatia, the proportion of pensioners in the primary system is less than the proportion of persons age 60 and older. While some pensioners in secondary systems may be excluded from the index°, some elderly persons may not be receiving any pension. Comparing Slovakia and Croatia suggests that eligibility rules in some countries allow them to reduce their fund costs to compensate for unfavorable demographics, while others tend to worsen their pension burden. In terms of demographic burden, Croatia has a higher proportion of aged, but it supports a smaller number of pensioners than Slovakia. Similarly, while Albania and FYR Macedonia both have favorable demographic conditions, pension eligibility is more lax in Albania than FYR Macedonia. This is reflected by a higher proportion of pensioners in total population relative to aged in total population in Albania. * Too Many New Pensioners. The higher proportion of early retirees throughout the region justifies concerns about the fiscal stability of pension systems which have both early normal retirement ages and special occupational exceptions. The issue of early retirement can be evaluated based on the proportion of old age pensions received by individuals less than 60 years of age to total pensioners in the first year of retirement. Figure 21 shows that this proportion in very high for all countries in Central and Eastern Europe. On average, 60 percent of all pensioners are below age 60 when they begin retirement. The high percentages in Croatia (90%) and Slovakia (76%) are likely to be a result of generous early and special retirement policies. In part, this reflects earlier retirement ages of women. These findings suggest that the pension burden in Croatia is intensified by early retirement. At the same time, some portion of the elderly population may not be receiving pensions, reducing the burden. * Too Many Disability Pensioners. While some pension systems have many old age pensioners, others have a large proportion of individuals on disability benefits. Disability policy is a challenge since a lack of clear criteria can lead to system abuse. The identification of systems with heavy disability burdens represents a first step towards expenditure control. The proportion of pensioners on disability reflects the relative ease of obtaining and retaining these benefits in different countries. Thus, Poland and Romania have the greatest problem in enforcing disability in the region. Although the majority of countries have 19 percent to 23 percent of pensioners on disability, nearly 37 percent of all pensioners in Poland are on disability (Figure 22, previous This is a particular problem in FYR of Macedonia where not all pensioners are included. -18- FIGURE 21 Percentage of Old-Age Pensioners in First Year of Retirement Under Age 60 to Total Number of Pensioners, 1993 100% ..- 90% 80% 70% 600/ 40% 300%- 20% 10% 0% it X Croatia Slovak Czech Poland Bulgaria Macedonia FIGURE 22 Percentage of Pensioners Receiving Disablility Benefits, 1993 40/o % . _ . . ..-.-. 35% 30% 25% 210% 3 5 ON r20 10% 5% 0% Poland Slovak Hungary Romania Croatia Macedonia Czech Albania Bulgaria -19- 70% - _ - ---- l 60% _ 50% _-=s-_| 40% _ Czech Hungary Slovak Bulgaria Croatia Poland Macedonia Romania Albania Figure 23 Percenatge of Contributors to Working Age Population, 1993 NOTE: FYR Macedonia and Hungary: Employment proxied for Contributors page). The situation in Poland reflects ease of recipiency and lack of monitoring control. In Romania, eligibility standards are known to have been severe; in fact, in recent years, as conditions have eased, there has been a relative influx of disabled pensioners. In contrast, Albania and Bulgaria have the smallest proportion of disabled pensioners. In Bulgaria, with its aging population, special categories of early retirement substitute for disability payments, providing pensions to both the able-bodied and the disabled. In Albania, a favorable demographic profile may reduce the disability rolls, as disability is strongly correlated with age. Reduced Contribution Base. Pension systems with a will have greater problems in financing pension costs for a given lower ratio of contributors to working age population number of pensioners as compared to those with a larger contribution base. Figure 23 shows that the Czech Republic and Hungary have the highest contribution base,9 while Albania and Romania have the lowest ratio of contributors relative to the work force. The policies to tackle a low number of contributors to the pension system will differ depending on whether a low number of contributors is the result of the participation of workers in the informal economy or the result of high rates of unemployment and early retirement policies. In the former case, better enforcement will be the key to increasing the number of contributors while in the latter, economic growth will raise the contribution base. * Increase in Informal Activity. An indication of the extent of worker participation in the informal economy can be gauged by comparing the ratio of contributors and employed to the working age population. Countries with a lower ratio of contributors to employed are those where worker participation in the informal economy is largely responsible for low level of contributors. Figure 24, next page, shows that the informal economy plays a significant role in reducing contributors in Albania, Poland and Romania. In these countries the number of employed far exceeds the number of contributors to the pension system. Thus, better enforcement should help reduce pension costs in these countries. * Decline in Economic Activity. In the remaining countries, contributors are roughly the same as the number of employed. Thus, low employment levels and the attrition of contributors as a result of early retirement are the reasons for the differing levels of contributors to working age population. In these countries, an increase in employment, as a result of economic growth, will be the key to increasing contributors to the pension system. 9In FYR Macedonia and Hungary, we have proxied the number of contributors by the number of employed. -20- 80% _______ _______ 70% 60% 50% El Employment 40% U Contributor 30% 20% 10% 0% Romania Czech Slovak Bulgaria Croatia Albania Poland Figure 24 Percentage of Employed and Contributor to Working Age Population, 1993 NOTES: Romania: 1992 data; Hungary and FYR Macedonia: Contributor data absent. (ii) Pension Fund Generosity. Even pension systems that have not overburdened their pensioner rolls may have financial difficulties if pension benefits are too high relative to wages. Similarly, countries with high system dependency rates may have low expenditures if pension benefits are not very generous. The relative generosity of a pension system can be measured by the ratio of average pensions to the average wage (replacement rate) prevailing in the country. Countries with generous benefit formulae and benefit indexation policies are more likely to have fiscal problems than those with stringent controls. The ratio of benefits to net wages for 1993 are given in Figure 25. FYR Macedonia and Poland have the most generous pension systems with an average replacement rate of over 75 percent. Albania, Bulgaria and Romania have the lowest replacement rates in the region, where pensions are around 40 percent of wages in Albania and Bulgaria and 25 percent of wages in Romania. Fiscal problems in Poland and FYR Macedonia are therefore exacerbated by excessive benefit generosity despite their relatively favorable demographic profiles. FYR Macedonia and Croatia provide an interesting contrast. Both countries inherited the same pension system from former Yugoslavia. The system provided an 85 percent target replacement rate. However, while Croatia has reduced pension benefits relative to wages (to 62% in 1993) using less generous pension adjustments for inflation to wages, FYR Macedonia allowed pension benefits to rise relative to wages (to 82% in 1993),° creating financial problems for the pension fund. 90% . _ __ ___ I C80 % __ _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __ l 70% _ 60% * * _ 50% _____ 40% 30% 20%- 10%. 0% Macedonia Poland Croatia Hungary Slovak Czech Albania Bulgaria Romania Figure 25 Replacement Rates (Net Wages), 1993 10 FYR Macedonia did decrease benefits in 1994-95. -21- 14% - 12% E _ _ __ _ o OPension 10% - _ - Expenditure Figure 26 P nTotal Expenditure 6% 4% 2% 0% Macedonia Bulgania Poland Hungary Slovak Czech Albania Romania Croatia Figure 26 Pension Benefit Expenditures and Total Pension Expenditures as a Percentage of GDP, 1993 (iii) High Non-Benefit Expenditures. One way to reduce the total expenditures of pension funds is to cut non-benefit expenditures.'1 Total expenditures include payments of pension benefits as well as administrative costs, supplementary benefits, aids to the handicapped, spa benefits, etc. A comparison of pension deficits across countries based on total pension fund expenditures and pension benefit expenditures indicates the level of expenditures on these 'other benefits' across countries. Figure 26 contrasts pension fund deficits based on total expenditures and pension expenditures for the same year. The difference between pension fund deficits according to the two measures is the highest for FYR Macedonia (nearly 5% of GDP)12 and sizable in Bulgaria and Hungary indicating that these countries spend a great deal on other items than simply paying out pension deficits.'3 In summary, the large share of pensions in government expenditures in the region indicate a critical need to reform pension systems. Pension reform is all the more critical in countries where pension systems have the most adverse impact on fiscal discipline. Reform efforts to contain pension costs will have to address the specific determinants that push up pension expenditures in each country. Thus, pension reform efforts will have to address: (i) aging populations in Hungary and Bulgaria, (ii) favorable pension eligibility conditions in Albania, Hungary, Bulgaria, Poland and Slovakia, (iii) too many new pensioners in Albania and FYR Macedonia, (iv) a large number of disability pensions in Poland, Romania and Hungary, (v) generous pension benefits in Poland, FYR Macedonia and Croatia, and (vi) high non-benefit expenditures in Bulgaria, Hungary and FYR Macedonia. I Note that total expenditures contain benefit expenditures such as payment to anny, police and veterans and would continue to be a part of govemment expenditures. 12 FYR Macedonia eliminated many of these benefits in 1995. 13The deficit measure for Czech Republic and Slovakia appears to have changed the least, but this is mainly because we do not have administrative or other expenditures for these countries -22- 4. Conclusions The economic restructuring in Eastern and Central Europe increased unemployment and forced firms to pension-off redundant workers, raising dependency rates in all countries across the region. The paper shows that system dependency rates increased primarily because of a decline in the number of contributors relative to the working age population. The increase in pensioners has been a more important determinant in raising system dependency rates in FYR Macedonia, Poland and Romania. Not surprisingly, the aging of the population did not increase pension costs in the short time span under study. Contrary to common perception, not all countries realized and increase in pension costs. Most countries reduced the generosity of their pension systems to offset the growth in system dependency rates on pension costs. The divergent trends pension expenditures relative to GDP realized in the region reflect the varying degrees of success with which governments were able to curtail growth in pension fund generosity to offset the growth in system dependency rates. Thus, Albania and Croatia realized declines in pension expenditures14 relative to GDP because both countries reduced the generosity of the pension system to more than offset the impact of increased system dependency rates on pension costs. Romania and the Czech Republic had a slow growth in pension costs over the transition because declines in system generosity mitigated the impact of the rise in system dependency rates. However, in Bulgaria and Slovakia the reduced generosity of the pension system did not offset increases in system dependency rates, so that pension costs increased sharply in these countries. In Poland and FYR Macedonia, the government increased replacement rates compounding the increase in pension costs. In 1993, all countries (with the exception of Romania) realized deficits in the pension fund. In large part, these deficits contributed to overall government budget deficits and reduced fiscal discipline and macroeconomic stability in the region. Reducing pension deficits will therefore be a critical component of a strategy for promoting sound economic policies and stimulating economic growth. Three policy conclusions emerge from this study. First, reducing the level and rate of growth of pension costs will be the main instrument for reducing pension deficits. We find that tax compliance is a major problem in all countries in the region and improving tax compliance will reduce pension deficits. However, even with full compliance, high tax rates prevailing in the region would remain, distorting labor markets and dampening the prospects for economic growth. Our second conclusion is that reducing the sharp growth in pension costs in FYR Macedonia, Poland, Slovakia and Bulgaria will require different approaches. Containing the growth of pension benefit expenditures in Poland requires a sharp reduction in pension fund generosity; Poland has already controlled the growth rate in new old age pensioners. An increase in the number of contributors should help contain growth in pension costs in Bulgaria and Slovakia. Replacement rates have declined in both countries and an increase in pensioners has been negligible. In FYR Macedonia, growth in pension costs can be reduced if the number of new pensioners entering the system is sharply restricted and the growth in the replacement rate is curtailed. 15 Our third conclusion is that strategies to reduce pension costs cannot be generalized. Instead, pension reform should focus on the factors putting upward pressure on pension costs in each country. Policies to reduce pension costs need to primarily focus on reducing pension benefits in FYR Macedonia, Poland and Croatia; restricting eligibility to the pension system in Albania, Bulgaria, Slovakia, Poland and Romania, monitoring disability benefits in Romania and Poland and curtailing the entry of new pensioners in Albania, Croatia and FYR Macedonia. A reduction of non-pension benefit related expenditures would help reduce pension deficits in Bulgaria, Hungary and FYR Macedonia. 4 In all cases financing variable paths are compared to GDP I5 The country has taken steps to control these recently and pension costs have declined. -23- Statistical Annex Annex 1: Data This paper uses primary data collected from pension institutes in ten Central and Eastern European countries: Albania, Bulgaria, Croatia, the Czech Republic, Hungary, FYR Macedonia, Poland, Romania, Slovakia and Slovenia. The collection of pension data was part of a larger SCT (Social Challenges of Transition) exercise to provide data on key indicators for all the social sectors over the transition. The pension data cover developments in the primary pension fund of each country, that is, from the fund that is the largest in terms of coverage and fiscal resources from 1989-93. In many cases, data on secondary pension systems are not readily available. For most countries, the exclusion of secondary systems is not material, as they are extremely small. However, Poland and FYR Macedonia both have large secondary systems and their exclusion should be borne in mind when comparing results for these countries. Pension data were collected on four broad topics for the years 1989-93: (i) The number of contributors to the pension fund by sex and occupation (wage earners, self-employed and farmers), (ii) the number of existing and new beneficiaries by type of pension (old age, disability, survivor), (iii) average, legal minimum and legal maximum pensions by type of pension and sex of recipient and (iv) revenues and expenditures of the pension fund. These data, coupled with the demographic and macroeconomic data, allowed the construction of key indicators demonstrating the financial status of the pension funds presented in this paper. The availability of the data varies considerably by country. The most complete data were provided by Bulgaria, FYR Macedonia, Croatia, Hungary, the Czech Republic and Slovakia while Albania, Romania and Slovenia provided the least data. In Albania, the lack of computerization of records meant that data would have to be compiled from individual folders. In Slovenia, the paucity of data reflects a lack of coordination between the Pension Institute and the Macroeconomic Institute. We were unable to construct indicators consistent with other countries for Slovenia so it has been omitted from the analysis in this paper. In contrast to the data from the other ex-Yugoslav Republics, data from Croatia and FYR Macedonia are quite complete and appear reliable. In Romania, the Pension Institute reported that certain detailed information requested was not computerized and, therefore, not easily accessible. Despite the substantial amount of data available in other areas, Poland could not provide age distribution of pensioners by type and gender. The only source for these data for Poland is a sample survey of pensioners conducted in 1992. The availability of primary data also varied by category of pension data required. The level of average pensions and the distribution of pensioners by sex and type of pension were by far the most commonly available data. Data on the number of contributors disaggregated by sex and occupation were not available and data on the legal limits of minimum and maximum pensions by country over time were difficult to obtain. Revenue and expenditure data were available for most countries but not always for all the years under study. Pension revenues were not available in Hungary, the Czech Republic and Slovakia before 1992 because their pension funds were financed directly from general revenues. Trends in payroll tax rates and gross wages -- critical inputs to an analysis of revenue trends -- were not available for Albania and Hungary. In addition, data on net wages -- necessary for comparison with pension benefits in order to determine the generosity of the pension fund -- were not available for the Czech Republic, Slovakia and Bulgaria. The years for which data are available also vary across countries. In large part, pension data were available for most countries from 1990 to 1993. Therefore, the paper evaluates pension financing trends over these four years. Despite these limitations, the data allow us to portray trends in pension financing for the majority of the countries in the region over the first few years of the transition. -24- Annex 2: The Model We evaluate pension fund financing by analyzing the levels and rates of change in pension revenues, expenditures and deficits relative to economic output. Pay-as-you-go pension systems finance benefits to pensioners by imposing a payroll tax on the wages of current employees. Thus, differences in economic and demographic factors and government policy across countries (or changes in these factors over time) that affect the proportion of contributors to beneficiaries or alter the level of benefits to worker earnings have the potential to determine the level and rates of change in pension expenditures relative to pension revenues and affect the system's financial viability. Pension Expenditures The level and rate of change in pension expenditures relative to economic output is a key indicator of financing problems in pension funds. Countries with high (or increasing) expenditures are supporting (or increasingly supporting) pension systems that are too expensive relative to their income. Even if these countries can raise enough revenues to finance pensions through payroll taxes on current workers so that there are no deficits in pension funds (defined as payroll tax revenues less total pension expenditures), these tax rates are bound to be high and will impose considerable costs on current workers and the economy. Expenditures on pension benefits comprise the largest share of total pension expenditures. The remainder of pension outlays are spent on the administrative cost of running the pension fund and on numerous small benefits such as spa care for the elderly and special care for the handicapped. Changes in total pension expenditure over time can therefore be approximated fairly closely by changes in expenditures on pension benefits. The ratio of pension (benefit) expenditures relative to GDP can be expressed as follows: Pension Expenditures = Pensioners * (Pension Earenditures/Pensioneri (I) GDP Contributors GDP/Contributor If we define the ratio of pensioners to contributors as the system dependency rate and define the ratio of average pensions to GDP/contributor as the transfer ratio , then equation (1) can be written as: = System Dependency Rate * Transfer Ratio (Ia) The transfer ratio can be further decomposed (from equation 1) as follows: Transfer Ratio = Replacement Rate* Wage Share (2) where the replacement rate is the ratio of average pensions to average wage, and the wage share is the ratio of average wage to GDP/contributor. Equation 2 shows that pension costs are positively related to system dependency rates and the transfer ratio. The system dependencv rate measures the number of beneficiaries relative to the number of contributors in the pension system. Pension costs will be higher (or increasing) in countries that have a higher (or increasing) proportion of pensioners relative to contributors. The system dependency rate can vary across countries as a result of differences in pension eligibility conditions, the demographic burden and economic conditions that affect the level of employment in the economy. Specifically, the system dependency rate can be decomposed as follows: 16 Note that this is not GDP/employed (as in Holzman) but GDP/contributor. An analysis of employed and contributor trends shows that both trends are virtually identical for most of the countries in the study. -25- System Dependency Rate = Pensioners * Pp 60+ *Total Pop * Pop 15-59 (3a) Total Pop Pop 15-59 Pop 60+ Contributors = Pensioners * Pop 15-59 (3bJ Pop 60+ Pop 15-59 Contributors Equation 3a shows that a higher proportion of pensioners in total population -- generally, the result of favorable early retirement policies and low retirement ages -- will result in higher system dependency rates and higher pension costs. Similarly, countries with demographic profiles that feature a higher proportion of aged relative to the total population or a higher ratio of pensioners to persons of working age will have a greater proportion of pensioners relative to contributors. Finally, system dependency rates and pension costs will be higher in countries with a lower ratio of working age population to contributors. This could occur in countries where high tax rates force workers to join the informal economy, where employment opportunities are more limited and where a larger number of workers have been pensioned off. The relative importance of the informal economy can be estimated by comparing the ratio of contributors and employed to working age population. Countries with higher employed in working age population relative to contributors are those where the informal economy, rather than high rate of unemployment, is the reason behind a small number of contributors to the pension system. Equation 3b can also be used to evaluate differences in the levels and rates of change in system dependency rates. The only difference between equation 3a and 3b is that in 3b, changes in eligibility are captured by changes in pensioners to total population over 60 years of age. Each ratio has its strengths and weaknesses in estimating differences in the eligibility levels of pension funds. The advantage of using pensioners to total population over 60 years of age to capture coverage of the pension system is that it is the most commonly used in the literature. However, in most countries the proportion of pensioners in total population over 60 years of age is greater than one because individuals less than 60 years of age (e.g. women, disability and survivor pension recipients) also receive pensions. The ratio of pensioners in total population used as an eligibility indicator in equation 3b also does not strictly reflect pension eligibility because it confounds demographic factors. For the same number of pensioners and population over 60, this ratio will be lower in countries with a higher population under 60 years of age. The ratio can be used in conjunction with the ratio of persons over 60 in total population to obtain the eligibility ratio used in equation in 3b. The tran4fer ratio measures the generosity of the pension system. It reflects the transfer of income from contributors to pensioners occasioned by the pension system. The greater the ratio of pensions to income, the more generous the pension system relative to the economic output of the country. Equation 2 identifies the two components of the transfer ratio as the replacement rate and the wage share. The replacement rate (ratio of average pensions to average net wages) is also an indicator of pension fund generosity, but it is a somewhat narrower concept, one that compares changes in pension benefits only to wages and not the income of workers. However, individual pensions are financed out of a tax on wage earnings. Therefore, the replacement rate can be used to indicate the cost to current workers of financing pension benefits to the retired. While variations in the level of replacement rates across the region measure the relative generosity of the pension system, changes in the replacement rate over time can be used to assess government policies to increase or decrease the generosity of the pension benefits. Specifically, increases (decreases) in replacement rates reflect government policies to raise (decrease) pension benefit relative to wages. In essence, the replacement rate is a somewhat narrower definition of pension system generosity. -26- The second component of the transfer ratio is the wage share. As noted above, the wage share is simply the ratio of average wages to average income, where average income is defined as GDP per contributor. Equation 2a shows that the impact of changes in replacement rates on pension expenditures can be mitigated by movements in the wage share. For example, government policies to contain pension costs by reducing replacement rates can be frustrated if wages grow relatively more than income. In this case, while pensions will decline relative to wages, they may actually increase relative to income forcing up pension costs relative to GDP. Similarly, countries that increase replacement rates may not realize large increases in pension expenditures relative to GDP if wage growth is less than income growth. In this case, benefits will increase relative to wages but decline relative to income, forcing down pension costs relative to income. To a large extent, governments can control the wage share by incomes policies that constrain the growth of wages relative to income. But the wage share is also influenced by labor and capital market developments that alter the share of labor income in economic output. Tables 1, 2 and 3 (pages 28-30) decompose the changes in pension expenditures relative to GDP. Table I is based on equation 3b, Table 2 on equation 2 and Table 3 evaluates changes in pension expenditures relative to GDP based on equation I a. Pension Revenues The pension fund obtains revenue through payroll tax rates levied on current workers. Thus, pension fund revenues relative to GDP can be written as: Pension Revenues = Potential Revenues (Actual Revenues/Potential Revenues) (4) GDP GDP where, Potential Revenues = Payroll Tax Rate *Gross Wage *Employed (5) and, Actual Revenues = Tax Compliance Rate (6) Potential Revenues where the tax compliance rate reflects the ability of governments to collect revenues from employees and employers. Tax compliance will be lower (or will decline) if reported wages of contributors diverse from actual wages and if not all employed contribute to the system. Therefore, Pension Revenues = (Tax Compliance Rate* Payroll Tax Rate) * (Gross Wage*Emyloyed) (7) GDP GDP GDP = Effective Tax Rate * Wage Bill/GDP (8) = Effective Tax Rate * Tax Base (9) where, Effective Tax Rate = Tax Compliance Rate *Payroll Tax Rate (JO) All else equal, revenues will decline if there is a fall in the tax base as a decline in the number of employed, or as a result of government incomes policies that reduce the growth of wages relative to GDP. Labor market conditions that reduce the share of wages in economic output or increased under- reporting of wages to avoid high taxation will also tend to reduce the tax base. Revenues will also fall if there is a decline in the effective tax rate. Since payroll tax rates have increased or remained the same in all countries, declines in effective tax rates are likely to reflect a reduction in tax compliance. -27- Pension Deficits The difference between revenues and total pension expenditures is defined as the pension fund deficit. Pension fund deficits are usually financed by the government through budgetary transfers. Pension fund deficits can contribute to overall budget deficits if governments cannot raise enough tax revenues to cover the pension fund deficit and other budgetary obligations. If we define government revenues, R, as the sum of payroll tax revenues for pensions and E, expenditures, then Dt, the total government budget deficit, can be written as: Dt=R-E (II) where E, total government expenditures, can be written as E=Eo+Ep, where Ep is expenditures on pensions and Eo is all other expenditures. Total government revenues, R, can be similarly expressed as R= Rp + Ro. The pension fund deficit can be written as: Dp=Rp-Ep (12) The total government deficit is: Dt=Ro-Eo-Dp (13) Thus, if Ro Eo+Dp, indicating that governments can raise enough revenues to finance pension deficits as well as other expenditures. An alternate definition of pension fund deficit is the difference between expenditures on pension benefits and payroll tax revenues. These two measures of the deficit allow us to compare the level of 'non-benefit' expenditure incurred by pension funds. Large non-benefit expenditures in pension funds indicate the use of payroll tax revenues to finance other benefits, such as grants to orphans, etc., that do not represent wage replacement. -28- TABLE I (1) (2) (3) (4) (5) | | Pensioner/ | Pensioner/ Pop. 60+/ Pop. 15-59/ Sum of | | Contributor Pop. 60+ Pop. 15-59 Contributor (2),(3),(4) Albania 1993 (1) 1.06 1.73 0.13 4.80 1990 (2) 0.38 1.35 0.12 2.30 % change (3) 1.80 0.29 0.04 1.09 1.42 % contribution (4) 1.00 0.16 0.02 0.61 0.79 Bulgaria 1993 (1) 0.80 1.39 0.35 1.66 1990 (2) 0.55 1.34 0.33 1.23 % change (3) 0.46 0.04 0.06 0.34 0.44 % contribution (4) 1.00 0.08 0.13 0.74 0.95 Croatia 1993 (1) 0.43 0.97 0.25 1.76 1990 (2) 0.30 0.74 0.28 1.49 % change (3) 0.43 0.30 -0.09 0.18 0.40 % contribution (4) 1.00 0.71 -0.21 0.43 0.93 Czech 1993 (1) 0.51 1.30 0.29 1.37 Republic 1990 (2) 0.42 1.27 0.29 1.14 % change (3) 0.22 0.02 -0.01 0.20 0.22 % contribution (4) 1.00 0.11 -0.03 0.91 0.99 Hungary 1993 (1) 0.66 1.44 0.31 1.47 1992 (2) 0.58 1.41 0.31 1.33 % change (3) 0.13 0.03 -0.001 0.10 0.13 % contribution (4) 1.00 0.19 -0.01 0.80 0.98 FYR Macedonia 1993 (1) 0.28 0.82 0.18 1.83 1990 (2) 0.22 0.73 0.17 1.82 % change (3) 0.24 0.13 0.09 0.01 0.23 % contribution (4) 1.00 0.53 0.38 0.04 0.94 Poland 1993 (1) 0.53 1.36 0.22 1.79 1990 (2) 0.40 1.21 0.21 1.59 % change (3) 0.33 0.13 0.04 0.13 0.30 % contribution (4) 1.00 0.40 0.13 0.38 0.91 Romania 1993 (1) 0.49 0.85 0.28 2.09 1990 (2) 0.34 0.73 0.26 1.78 % change (3) 0.45 0.16 0.07 0.17 0.40 % contribution (4) 1.00 0.35 0.15 0.38 0.89 Slovakia 1993 (1) 0.53 1.36 0.24 1.64 1990 (2) 0.45 1.32 0.25 1.37 % change (3) 0.19 0.03 -0.05 0.20 0.18 % contribution (4) 1.00 0.17 -0.27 1.06 0.96 Notes: *column (5) should equal corresponding row entry in column (1) *row (3) = [(row I - row 2)/row 21 * 100 *row (4) = row 3 for columns 2-4 divided by row 3, column I -29- TABLE 2 (1) (2) (3) (4) Pensions/ Replacement Wage Sum of Income Rates Share (2),(3) Albania 1993 (1) 0.06 0.42 0.14 1990 (2) 0.18 0.53 0.34 % change (3) -0.68 -0.21 -0.59 -0.80 % contribution (4) 1.00 0.31 0.87 1.18 Bulgaria 1993 (1) 0.13 0.40 0.33 1990 (2) 0.14 0.43 0.33 % change (3) -0.06 -0.07 0.00 -0.07 % contribution (4) 1.00 1.17 0.00 1.17 Croatia 1993 (1) 0.13 0.62 0.21 1990 (2) 0.30 0.73 0.42 % change (3) -0.58 -0.15 -0.51 -0.66 % contribution (4) 1.00 0.26 0.87 1.13 Czech 1993 (1) 0.16 0.43 0.36 Republic 1990 (2) 0.18 0.48 0.38 % change (3) -0.14 -0.10 -0.06 -0.17 % contribution (4) 1.00 0.73 0.44 1.16 Hungary 1993 (1) 0.13 0.47 0.27 1992 (2) 0.15 0.48 0.31 % change (3) -0.14 -0.02 -0.12 -0.14 % contribution (4) 1.00 0.12 0.90 1.02 FYR Macedonia 1993 (1) 0.47 0.82 0.57 1990 (2) 0.39 0.79 0.49 % change (3) 0.22 0.05 0.17 0.22 % contribution (4) 1.00 0.21 0.75 0.96 Poland 1993 (1) 0.24 0.77 0.31 1990 (2) 0.18 0.59 0.31 % change (3) 0.31 0.30 0.00 0.30 % contribution (4) 1.00 0.98 -0.02 0.97 Romania 1993 (1) 0.11 0.26 0.43 1990 (2) 0.16 0.42 0.37 % change (3) -0.29 -0.38 0.15 -0.23 % contribution (4) 1.00 1.33 -0.52 0.80 Slovakia 1993 (1) 0.18 0.46 0.38 1990 (2) 0.19 0.48 0.39 % change (3) -0.06 -0.04 -0.03 -0.07 % contribution (4) 1.00 0.65 0.47 1.12- Notes: *column (4) should equal corresponding row entry in column (1) *row (3) = [(row 1 - row 2)/row 21 * 100 *row (4) = row 3 for columns 2-4 divided by row 3, column I *Bulgaria, Czech Republic, Slovak Republic: (Gross Wages) -30- TABLE 3 (1) (2) (3) Pension Exp./ Pensioners/ Pensions/ GDP Contributors Income Albania 1993 0.062 1.06 0.06 1990 0.068 0.38 0.18 % change -0.096 1.80 -0.68 Bulgaria 1993 0.108 0.80 0.13 1990 0.078 0.55 0.14 % change 0.381 0.46 -0.06 Croatia 1993 0.055 0.43 0.13 1990 0.093 0.31 0.30 % change -0.415 0.40 -0.58 Czech 1993 0.080 0.51 0.16 Republic 1990 0.076 0.42 0.18 % change 0.044 0.22 -0.14 Hungary 1993 0.085 0.66 0.13 1992 0.087 0.58 0.15 % change -0.022 0.13 -0.14 FYR Macedonia 1993 0.131 0.28 0.47 1990 0.086 0.22 0.39 % change 0.523 0.24 0.22 Poland 1993 0.127 0.53 0.24 1990 0.073 0.40 0.18 % change 0.737 0.33 0.31 Romania 1993 0.055 0.49 0.11 1990 0.053 0.34 0.16 % change 0.036 0.45 -0.29 Slovakia 1993 0.094 0.53 0.18 1990 0.074 0.39 0.19 % change 0.268 0.36 -0.07 -31- Annex 3: Data for Figures FIGURE 1: Total Pension Expenditures as a Percentage of GDP, 1990-1993 _ 1990 1991 1992 1993 Albania 8.56% 12.03% 6.96% 6.46% Bulgaria 10.92% 12.41`0 12.18% 14.09% Croatia 13.60% 10.62% 6.04% 6.20% Czech R. 7.66% 7.500o 7.93% 7.98% Hungary #N/A #N/A 11.12% 10.63% FYR Macedonia 10.41% 11.43% 11.44% 15.62% Poland 7.64% 11.50% 13.59% 13.43% Romania 6.94% 7.04% 6.76% 6.24% Slovakia 7.41% 8.1 1 %0 9.27% 9.38% FIGURE 2: Trends in Total Pension Expenditures as a Percentage of GDP Relative to 1990 1990 1991 1992 1993 Albania 1.00 1.40 0.81 0.75 Bulgaria 1.00 1.14 1.12 1.29 Croatia 1.00 0.78 0.44 0.46 Czech R. 1.00 0.98 1.04 1.04 Hungary #N/A #N/A 1.00 0.96 FYR Macedonia 1.00 1.10 1.10 1.50 Poland 1.00 1.51 1.78 1.76 Romania 1.00 1.01 0.97 0.90 Slovakia 1.00 1Il9 1.25 1.27 FIGURE 3: Pension Expenditures as a Percentage of Government Expenditure 1989 1990 1991 1992 1993 Albania #N/A 10.99% 16.14% 14.00% 15.32% Bulgaria 12.49% 12.67% 19.18% 21.28% 20.94% Croatia #N/A #N/A 4.71% 3.20% 3.30% Czech R. #N/A #N/A #N/A 16.50% 16.42% Hungary #N/A #N'A #N/A 14.19% 14.93% FYR Macedonia #N/A #N/A 18.76% 19.27% 24.80% Poland 14.61% 16.75% 22.31% 24.77% 25.40% Romania 10.80% 13.72% 14.11% 12.04% 15.55% Slovakia #N/A #N/A #N/A 14.53% 16.86% -32- FIGURE 4: Trends in Contributors per Pensioner Relative to 1990 1990 1991 1992 1993 Albania 1.00 0.81 0.61 0.36 Bulgaria 1.00 0.85 0.71 0.69 Croatia 1.00 #N/A 0.75 0.71 Czech R. 1.00 0.91 0.84 0.82 Hungary 1.00 0.94 0.81 0.72 FYR Macedonia 1.00 0.92 0.87 0.80 Poland 1.00 0.88 0.81 0.75 Romania 1.00 0.89 0.78 0.69 Slovakia 1.00 0.88 0.78 0.74 NOTE: Hungary, FYR Macedonia numbers: Employed to Pensioners; Albania is net of fanners FIGURE 5a: Trends in Number of Pensioners to Population Age 60 and Over Relative to 1990 1990 1991 1992 1993 Albania 1.00 1.12 1.22 1.29 Bulgaria 1.00 1.04 1.05 1.04 Croatia 1.00 #N/A 1.11 1.31 Czech R. 1.00 1.01 1.02 1.02 Hungary 1.00 1.03 1.07 1.10 FYR Macedonia 1.00 1.05 1.09 1.13 Poland 1.00 1.08 1.11 1.13 Romania 1.00 1.06 1.08 1.16 Slovakia 1.00 1.04 1.06 1.07 FIGURE 5b: Trends in Number of New Pensioners Relative to 1989 I 1989 1990 1991 1992 1993 Albania 1.00 1.86 3.28 4.00 4.07 Bulgaria 1.00 1.09 1.44 1.16 0.82 Croatia 1.00 2.08 #N/A 1.88 1.72 Czech R. 1.00 0.89 1.03 0.94 0.86 Hungary #N/A #N/A #N/A #N/A #N/A FYR Macedonia 1.00 1.47 1.12 1.59 1.74 Poland 1.00 1.49 2.09 1.15 0.95 Romania #N/A #N/A #N/A #N/A #N/A Slovakia 1.00 0.98 1.21 1.11 0.93 -33- FIGURE 6: Trends in the Ratio of Contributors to Working Age Population Relative to 1990 1990 1991 1992 1993 Albania 1.00 0.91 0.75 0.48 Bulgaria 1.00 0.89 0.77 0.75 Croatia 1.00 0.94 0.86 0.85 Czech R. 1.00 0.92 0.86 0.83 Hungary 1.00 0.96 0.87 0.79 FYR Macedonia 1.00 1.00 0.99 0.99 Poland 1.00 0.96 0.93 0.89 Romania 1.00 0.96 0.89 0.85 Slovakia 1.00 0.91 0.82 0.76 FIGURE 7: Trends in the Ratio of Population Age 60 and Over to Working Age Population Relative to 1990 1990 1991 1992 1993 Albania 1.00 1.00 1.02 1.04 Bulgaria 1.00 1.01 1.04 1.06 Croatia 1.00 1.03 1.03 0.91 Czech R. 1.00 1.00 1.00 0.99 Hungary 1.00 1.00 1.00 1.00 FYR Macedonia 1.00 1.03 1.05 1.09 Poland 1.00 1.02 1.03 1.04 Romania 1.00 1.02 1.05 1.07 Slovakia 1.00 0.99 0.99 0.96 FIGURE 8a, b: Replacement Rates (Net Wages) I 1990 1991 1992 1993 Albania 52.60% 51.62% 36.33% 41.51% Bulgaria 42.81% 49.72% 38.68% 40.17% Croatia 72.99% 62.77% 56.61% 62.02% Czech R. 47.57% 49.57% 47.40% 43.35% Hungary #N/A #N/A 48.22% 47.28% FYR Macedonia 78.54% 67.29% 75.76% 82.27% Poland 59.03% 67.84% 77.45% 76.75% Romania 41.90% 46.25% 27.20% 25.94% Slovakia 48.25% 50.84% 48.46% 46.35% FIGURE 9a, b: Trends in Wage Share/Tax Base Relative to 1990 1990 1991 1992 1993 Albania 1.00 1.21 0.85 0.41 Bulgaria 1.00 0.81 0.92 1.00 Croatia 1.00 0.96 0.56 0.49 Czech R. 1.00 0.85 0.88 0.94 Hungary 1.00 1.08 0.96 0.85 FYR Macedonia 1.00 1.20 1.00 1.17 Poland 1.00 1.14 1.08 1.01 Romania 1.00 0.82 1.21 1.15 Slovakia 1.00 0.91 0.97 0.97 -34- FIGURE 10: Ratio of Pension Revenue as a Percentage of GDP, 1990-1993 I 1990 1991 1992 1993 Albania 5.75% 8.74% 3.64% 2.74% Bulgaria 10.95% 9.33% 10.89% 10.78% Croatia 12.21% 9.11% 4.82% 4.72% Czech R. #N/A #N/A 6.16% 6.63% Hungary #N/A #N/A 9.07% 8.75% FYR Macedonia 8.98% 9.41% 8.79% 9.87% Poland 9.10% 11.06% 12.05% 11.82% Romania 6.20% 7.37% 7.81% 6.85% Slovakia #N/A #N/A 5.30% 6.66% FIGURE 11 & 12: Pension Deficit as a Percentage of GDP (Total Pension Expenditures) _ 1990 1991 1992 1993 Albania -2.81% -3.28% -3.32% -3.72% Bulgaria 0.03% -3.08% -1.29% -3.30% Croatia -1.39% -1.51% -1.22% -1.48% Czech R. #N/A #N/A -1.77% -1.35% Hungary #N/A #N/A -2.05% -1.88% FYR Macedonia -1.43% -2.02% -2.65% -5.75% Poland 1.46% -0.44% -1.54% -1.61% Romania -0.75% 0.34% 1.05% 0.62% Slovakia #N/A #N/A -3.97% -2.71% FIGURE 12:Government Deficit as a Percentage of GDP _ 1990 1991 1992 1993 Albania 1.54% 3.34% 2.19% 1.47% Bulgaria 9.15% 3.29% 5.12% 14.04% Croatia #N/A 42.76% 1.54% 4.33% Czech R. #N/A #N/A 4.98% -5.52% Hungary -0.46% 2.28% 5.49% 6.04% FYR Macedonia #N/A 4.42% 9.87% 13.15% Poland -3.41% 6.67% 6.64% 2.30% Romania -1.28% -0.45% 6.69% 0.08% Slovakia #N/A #N/A 13.09% 7.04% FIGURE 13: Total Pension Expenditures as a Percentage of GDP I 1990 1991 1992 1993 Albania 8.56% 12.03% 6.96% 6.46% Bulgaria 10.92% 12.41% 12.18% 14.09% Croatia 13.60% 10.62% 6.04% 6.20% Czech R. 7.66% 7.50% 7.93% 7.98% Hungary #N/A #N/A 11.12% 10.63% FYR Macedonia 10.41% 11.43% 11.44% 15.62% Poland 7.64% 11.50% 13.59% 13.43% Romania 6.94% 7.04% 6.76% 6.24% Slovakia 7.41% 8.11% 9.27% 9.38% -35- FIGURE 14: Pension Deficit as a Percentage of GDP (Total Pension Expenditures) I 1990 1991 1992 1993 Albania -2.81% -3.28% -3.32% -3.72% Bulgaria 0.03% -3.08% -1.29% -3.30% Croatia -I.39% -1.51% -1.22% -1.48% Czech R. #N/A #N/A -1.77% -1.35% Hungary #N/A #N/A -2.05% -1.88% FYR Macedonia -1.43% -2.02% -2.65% -5.75% Poland 1.46% -0.44% -1.54% -1.61% Romania -0.75% 0.34% 1.05% 0.62% Slovakia #N/A #N/A -3.97% -2.7 1% FIGURE 15: Payroll Tax Rates I 1990 1991 1992 1993 Albania #N/A #N/A #N/A 31.7% Bulgaria 30.0% 35.0% 35.0% 35.0% Croatia 16.7% 18.4% 22.0% 22.0% Czech R. 27.2% 27.2% 27.2% 27.2% Hungary 31.7% 31.7% 31.7% 31.7% FYR Macedonia 16.9% 16.6% 18.0% 18.0% Poland 43.0% 43.0% 45.0% 37.0% Romania 14.5% 20.0% 20.0% 28.0% Slovakia 26.5% 26.5% 26.5% 26.5% FIGURE 15: Effective Tax Rates ____ ____ ___ 1993 Albania 19.5% Bulgaria 27.0% Croatia 12.4% Czech R. 18.6% Hungary 21.6% FYR Macedonia 17.2% Poland 30.2% Romania 21.1% Slovakia 17.1% FIGURE 16: Actual Deficit as a Percentage of GDP (Total Pension Expenditures) 1990 1991 1992 1993 A]bania -2.81% -3.28% -3.32% -3.72% Bulgaria 0.03% -3.08% -1.29% -3.30% Croatia -1.39% -1.51% -1.22% -1.48% Czech R. #N/A #N/A -1.77% -1.35% Hungary #N/A #N/A -2.05% -1.88% FYR Macedonia -1.43% -2.02% -2.65% -5.75% Poland 1.46% -0.44% -1.54% -1.61% Romania -0.75% 0.34% 1.05% 0.62% Slovakia #N/A #N/A -3.97% -2.7 1% -36- FIGURE 16: Potential Deficit as a Percentage of GDP (Total Pension Expenditures) 1993 Albania -2.00% Bulgaria -0.10% Croatia 2.19% Czech R. 1.71% Hungary 2.22% FYR Macedonia -5.29% Poland 0.74% Romania 2.86% Slovakia 0.95% FIGURE 17: Pensioners as a Percentage of Contributors (System Dependency Rates) 1990 1991 1992 1993 Albania 0.22 0.27 0.35 0.97 Bulgaria 0.55 0.65 0.78 0.80 Croatia 0.31 #N/A 0.40 0.43 Czech R. 0.42 0.46 0.50 0.51 Hungary 0.47 0.50 0.58 0.66 FYR Macedonia 0.22 0.24 0.26 0.28 Poland 0.40 0.45 0.49 0.53 Romania 0.34 0.38 0.43 0.49 Slovakia 0.39 0.45 0.50 0.53 FIGURE 18: Percentage of Persons Age 60 and Over in Total Population I 1990 1991 1992 1993 Albania 7.28% 7.33% 7.45% 7.65% Bulgaria 19.82% 20.17% 20.69% 21.03% Croatia 17.38% 17.81% 17.82% 15.86% Czech R. 17.70% 17.90% 17.98% 17.98% Hungary 19.07% 19.21% 19.25% 19.31% FYR Macedonia 10.72% 11.00% 11.26% 11.68% Poland 12.19% 12.42% 12.61% 12.80% Romania 15.65% 16.00% 16.41% 16.89% Slovakia 14.90% 14.92% 15.00% 14.75% -37- FIGURE 19: Ratio of Working Age Population to Persons Age 60 and Over l 1990 1991 1992 1993 Albania 8.19 8.18 8.06 7.87 Bulgaria 3.02 2.97 2.90 2.86 Croatia 3.61 3.51 3.51 3.98 Czech R. 3.45 3.44 3.45 3.48 Hungary 3.20 3.21 3.21 3.21 FYR Macedonia 5.93 5.78 5.64 5.44 Poland 4.84 4.76 4.69 4.64 Romania 3.88 3.81 3.71 3.63 Slovakia 4.03 4.06 4.06 4.21 FIGURE 20: Percentage of Population Receiving Pensions from Primary Pension System 1990 1991 1992 1993 Albania 9.81% 11.04% 12.20% 13.25% Bulgaria 26.55% 28.00% 29.15% 29.21% Croatia 12.84% #N/A 14.66% 15.33% Czech R. 22.49% 23.02% 23.29% 23.40% Hungary 24.98% 25.95% 27.10% 27.88% FYR Macedonia 7.78% 8.39% 8.88% 9.57% Poland 14.72% 16.15% 16.96% 17.43% Romania 11.51% 12.44% 13.01% 14.39% Slovakia 18.90% 19.66% 20.21% 20.07% FIGURE 21: Percentage of Old-Age Pensioners in First Year of Retirement Under Age 60 (Primary System) to Total Number of Pensioners 1990 1991 1992 1993 Albania #N/A #N/A #N/A #N/A Bulgaria 73.77% 79.25% 68.70% 62.70% Croatia 64.46% #N/A 86.56% 90.87% Czech R. 51.87% 62.83% 67.76% 67.54% Hungary #N/A #N/A #N/A #N/A FYR Macedonia 37.02% 46.59% 36.89% 44.46% Poland #N/A #N/A 67.80% 67.30% Romania #N/A #N/A #N/A #N/A Slovakia 52.23% 68.52% 74.46% 76.40% -38- FIGURE 22: Percentage Distribution of Pensioners (Primary System) by Type of Pension I1990 1991 1992 1993 Albania Old-Age 75.17% 77.11% 79.06% 79.93% o Disability 8.36% 7.05% 6.45% 5.93%sio Survivors' 16.47% 15.85% 14.49% 14.14% Bulgaria Old-Age 83.01% 84.67% 85.56% 85.83%0 Disability 8.34% 8.02% 7.72% 7.620%0 Survivors' 6.32% 5.05% 4.46% 4.29096 Croatia Old-Age 46.07% #N/A 51.99% 52.36° Disability 26.64% #N/A 22.720o0 22.1804 Survivors' 27.29% #N/A 25.30% 25.16% Czech R. Old-Age 54.50% 54.90% 55.07% 54.94% Disability 18.42% 18.53% 18.74% 19.11% Survivors' 27.09% 26.56% 26.20% 25.95% Hungary Old-Age 56.51% 56.70% 55.18% 54.55% Disability 20.98% 21.50% 22.58% 22.96% Survivors' 14.37% 13.47% 12.64% 11.79/o FYR Macedonia Old-Age 51.51% 52.3 8% 52.78% 54.05% Disability 20.79% 21.14% 21.23% 20.84% Survivors' 27.70% 26.48% 25.99% 25.11% Poland Old-Age 41.94% 44.92% 45.84% 45.96%/o Disability 38.98% 37.53% 37.43% 37.25% Survivors' 18.09% 16.71% 16.36% 16.28% Romania Old-Age 74.76% 75.18% 73.87%/ 72.69% Disability 7.31% 7.63% 8.76% 10.31% Survivors' 17.93% 17.18% 17.37% 17.00% Slovakia Old-Age 50.43% 51.09% 51.01% 50.71%/o Disability 22.19% 22.10% 22.66% 23.10% Survivors' 27.37% 26.80% 26.34% 26. 9% -39- FIGURE 23 & 24: Percentage of Contributors to Working Age Population _ 1993 Albania 0.23 Bulgaria 0.60 Croatia 0.57 Czech R. 0.73 Hungary #N/A FYR Macedonia #N/A Poland 0.56 Romania 0.50 Slovakia 0.61 FIGURE 24: Percentage of Employed to Working Age Population _ 1993 Albania 0.50 Bulgaria 0.61 Croatia 0.57 Czech R. 0.74 Hungary 0.68 FYR Macedonia 0.55 Poland 0.65 Romania 0.75 Slovakia 0.63 FIGURE 25: Replacement Rates (Net Wages) 1990 1991 1992 1993 Albania 52.60% 51.62% 36.33% 41.51% Bulgaria 42.81% 49.72% 38.68% 40.17% Croatia 72.99% 62.77% 56.61% 62.02% Czech R. 47.57% 49.57% 47.40% 43.35% Hungary #N/A #N/A 48.22% 47.28% FYR Macedonia 78.54% 67.29% 75.76% 82.27% Poland 59.03% 67.84% 77.45% 76.75% Romania 41.90% 46.25% 27.20% 25.94% Slovakia 48.25% 50.84% 48.46% 46.35% -40- FIGURE 26: Total Pension Expenditures as Percentage of GDP 1990 1991 1992 1993 Albania 8.56% 12.03% 6.96% 6.46% Bulgaria 10.92% 12.41% 12.18% 14.09% Croatia 13.60% 10.62% 6.04% 6.20% Czech R. 7.66% 7.50% 7.93% 7.98% Hungary #N/A #N/A 11.12% 10.63% FYR Macedonia 10.41% 11.43% 11.44% 15.62% Poland 7.64% 11.50% 13.59% 13.43% Romania 6.94% 7.04% 6.76% 6.24% Slovakia 7.41% 8.11% 9.27% 9.38% FIGURE 26: Pension Benefit Expenditures as a Percentage of GDP L 1990 1991 1992 1993 Albania 6.83% 10.00% 6.59% 6.17% Bulgaria 7.85% 8.74% 9.25% 10.77% Croatia 9.31% 8.68% 5.35% 5.45% Czech R. 7.62% 7.47% 7.91% 7.96% Hungary #N/A #N/A 8.74% 8.55% FYR Macedonia 8.57% 9.53% 9.55% 13.05% Poland 7.30% 10.92% 12.83% 12.69% Romania 5.29% 5.42% 5.28% 5.49% Slovakia 7.41% 8.11% 9.27% 9.38% Distribtultors of' C1NAn GEECE ITALY NORWAY BINIAPOIRE, TAIWAN. Vem ee mEd 11e Tedeiq Cin4 rFhv It Ewa Ppaw- S A lxrsa Commaeshinn t n SPA tioanesrn b nlionwx Caente dYAN14AR, BnUtiEt Ch de LwcVe 41 W orld Bank Atdidwngl 35l Sasala Sair ViaSDixraDiCalta IUl BokDpradIer4 A-airgamlPierlPngAri CAsiaHIM7n y a. .a ro SiFDong Je 106 82 Athens Caseldo Post*eSS2 PO E8x6125Elhenslad PPch4FPe LtM Tel 021)9432673 Publications 8eqg Tel (i)364 1a26 50125Firenze 1106020do6 41 KleNranldruw ngfladl0403 Fae,021)9433605 Prices aid cirdil etrims imy Tel f11333M8257 Fe- 1) 364 8254 Tel: (55)645 415 Td (22)57-3300 GdotenmireWelBrAc g fn'iii COiililry io oitiiily. rFax: (1) 4017365 Fax )55)641-257 Fee: (22)68 1901 Sinppre 349316 TANZAKA HO"G KONG. MACAO ~~~~~~~~~~~~~~~~~~~~Tel: (65) 741 5166 OdaRd UiJendreA Press Consuill yoer 1corl distiiltor COLOMBIA Asia 2000 Lid JAMAICA PAKISTAN Fee (65) 742 0356 MaIla belStr P befole dlaciig anl order: idoDaXce Llda Sales & Cwculabm Deparsreo le Rde Piises itd Meza Bed Agenry eamri ashgile@asiancnnec carom POSes 5299 AjraloAereo 34270 Seabdd Hrmse, xre 1101 02 2060 ld-Hre Roed 65. Shilahr-e-Ouaid e-Azm Darues Salaam Begd D.E. 2226 Wyntdham Streel CearSf Knglon 6 PO Box tlI 729 SLOVAK REPUBLIC Td (i) 29209 ALBAlA Tel- () 2M5-2798 Hong Keng Tel 80927-20W5 Lehoe 54000 SInvarG TG Lid Fix (51i 46822 Aaont Lid Fax (i1285-2795 Tel: 6522S301409 Fa: 809977.0243 Tet (42)7353601 Knpinsta4 Pea- Sir. Tax 85225261107 Fax: (42) 7595293 POBox 152 THAILAND Pal. 9.& 1 Ap. 4 COTE DtvotE UlIl tqrpJhMnwsede&4arla2ODacrrMhk JAPAN B52 99 alrrlaneS Central Boels D iwnbu Thira C*e d Eddxon el del rrtk EarRlem Bree Smire Onierd ltnlxsiy Press Tel (716394H2 306 Slnm Road Tel (42) 274 1 221 72 Ahkrinesr CEDA) HtUNGARY lnge 3 Clrome S Bwgalexe towe Fax (7) 839405 B-agk Fe.. (42) 214 19 04BP541 Foundahloali Maritl Budyo-xk 113 Shaae FrAisal Td (2)235540D AM4e 04 Plxieau Economy Tokyo POSBe 13033 SOUTH AFRICA, BOTSWANA Fax (2)237-832i ARGENTtNA Tel 225-24-6510 Debrear Uli 1719 Tel. (03)38180861 Karacht 75350 Fe ole ls Okna del LtoktremaconW Far 22525-0567 H HIT BSdqseoi Fax. (03( 381 0064 Tel (21) 446307 OxoedUn,,e,Iri Press TRItNtADA TOBAGO, JAM. An. Condb4 1877 Tel 361 204 29Hi a UnL ht1Ip1ebeikorme.cx.V-sAiebs Fax (21)454-7640 SeoAhnAhra Syslenmatics 9uJes Un 1120 BuleaxnAkne CYPRUS 361 204 2940 PO Boe 1141 1egWagStSee Tel (I)6158156 CeleedAppliedResewch Fx: 3612042953 KENYA PERU Cap."eerWn0M Crepe Fa (l) alS8354 Cypm- Co"e Alicaie nk Semxee (F A) Lid Edlexal iestanxralo SA Tel (21)457266 Trieidad. Wil Indies 6. Dieogeesn Siree. Eng_i IND4A Ouranie lee, Mlagana Soet Apalado 3824 Fas. (21)45 7265 Td Mg662-5654 AUSTRALUA, FrIt PAPUANEW GUINEA, rO.Br- 2D6 Ared Puishdrmrs lId Po Box 45245 Lna I Fa. 89g662-5654 SOLOMON LANDS. VANUATU. ANID Nxosa 751 lAuna Trad NMiri Tcd 114)2650W0 rsxhrc4larin .ers WESTERNSAMOA Tel 244-1730 M.-iARs - 60D2 Tld (2)23611 Fee (14(285628 Weriat,nalShdsurwonSerce UGANDA DA emston Seerres Fax: 246 2051 Tel. (44) 852 3938 Fex. (2)33W272 PO B.x41095 GustomLtd 648 Whdwelx o red Faa (44 6529049 PHILIPPINES Ci,Aarglr MaCadSiBaing Mtd:ham 3132 CZECH REPUBLIC KOREA, REPUBLtC OF Inleroal"rel Boolsrcune Cooler Ic Jdnnee g 2024 PO Bee 9997 Vicd8 Naiienal Inormatbion CenKe, INDONESIA Dxeon Iradug Co Lid Sune 720, C4yland IT Tel (lI)0BO.1449 Pot 164 Jiq Rd Tel: p1)392107777 pmexxu. Keekbka S Fl WiseaLr*ed PO Bo-34 CoadminiumTiTne 2 Fex. (lI) No6248 Kaxe4a Fan (61)39210 7760 CS - 113 57 PMagare I Jlan Bobud. 29 Yerod id I V dda Cosna, comer TeVFan: 41( 254763 URL WlF6grJ,edadireclcomue Tel (2)24229433 PO Be 18t Send ValeroSt SPAIN Fax (2) 2422-1484 Jakarla 10320 Tel- (2) 785 163114 Makaoh Metro Marma M -d P,enrs Lbmn,s S A. UNtTED KINGDOM AUSTRIA UllL hpJ/enridsczl Tel (21)390 4290 Fex (2) 7840315 Tel (2)817-9676 CasIelo37 Micrdo Lld GeroldardCo Fax, (21) 421-4209 rox (2) 817 1741 2R0DI Madrid PO Bex 3 GSbe.31 DENMARK MAtAYSIA Tel (1(431-3399 ADa. HmepshnmeGU342PO A lOll Wren SamfndsLtneehl, IRAN Urnery ol Malaya Cooperaibe POLAND rae- (1)5753998 England Tel (1533 50-14-0 Rosenoems Al II Ko.atibPuUlolers BolkstVo. Lkied inlralenaldrutinshin Ser*ice hop /a-relsai es/grrsa Tel (1420)(8646 Fax (1) 512473129 DK-1970 Fredelkseg C Po Bee 19575511 PO Bn- 1127 Ul Pekloa 31/37 Fax (1420) O9M Tel (31) 351942 Tehrran Jalan Fenoi Bare 0 577 Wareara Mird P.ee,sa Barcelona BANGLADESH Fax (31) 357822 Tel. (21) 258 3723 597W0 KualalLunx Tel (2)(6280o9 Crsel de Cern. 391 ZAMBIA 1km Iic dustnes Deelooerel Fas- 98( 21 258 3723 Tel (3) 756 swn rae- (27 621 7255 0W0 Barcelona Uremesr Pernkshop Asrarice Soceley (MIDAS) ErYPT, ARAB REPUBLIC OF Fae (3) 755 4424 Td (3) 488 30CR Greal Easl Road Camepus FIneseS Rnad 16 Al Ahram Kelib Sara Co Pddlishers POlITUGAL Fae (31497 7659 PO Boe 32379 Dtearxxdi R!hna Al Gaiax SIreeE KEie*d Estenboli Ave. MEXICO LIVInnO Frloirgl Lusaka Ehaka 1209 Cuio 6th Shel diOlroC Tir Ti C-oe 7074 sRm LANKA, TIIE MALDIVES Tel it) 213221 Et 482 Tel. 12) 326427 Tel (2) 576 60E3 KuIsheh DedriR No 8 0 Apalada resloa 22 860 1200 1 lsitar Lake lleuse foroksrp Fax (2)611198 Fee: (2)i57v633 Tehran 14060 1aIran. TIl (l(3474982 PO) Bee 244 ZIMBABWE Tel 8717819rn 8716104 MerxcoOF Fx- (1) 3410764 10, Se CS. lairabamA Longman Zxartete (Pie 9Id BELGIUM lhe MddeEeosiObsreri Fe. 8062479 Tel (5 606 0011 Gaedner Mnralha Toue Read. Ahdlr JeanEDeLuary 41, Sher Seel Fex: (5)6060386 ROMANIIA Celonto 2 PO Box ST125 An du flo 202 Cako IRELAIID Cromipai DeTibLraniBucursl S A. Tel (1( 32105 Southeoeon 06oEBrnssels Fel. t2)3939732 Gererrxmenil SrlesAgeney NETIHERLANDS Sir Ipscarlne 26, sedle3 Fe. (1(432104 Harare Te 2) e38 5169 Fee. f2) 393-9732 01mg an ISothnair De LeWdebeonel0-FPubiokelres Bircxl, sl Td (41)6216617 Far (2) 3tt0641 4-5 Ilecml Road PF B. 202 Tel (116130015 SwEnEN Faa 14(621670 FINLAND Du[ian 2 74000AE lxaaksbergen Fax (1i 312 4000 fr`les Customer Senrce BRAZIL Akaleenren Kkjak.upa Tel (1)401-3111 Tel (53) 574 0104 Regenrg oale 12 Publiaces TKneas Intenmcionals PO. Boo 23 Fao (1) 475 2670 Fae (53)572-9296 RUSSIAN FEDERATION 5 106 47 Slrlrholm Lka Fel 00371 HetkIl lsdnldslwo *Vees M-r Te (aj 690 90 90 Rux PeienolGoede, 209 Tel (0(12141 ISRAEL HEW ZEALAND 9a. Kolpad,ro reulor Fax (9(214777 O04095SoPb. SP Far (0)121-4441 Yo,molrtieralreLUd EBSCONlZLid Mrion 101031 Tel 11)2590644 ULV- hilphbooknelcudtilelthaka/ PO boxO55 PFale Mai Bag 99914 Tel (15(9178749 Weonexgen Wdahnms AB Fee (i) 25806990 Tel Am+, 61560 Ner Ma.kel Fa. (95) 917 92 59 P 0 Boo 1305 FRANCE Tel. (31 5285 397 AucLuand S 171 25 Solna CANADA Word dark Pubbcalions Fae (3j5285 397 Tel. (90524 8119 SAUDIAI IABIA, OATAR Tel (6(7059750 Reerd PLib h Co Lid 66 axrne d dea Fae (9(5240067 Jai, Book Stlre Fx (08) 2700 71 1294 Alra Pad 75116 Pon$ ROY oleoraleonal ro One 3196 Oeaaea OlantKIB I3W Tel. JI(40693085657 POBox 13056 NIGERIA Mllyndh 11411 SWITZERLAND Tel.613741-43,3 Fe. (1(40-6930-66 TeloAv 61130 ULioeroly Press Urrxld Tel (1(4773440 Ubnmeie Payol Fax 613-7415439 Tel (3) 5461423 ThreerCoons Raiding Jeclr F ax o1)(477 2940 Sel*e lrslilvrleal GERMANY Fae (3( 5461442 Pinrale Mad Bag 5095 Cloes de Monibene 30 UttO Veiag badan 1002 Lxasanne PoppelsdafreAlee 55 Palesinan AulhontlyM edle.Eol TIl (22(41-1356 Tel. (021) 341 3229 53115Ben rIdcxilonIrn,lion Srvices Fae (22)412056 Fae (021(341 3235 Tel (2261212940 PFOB 19407 J.uraIem Frx (220)217492 Tel (21271219 RECENT WORLD BANK TECHNICAL PAPERS (conitimiied) No. 300 Oram and de Haan, Tcchlnologies tfor Rainifed Agriculture in Mciliterrancani Clitiatets: A Revicu, of Worldi Banki Experiences No. 301 Mohan, editor, BiNzo',,'raphln of Pulications: Tcchlinical Departnicnt- Africa Rcgion. liliI 19871 to April 1995 No. 3(02 Baldry, Calarmari, and Yarmeogo, Etnra ronmeental Inipact Assessment of SettIenCit and De7eloynijpent inl t(li Upper Lerab1a Basin No. 303 Heneveld and Craig, Schools Coliit: World BaneSk Project Dcsignsand th,e Qiianlitii of Priniana , Edntlcationl in Siit-Saha ran Africai No. 304 Folev, Photocoltaic Applicationis inl Rntril Areas of(ltile Dcevelopig Vv' World No. 305 Johnson, Ed neation andl Training ot Accoiin aillts in Sn -Sh1ti r1-anl A nglo phone A frica No. 306 Muir and Saba, Iniproving Stntct En tirprise Perfltornia cc: Thec Role of Initerinal an d Exterinal hliceo tics No. 307 Naravan, Toward Participatoril Risearb11 No. 308 Adamson and others, Elergit llsi'. Air Pahillitton i.ai andELn iin litc iit(l PolcY yll Krakowu: Cal Econoninic lbcenztincs Reaillif Help? No. 309 The World Bank/FOA/LNIDO/Industry Fertilizer Working Group, World a71nd1 RCgiolnal Slippll andlL Dcciniaiid Balanices for Nitrogeni, Phosphate, andt Potash, 1993/94-1999/200)() No. 310 Elder and Cooley, editors, Siistama illlc Settlnicict (nlad Dcucloptieiit of tlc t'Onchocerciasis Control Pro granininc Arna: Proccedinigs of a .Vliiiistcriill Meting No. 311 Webster, Riopelle and Chidzero, World Baiik Leiidingfor S,niall Entcrprises 19849-1993 No. 312 Benoit, Project Fina lice at (lit' World Baii k: A li Ovtcr. 'ico of Policies aii0 ltrisuililieilts No. 313 Kapur, Airport Infrastrnictire: The ELicrginig Role of t(e PricatC Scctor No. 314 Valdes and Schaeffer in collaboration with Ramos, SiirOicflaiice of Agricultuiral Price andlf Trade1 Policles: A -landl,book tor Eciuador No. 316 Schware and Kimberlev,1 nforination Tceliniology aii Na1tioital Trade Facilitationl: Makiiig t(le Most of Glottal Trtide No. 317 Schware and Kimberlev, lIfriiiatiou Tcciiolo,,giy and Natiolll Irade Facilitation: GuidLe to Best Practice No. 318 Taylor, Boukambou, Dahniya, Cual(Vogode, Avling, Abdi Noor, and Toure, Streiigthenibig National Agricultural Research Silstetis iii t(ie Hii mid aniid Su l-hldiii u1 Zonets of Wcst niil Ccltlia1 Africa: A Franicicork tor Actiou No. 320 Srivastava, Lambert and Vietmever, Meduciiual PlaiLts: Ani L.xpaiidniig Rolc ii Developiment No. 321 Srivastava, Smith, and Forno, Bioduotcrsitl atid1 Agricit iluic: Iiptlieti(ions fOr COluserCatiolu aitid Devzelopiicnit No. 322 Peters, The Ecologi a I ' Mani nageicnn t of Non - ritid cr Forest RPCtOreC's No. 323 Pannier, editor, Corpora t' Goicr,nailcC of PuiNllic Eu tcrpriscs iii Trm iisitioiual Ecoiooiniics No. 32)4 Cabraal, Cosgrove-Da ies, and Schaeffer, Best PractuctS for Phiotozvoltaic Ho usehold Electrification Progranis No. 325 Bacon, Besant-Jones, and Heidarian, Estintating CoIistrictioll Costs andld Scihduiles: Experience ,vith Pot'er Geticratioiu Projects inl Deoelopiig Coui ii tries No 326 Colletta, Balachander, Liang, The Coi17ditioi Oft Yoiiig Chillr.i Ill Silil-Sahllanll Africr: Tae Contilthi, Niitritioii, and Earhly Eduicatioii No. 327 Valdes and Schaeffer in collaboration with Martin, Suirotcillancc of Agriciultural Price and Trade Policies: A Handbook tor Pa ragiaYi No. 328 De Gevndt, Socitl Developnuieit inud Absol ii te Ptoztcrt iin Ash land Lnitin Al ierica No. 329 Mohan, editor, Bubliograpliln of P1ulicationIs: Tecliiuil17 Dep(a7rtiie)it, Afr ica Regioii, hiili 1987 to April 1996 No. 332 Pohl, Djankov, and Anderson, Rtstruicturinig Lrgte ldiclitstriiiil Finns ii Cc itrialatiti Eastern Europef: Anl Enllpricil Analpisis No. 333 Jha, Ranson, and Bobadilla, Mcastring (lit' Burden of Disease iiiudl t(e Cost-Effi'ctiveness of Hiealth Iiuterne,itioins: A Case Stidly iil Guilsicln No. 334 Mosse and Sontheimer, Performiaice Moniitoring Indicators Haiuldlook No. 335 Kirmani and Le Moigne, Fost crzig Ripariani Cooperation 1in IiteCriatiotnal Rivcr Basiiis: ThIe World Bank it Its Best itn Dccvlopnment DiplonuacY, No. 336 Francis, with Akinwvumi, Ngwu, Nkom, Odihi, Olomajeve, Okunmadewa and Shehu, State, Coinuininity, anid Loci7l Deotelopnuicnt in Nigeria No. 338 Young, Measiiring Ecouolnic Beincftsfior Water liczStiiue,its atid Polilces - ~~~~~~~~~~~~~~~~~~~~~~~~~~ - s~~~~~~~~~~~~~~~~~~~~~~~~