10059 Reform ing Centralg and Eastern -European Economiies Initial RVesullts and Challenges edited by Vittorio Corbo Fabnizio Coricelli Jan Bossak A World Bank Symposium FILE COPY Reforming Central and Eastern European Economies A World Bank Symposium Reforming Central and Eastern European Economies: Initial Results and Challenges edited by Vittorio Corbo Fabrizio Coricelli Jan Bossak The World Bank Washington, D.C. ©1991 The International 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 September 1991 The findings, interpretations, and conclusions expressed in this study are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Direc- tors or the countries they represent. Because of the informality of this series and to make the publication available with the least possible delay, the manuscript has not been edited as fully as would be the case with a more formal document, and the World Bank accepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to repro- duce portions of it should be sent to Director, Publications Department, at the ad- dress 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 photocopy portions for classroom use is not required, although notification of such use having been made will be appreciated. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of sub- jects, authors, and countries and regions. The latest edition is available free of charge from the Publications Sales Unit at the address in the copyright notice or from Publications, World Bank, 66, avenue d'Iena, 75116 Paris, France. Library of Congress Cataloging-in-Publication Data Reforming Central and Eastern European economies: initial resultsandchalleng- es / edited by Vittorio Corbo, Fabrizio Coricelli, Jan Bossak. P. cm. Papers from a conference held in Pultusk, Poland, Oct. 4-5, 1990; organized by the Macroeconomic Adjustment and Growth Division of the World Bank. Includes bibliographical references. ISBN 0-8213-1893-4 1. Central Europe-Economic policy-Congresses. 2. Central Europe-Econom- ic conditions-Congresses. 3. Privatization-Central Europe-Congresses. 4. Eu- rope, Eastern-Economic policy-Congresses 5. Europe, Eastern-Economic conditions-1989-Congresses. 6. Privatization-Europe, Eastern-Congresses. I. Corbo, Vittorio. II. Coricelli, Fabrizio. III. Bossak, Jan. IV. International Bank for Reconstruction and Development. Economics Dept. Macroeconomic Adjustment and Growth Division. HC244.R3775 338.947-dc2091-27563 CIP Contents Preface xi Contributors xiii 1. Introduction Vittorio Corbo, Fabrizio Coricelli andJan Bossak Overview 1 Part 1. Central and Eastern Europe: An Historical Background 2 Part II. The World Bank Experience with Adjustment Lending 2 Part III. Speed of Adjustment and Sequencing of Policy Changes 2 Part IV. Stabilization: Design of Programs, Effects on the Economy 3 Part V. Structural Rigidities, Distortions and Inertia in Behaviors: The Supply Response to New Environments and Constraints in Central and Eastern Europe 3 Part VI. Adjustment and Reforms: Financial Market, Foreign Trade, Fiscal Sector 4 Part VII. Panel on Privatization 4 Part VHII. Summing up and Overview 5 References 6 PART I CENTRAL AND EASTERN EUROPE: AN HISTORICAL BACKGROUND 7 2. The Economies of Central and Eastern Europe: An Historical and International Perspective 9 Andrds Solimano Central and Eastern Europe Before Socialism: The Inter-war Period 9 The Collapse of Socialism in Central and Eastern Europe: The Late 1980s 18 Concluding Remarks and Summary 20 Notes 21 References 22 PART II THE WORLD BANK EXPERIENCE WITH ADJUSTMENT LENDING 25 3. World Bank-Supported Adjustment Programs: Lessons for Central and Eastern Europe 27 Vittorio Corbo v The Emerging Consensus on Program Design 28 The Main Findings of RAL-2: The Effectiveness of Adjustment Programs and Their Implementation 29 Main Results of RAL-2: Policies That Promote Growth 34 Lessons for Central and Eastern Europe 38 Conclusions 40 Notes 40 References 41 4. World Bank Adjustment Lending in Central and Eastern Europe 43 Ulrich R. W. Thumm A Brief Summary of the Lessons from World Bank Adjustment Lending 43 Crucial Economic Problems in Central and Eastern Europe 44 Key Elements of an Economic Transformation Program for Central and Eastern European Economies 50 World Bank-Supported Programs in Central and Eastern Europe 51 Conclusions 56 Notes 57 References 58 5. Comments on "World Bank Adjustment Lending in Central and Eastern Europe," by Ulrich R.W. Thumm 59 Miroslav Hmcir 6. Comments on "World Bank-Supported Adjustment Programs: Lessons for Central and Eastern Europe," by Vittorio Corbo and "World Bank Support for Adjustment Lending in Cen- tral and Eastern Europe," by Ulrich R.W. Thumm 63 Stanislaw Gomulka Notes 64 References 64 PART III SPEED OF ADJUSTMENT AND SEQUENCING OF POLICY CHANGES 65 7. Issues in the Reform of Socialist Economies 67 Stanley Fischer and Alan Gelb Initial Conditions 67 Elements of Reform 71 The Sequencing of Reforms 76 The Role of Foreigners 78 Conclusion 79 Notes 80 References 81 PART IV STABILIZATION: DESIGN OF PROGRAMS, EFFECTS ON THE ECONOMY 83 8. Stopping Inflation: The Experience of Latin America and Israel and the Implications for Central and Eastern Europe 85 Miguel A. Kiguel and Nissan Liviatan vi Inflation and Stabilization 87 Inflation in Central and Eastern Europe 90 Stabilization Strategies 94 The Stabilization Programs in Yugoslavia and Poland 97 Notes 98 References 99 9. Stabilization Programs in Eastern Europe: A Comparative Analysis of the Polish and Yugoslav Programs of 1990 101 Fabrizio Coricelli and Roberto de Rezende Rocha Background 102 Conditions Preceding Stabilization 103 A Comparison of the Two Stabilization Programs 105 Examination of the Initial Results 107 Assessment of the Main Issues 112 Appendix 1. Stabilization in Poland 113 Appendix 2. Stabilization in Yugoslavia 122 Notes 132 References 133 10. Comments on "Stopping Inflation: The Experience of Latin America and Israel and the Implications for Central and Eastern Europe" by Miguel A. Kiguel and Nissan Liviatan 135 John Williamson Notes 137 References 137 PART V STRUCTURAL RIGIDITIES, DISTORTIONS AND INERTIA IN BEHAVIORS: THE SUPPLY RESPONSE TO NEW ENVIRONMENTS AND CONSTRAINTS IN CENTRAL AND EASTERN EUROPE 139 11. The Ownership-Control Structure and the Behavior of Polish Enterprises During the 1990 Reforms: Macroeconomic Measures and Microeconomic Responses 141 Roman Frydman and Stanislaw Wellisz The Economic Reforms 141 The Socialized Sector 142 Effects of the Reforms on Enterprise Behavior 143 Incentives and Responses 153 Notes 154 References 155 12. Life after the Polish "Big Bang": Representative Episodes of Enterprise Behavior 157 Erika A. Jorgensen, Alan Gelb and Inderjit Singh The Methodology and the Sample 157 The 1990 Policy Package 159 The Effect of the Big Bang Policy Package at the Firm Level 160 The Responses of Industrial Firms to the Big Bang: Some Findings from a Field Trip 161 vii Did the Big Bang Create a New "Sink or Swim" Environment for Firms? 169 Conclusions 172 Notes 172 References 173 13. Comments on "Life after the Polish 'Big Bang': Representative Episodes of Enterprise Behavior," by Erika A. Jorgensen, Alan Gelb and Inderjit Singh 175 Jan Svejnar 14. Distortionary Policies and Growth in Socialist Economies 177 William Easterly Introduction 177 Distortionary Policies in Socialist Countries 177 A Model of Endogenous Growth with Distortionary Policies 180 Estimating the Growth Effects of Distortion 184 The Problems of Partial Liberalization 186 Conclusions 188 Notes 188 References 188 15. Comments on "Distortionary Policies and Growth in Socialist Economies" by William Easterly 191 Jacek Rostowski Notes 193 PART VI ADJUSTMENT AND REFORMS: FINANCIAL MARKET, FOREIGN TRADE, FISCAL SECTOR 195 16. Financial Aspects of Socialist Economies: From Inflation to Reform 197 Guillermo A. Calvo Bad Credit: A Money Machine 197 Deactivating the Money Machine 198 Good Credit 200 Financial Policy 201 Final Remarks 203 Notes 204 References 205 17. Adjustment, Trade Reform and Competitiveness-The Polish Experience 207 Jan W. Bossak Outward-Oriented Development: The Emerging Orthodoxy 207 Adjustment with Sustainable Growth: The Challenge 209 Some Issues Raised by Adjustment 210 Policy Reforms under the New Polish Regime 211 The Foreign Debt Problem 213 Conclusion 214 Notes 214 References 214 viii 18. Comments on "Adjustment, Trade Reform and Competitiveness-The Polish Experience" by Jan W.Bossak 217 Gdbor Oblath 19. Fiscal Issues in Economies in Transition 221 Vito Tanzi Restructuring and the Institutional Requirements 221 Restructuring and the Fiscal Institutions 222 Concluding Remarks 225 Appendix 1. Statistical Tables 227 Notes 228 References 228 PART VII PANEL ON PRIVATIZATION 229 20. Privatization in Eastern Europe: The Case of Poland 231 David Lipton and Jeffrey Sachs The Debate over the Pace of Privatization 232 Current Ownership Patterns in Poland 233 Corporate Governance and Financial Intermediaries 240 A Strategy for Privatization 242 Conclusion 248 Notes 248 References 250 21. Markets and Institutions in Large-Scale Privatization: An Approach to Economic and Social Transformation in Eastern Europe 253 Roman Frydman andAndrzejRapaczynski Why Privatize? 253 The General Principles of Privatization 255 The Problem of Valuation 256 The Sale Model 258 Free Distribution Models 260 The Problem of Control 262 Conclusion 267 Appendix 1. Valuation of State Enterprises in the Polish Economy: A Brief Case Study 268 Appendix 2. The Polish Government's Large-Scale Privatization Plan: A Preliminary Ananlysis 269 Appendix 3. Money and the Question of Voucher Denomination 272 Notes 273 References 274 22. A Note on the Privatization of Socialized Enterprises in Poland 275 Manuel Hinds The Government's Proposed Approach 275 The Requisites of a Privatization Scheme 276 ix How Vell Does the Government's Plan Meet Its Objectives and the Requisites of Privatization? 278 Desirable Modifications to the Proposed Approach 281 Notes 286 References 287 23. Comments on David Lipton and Jeffrey Sachs, "Privatization in Eastern Europe: The Case of Poland," Roman Frydman and Andrzej Rapaczynski, "Markets and Institu- tions in Large-Scale Privatization: An Approach to Economic and Social Transformation in Eastern Europe" and Manuel Hinds, "A Note on Privatization of Socialized Enterprises in Poland" 289 E. Borensztein Notes 292 References 292 PART VIII SUMMING UP AND OVERVIEW 293 24. The Symposium in Review and a Glance at the Future 295 Stanislaw Gomulka, Johannes F. Linn and Jan Svejnar Comments by Stanislaw Gomulka 295 Comments by Johannes F. Linn 296 Comments by Jan Svejnar 298 Notes 299 References 299 Participants 301 x Preface At the request of the World Economy Research Insti- Poland, Romania and Yugoslavia, as well as academics tute (WERI) of the Warsaw School of Economics, the from western countries and representatives of multilat- Macroeconomic Adjustment and Growth Division of the eral institutions. The Conference was organized by Vit- World Bank organized a Conference in Pultusk, Poland torio Corbo and Fabrizio Coricelli, of the Country on October 4-5, 1990. The aim of the Conference was to Economics Department of the World Bank, and Jan draw from reform experiences in developing countries Bossak of the WERI. Whitney Watriss has done an excel- some practical lessons for Central and Eastern European lent job of copy-editing the volume in a very short period countries. The Conference included papers that evaluat- of time. ed international experiences, proposals for reform and Although there are no precedents for the economic progress reports on the initial response to reforms in reforms being initiated or about to be initiated in Cen- some of the Central and Eastern European countries. tral and Eastern Europe, a careful evaluation of reforms Most of the papers published in this volume were pre- in specific areas that have been carried out in other sented and discussed at the Conference in Pultusk. The countries could provide important lessons for the policy- papers by Andres Solimano, providing an historical makers who are responsible for the reform programs in background, and by Eduardo Borensztein, comparing Central and Eastern Europe. We hope this volume will the three proposals for privatization contained in the contribute to providing some guidelines for the unprec- volume, were prepared after the Conference. edented reforms which are now underway. In addition to the authors of the papers, the Confer- ence brought together a range of distinguished officials Vittorio Corbo and academics from Bulgaria, Czechoslovakia, Hungary, Fabrizio Coricelli Acknowledgments This Conference was made possible thanks to the support of the following departments of the World Bank: Europe, Middle East and North Africa, Division 4 (EM4), Country Economics (CEC), Research Administration (RAD) and Development Economics (VDDEC). The staff of the World Economy Research Institute of the War- saw School of Economics and Barbara Ossowicka of the World Bank provided much support in the preparation and the implementation of the Conference. xi Contributors Eduardo Borensztein Stanislaw Gomulka International Monetary Fund, London School of Economics Research Department, Washington, D.C. Manuel Hinds Jan W. Bossak World Bank, Europe, Middle East and North Africa, World Economy Research Institute, Trade and Finance, Washington, D.C. Warsaw School of Economics Miroslav Hrncir Guillermo Calvo Institute of Economics, Prague International Monetary Fund, Research Department, Washington, D.C. Erika Jorgensen World Bank, Socialist Economies Reform Unit, Vittorio Corbo Washington, D.C. World Bank, Macroeconomic Adjustment and Growth Division, Washington, D.C. Miguel Kiguel World Bank, Macroeconomic Adjustment Fabrizio Coricelli and Growth Division, Washington, D.C. World Bank, Macroeconomic Adjustment and Growth Division, Washington, D.C. Johannes F. Linn World Bank, Country Economics Department, William Easterly Washington, D.C. World Bank, Macroeconomic Adjustment and Growth Division, Washington, D.C. Nissan Liviatan Hebrew University of Jerusalem and Stanley Fischer World Bank, Macroeconomic Adjustment World Bank, Office of the Vice President, and Growth Division, Washington, D.C. Development Economics and Chief Economist, Washington, D.C., and Massachusetts Institute Gabor Oblath of Technology, Economics Department, Cambridge, Institute for Economic and Market Research, Budapest Mass. Andrzej Rapaczynski Roman Frydman Columbia University, Law School, New York New York University, Department of Economics Roberto R. Rocha Alan Gelb World Bank, Macroeconomic Adjustment World Bank, Socialist Economies Reform Unit, and Growth Division, Washington, D.C. Washington, D.C. xiii Jacek Rostowski Vito Tanzi Polish Ministry of Finance, Warsaw International Monetary Fund, Fiscal Affairs Department, Washington, D.C. Jeffrey Sachs Harvard University, Department of Economics, Ulrich R.W. Thumm Cambridge, Mass. World Bank, Europe, Middle East and North Africa, Washington, D.C. Inderjit Singh World Bank, Socialist Economies Reform Unit, Stanislaw Wellisz Washington, D.C. Columbia University, New York, and Polish Ministry of Finance, Warsaw Andres Solimano World Bank, Macroeconomic Adjustment John Williamson and Growth Division, Washington, D.C. Institute for International Economics, Washington, D.C. Jan Svejnar University of Pittsburgh, Department of Economics xiv 1 Introduction Vittorio Corbo, Fabrizio Conicelli and Jan Bossak Overview alike. Clearly, a blueprint for the task is not available, and, given significantly different initial conditions across coun- In the last two years, countries in Central and Eastern tries, it may even be misleading to search for one. However, Europe have initiated major reform efforts with the final despite the unique character of the reform in each country, objective of creating a Western European-type market experience with reform in other parts of the world can pro- economy. Undoubtedly, implementing the reforms has vide some useful guidelines. Analyzing the issues raised by been very difficult. The difficulties have been of three reform in the socialist economies against the background kinds. First, after 40 or more years of a command econ- of this experience will not only provide insights and guid- omy, institutions and the system of incentives are far re- ance but will also help isolate the unique features of social- moved from those required to establish a Western ist economies. European-type market economy. Therefore, significant Over the last 10 years the World Bank has gained wide changes in institutions and the system of incentives are experience in the design and implementation of reform required. The significant changes needed in the produc- programs. In particular, two reports (Report on Adjust- tive and institutional structure are bound to result in ment Lending 1 [RAL-11 [World Bank 19881 and Report important short-term adjustment costs. Second, impor- on Adjustment Lending 2 [RAL-21 [World Bank 1990]), tant political economy problems will emerge as the un- completed in the past two years, provide comprehensive avoidable short-term costs of adjustment will occur at reviews of its experience with adjustment programs. The the same time that a fundamental redistribution of the original motivation of the Conference on Adjustment political regime takes place. Third, there are no prece- Lending: Lessons for Eastern Europe, the proceedings of dents for such far-reaching reforms from which to derive which are contained in this volume, was to draw from clear lessons on program design. those reports the main lessons for reform in Central and The conference that gave origin to this volume fo- Eastern Europe. As these reports do not focus specifically cused on the economic reforms required to create a mar- on Central and Eastern Europe, a decision was made to ket economy and, specifically, on the main challenges. complement the lessons from the reforms supported by The Central and Eastern European countries have to the Bank with a set of papers dealing with particular is- create the basic structure and institutions needed for a sues that reforms in the Central and Eastern European market economy from the ground up and at the same economies raised. time implement mac-oeconomic programs aimed at sta- The volume covers a wide range of issues, touching bilizing large macroeconomic imbalances. The interre- on key problems at both the macroeconomic and micro- lation between these structural issues, which are economic levels. In fact, the interplay between macro- generally of a microeconomic and institutional nature, economic and microeconomic issues is the leit-motif of and the macroeconomic issues raises fundamental ques- the volume. The structure of the volume reflects this tions about the design and implementation of reform perspective by including analyses of the problems of programs. Fundamental tensions between the micro- macroeconomic stabilization, the development and re- and macroeconomic dimensions of the reform can arise. form of key markets and, finally, privatization. While Answering these many questions that the transforma- containing a host of important suggestions derived from tion of the Central and European countries poses is an ex- the experience of developing countries, the volume also traordinary challenge for policy-makers and researchers discusses some of the unique features of reform in Cen- 1 tral and Eastern Europe, in particular the fact that reform itate the economic restructuring. Finally, institution- in these economies requires the creation of markets and building, including the capacity to formulate and execute market institutions, and not just the removal of distor- macroeconomic policies, reduction in the size of the pub- tions. lic sector, including moving out fast on privatization and The lack of the basic institutions and structures found a regulatory framework for the functioning of markets are in market economies makes the traditional models for fundamental conditions for successful reforms. studying macro and micro policies in market economies In "World Bank Adjustment Lending in Central and inadequate for analyzing the transition of socialist econo- Eastern Europe," Ulrich Thumm reconstructs the main mies. This volume is a first step in isolating some of the features of adjustment cum reform programs supported unique issues facing these countries in their transition by Sector Adjustment Lending (SALs) from the World and suggesting appropriate conceptual frameworks. Bank in Hungary, Poland and Yugoslavia. The paper also briefly reviews the incipient adjustment program in Part 1. Central and Eastern Europe: An Czechoslovakia. In his analysis of the recent economic Historical Background performance of countries in Central and Eastern Eu- rope-which are characterized by large macroeconomic Part I provides historical background, drawing espe- imbalances, external debt problems and low efficiency- cially on the period before the Second World War, when Thumm notices the prima facie similarity of this experi- these Central and Eastern European countries had mar- ence with that of other countries, which is very familiar ket economies. The paper also provides an overview of the to the World Bank. Despite this similarity in performance, basic conditions in these countries on the eve of the cur- he describes a set of systemic features of socialist econo- rent reform effort. mies that make the design and implementation of adjust- ment in Central and Eastern Europe unique. The lack of Part II. The World Bank Experience with private ownership-which is of critical proportions-has Adjustment Lending been the main stumbling block to a successful restructur- ing of these economies. Moving from this premise, Thumm argues that, while the socialist economies have Part II consists of two papers that focus on the main undertaken fundamental macroeconomic measures and lessons for reforming socialist economies, as drawn from important reforms in the areas of price and trade liberal- the World Bank's experience with adjustment lending. ization, the transformation of enterprises with respect to The main objectives are to highlight aspects of adjust- both their restructuring or privatization, has been pro- ment that have general applicability and, at the same ceeding slowly. While recognizing that their structural, time, to stress some specific features of reforming social- systemic transformation will necessarily take time, ist countries that require deviations from the approaches Thumm concludes that important steps to create a criti- pursued by other countries, such as in the areas of se- cal mass of private ownership that can initiate the move quencing, the interrelation of macro and micro aspects, toward a market economy have to be taken at the outset and the pace and intensity of liberalization and reforms. of the program, together with macroeconomic measures. In "World Bank-Supported Adjustment Programs: Les- sons for Central and Eastern Europe," Vittorio Corbo Part III. Speed of Adjustment and Sequencing draws from the accumulated experience of the Bank four of Policy Changes main lessons for Central and Eastern Europe. First, where high inflation-even if repressed-and severe macroeco- nomic imbalances are present, the adjustment program Part III establishes a precise framework for the se- has to tackle them at the beginning. Second, to achieve a quencing of policy measures and structural reforms un- supply response early on in the reform, trade liberaliza- derlying the transition to a market economy. In "Issues tion has to be accompanied by reforms reducing the re- in the Reform of Socialist Economies," Stanley Fischer strictions on labor mobility and the entry and exit of and Alan Gelb emphasize that the path of the transition firms. In this regard, labor market policies, as well as and the sequencing of the measures depend crucially on housing policies, and anti-monopoly legislation have to initial conditions. They classify these initial conditions on be addressed at the outset of the reform program. Third, the basis of: the degree of internal and external macroeco- although the creation of a full-fledged financial system nomic imbalance; and the degree of economic centraliza- should be postponed to the later stages, a minimal finan- tion. Given that the reform will deal with both cial structure capable of channelling working capital fi- macroeconomic stabilization and economic decision- nancing on the basis of an appropriate assessment of making, the degree of difficulty of the reform and the se- credit risk has to be established at the beginning to facil- quencing of measures in different countries will be deter- 2 mined by their relative position in terms of those factors. been substantial. Coricelli and Rocha argue that, notwith- Fischer and Gelb then go in detail into the main areas standing the apparent similarities, the initial outcomes of where reforms will be needed. They identify eight main the two programs are significantly different, a result that areas and suggest a phased set of policy measures, with a can be explained by the different initial conditions and the time horizon for each. The eight areas are macrostabiliza- different sequence and scope of some policy measures. Fi- tion, a social safety net, institutional reform, price and nally, the authors discuss the main issues to be addressed market reform, small-scale privatization and develop- in the second stage of the programs, including the un- ment of the private sector, large-scale restructuring and freezing of the nominal anchors (wages and exchange privatization, development of an autonomous banking rates) and crucial structural issues whose solution will system, and establishment of other financial markets. determine the prospects for a sustained supply response. Based on an evaluation of the outcomes of both programs, Part IV. Stabilization: Design of Programs, Coricelli and Rocha state that macroeconomic policies Effects on the Economy alone cannot generate more efficient allocation of re- sources and efficient natural selection among enterprises. Part IV addresses the design of stabilization programs They conclude that the sequencing model traditionally and the choice of policy mix and policy instruments. The applied in Latin American countries, in which structural two papers provide an interesting perspective by compar- reforms are relegated to the later stages of adjustment, ing the experience with stabilization programs in Latin does not seem to be applicable to reforming socialist American countries and Israel, on the one hand, and that countries, where stabilization and structural reforms- of two socialist economies, Poland and Yugoslavia, which and thus macroeconomic and microeconomic issues- carried out similar stabilization programs in 1990, on the are more closely intertwined. other. In "Stopping Inflation: The Experience of Latin Amer- Part V. Structural Rigidities, Distortions and ica and Israel and the Implications for Central and East- Inertia in Behaviors: The Supply Response to ern Europe," Miguel Kiguel and Nissan Liviatan New Environments and Constraints in Central emphasize the common elements between Latin America and Eastern Europe and Central and Eastern European countries such as Po- land and Yugoslavia in terms of both the sources and per- Part V provides an initial assessment of firms' respons- sistence of inflation and the design of stabilization es to the reforms in Poland. Because their response to the programs, including the main problems relating to their new institutions and incentives will determinate the ulti- sustainability. Kiguel and Liviatan stress the difficulty of mate success or failure of the reforms, it is of central im- stopping inflation in chronic high-inflation countries portance to obtain a progress report on how firms are such as Argentina and Brazil. They divide stabilization adjusting to the reforms. In particular, it is helpful to programs into two main groups. One, labeled "orthodox," know if the response is more of a restructuiring tyre or of is based on fiscal adjustment and the use of money or the a wait-and-see type. exchange rate as "anchors." The other, "heterodox," can In "The Ownership-Control Structure and the Behav- be defined as a fiscal adjustment with the addition of price ior of Polish Enterprises During the 1990 Reforms: Mac- and/or wage controls to deal with the inertial aspects of roeconomic Measures and Microeconomic Responses," inflation. Kiguel and Liviatan classify the Polish and Yu- Roman Frydman and Stanislaw Wellisz, on the basis of goslav stabilization programs as heterodox. They note, very rich empirical evidence, argue that the stabilization however, that in both countries the initial response of the and reform program launched in Poland in 1990 forced economy differed from that found with typical heterodox firms to adjust and in particular to adopt pricing that programs. In particular, the recessionary impact was largely eliminated shortages in the economy. They also larger, and, in the case of Poland, the improvement in the argue, however, that although enterprise behavior is ra- current account did not match the usual experience. tional in the present framework of institutions and incen- In "Stabilization Programs in Eastern Europe: A Com- tives, it is not conducive to greater efficiency because of parative Analysis of the Yugoslav and Polish Programs of the structure of rewards and ownership. In particular, 1990," Fabrizio Coricelli and Roberto Rocha discuss the they note the persistence of labor hoarding and the recent stabilization programs implemented in Poland and Yugo- resurgence of wage pressures, validated by an expansion slavia in 1990 at almost the same time. Interest in a com- in credit. Frydman and Wellisz conclude that in the ab- parative analysis of the design, performance and sence of deep changes in the structure of ownership and sustainability of the two apparently similar programs be- control of enterprises, a restrictive aggregated demand ing implemented by two reforming socialist countries has policy cannot be effective in achieving rationalization and 3 growth. However, an expansionary demand policy will ation and development in the overall reform in socialist also be ineffective and will result in inflationary pres- economies. Part VI effectively illustrates the size of the sures. Indeed, they claim that state-owned firms that are problems that Central and Eastern European countries managed de facto by workers, will respond to increased face in trying to move toward market economies given demand by raising wages and prices rather than output. their lack of financial markets, the particular structure of Therefore, to restore economic growth, the most needed their foreign trade and the inadequacy of the fiscal sector. reform is the development of a well-defined system of In "Financial Aspects of Socialist Economies: From In- ownership and control of enterprises. flation to Reform," Guillermo Calvo focuses on some fi- In "Life after the 'Big-Bang': Representative Episodes," nancial issues connected with the transformation from Erika Jorgensen, Alan Gelb and Inderjit Singh analyze the planned to market economies. The central theme is that preliminary results of the Polish reform based on a study credit in socialist economies has generally taken the form using a small sample of enterprises. From this sample of "bad" credit, i.e., credit that finances ultimately unpro- they observe that after initial chaos in the first months of ductive activities, such as, for instance, the accumulation 1990, firms began to get a sense of their markets for both of excessive inventories. The reason for this problem is a inputs and output. The nine case studies of Polish firms soft budget constraint. In moving to a market-oriented indicate that, despite the large, negative shocks, the firms economy, the authorities face the very complicated prob- have reacted strongly and positively to the Big Bang. The lem of stopping bad credit and simultaneously making authors identify positive responses in the areas of market- sure that potentially profitable activities have access to fi- ing-distribution, cost-cutting and finance. They also note nancing (to "good" credit). Calvo argues that the lack of that the firms with previous exposure to Western markets information and institutions precludes the expansion of were in the best position to undertake the needed adjust- good credit in reforming socialist economies. Simple ments in their operations. However, the authors con- tight monetary policies, while helping cut bad credit, can- clude-in agreement with Frydman and Wellisz-that if not solve the problem of the "efficiency" of credit. Calvo the issue of ownership and control of enterpises is not re- concludes that the development of a banking system solved, the adjustment may stall, despite the willingness should have priority in the reform. He suggests that if of firms to modify their behavior. monetary/credit policy relies too heavily on a credit In "Distortionary Policies and Growth in Socialist crunch, inflation may be slowed at the cost of equally, if Economies," William Easterly draws on the insights of not more disturbing economic problems, such as deep the new growth theory to analyze the interplay between and protracted recession. distortions and growth in Central and Eastern Europe. In "Adjustment, Trade Reform and Competitiveness: Using this framework, he presents a model that analyzes The Polish Experience," Jan Bossak reviews some of the how distortionary policies can affect growth, and not just trade policy issues in the reform. He argues that liberal- the level of output. He shows that the relation between ization per se may not be sufficient to dismantle the high- distortions and growth is very non-linear, the implication ly monopolistic structure of industries in socialist being that in highly distorted situations-such as those economies. He also suggests that the technological and in most socialist economies-small reductions in distor- quality gap between East and West is so large that some tions can have negative effects on growth. An important form of stimulus for specific activities should be intro- insight of this work is that, to be effective, reforms must duced. reduce distortions below a critical level. Thus, major re- In "Fiscal Issues in Economies in Transition," Vito ductions in large distortions can have important positive Tanzi gives an overview of the main aspects of fiscal policy effects on growth. Easterly presents an extension of the associated with adjustment programs supported by the model that illustrates the effects that the misallocation of International Monetary Fund (IMF). In connection with capital has had on growth in the Soviet economy. This ex- Central and Eastern Europe, Tanzi emphasizes the prob- tension is used to show that a partial privatization can be lem of institution-building. He stresses that fiscal reform harmful for growth, whereas a close to complete or com- will be a lengthy and effort-intensive undertaking, as the plete privatization of the economy can benefit growth. basic fiscal institutions (such as tax and budgetary sys- tems) are either non-existent or very primitive. Part VI. Adjustment and Reforms: Financial Market, Foreign Trade, Fiscal Sector Part VII. Panel on Privatization Part VI includes three papers that address issues relat- Privatization has become the Achilles' heel of most of ing to reform of financial markets, foreign trade and fiscal the reform efforts in Central and Eastern Europe. Part VII areas. They look at the role of markets and of their cre- contains four papers, three of which include proposals for 4 privatization in Central and Eastern Europe, while the es in Poland," Manuel Hinds shares the philosophy of the fourth provides a comparative assessment of the three other two papers, stressing the importance of a free distri- proposals. Although they focus specifically on the Polish bution of shares for both equity and macroeconomic con- case, the papers offer a general framework for privatiza- siderations (the lack of domestic savings) and the tion in reforming socialist countries. All three papers ar- importance of rapid privatization. Some of his recom- gue for rapid and large-scale privatization. They also mendations, however, differ from those put forward in the share the idea that, because of a lack of financial markets, other two papers. In particular, he strongly opposes dif- and the low level of savings, and for political reasons, ferent voting rights on shares (which Lipton and Sachs shares in firms should be distributed freely. Finally, they propose for the shares held by the treasury) and the hold- call in common for the creation of financial intermediar- ing of shares by commercial banks. Hinds also suggests a ies as an intermediate layer between shareholders and clear separation between firms and banks and the cre- management. These intermediaries would facilitate the ation of capital markets to channel funds to the enterpris- control and effective management of the enterprises. es. In contrast, the other two papers envisage the Notwithstanding these common elements, the three development of a financial sector in parallel with privati- papers differ in important aspects. The main difference zation and having the financial intermediaries, including relates to the role to be played by the state during the the banks, gain control of the enterprises. transition period. In his comments on the above three papers, Borensz- In "Privatization in Eastern Europe: The Case of Po- tein highlights their main points of agreement. He also land," David Lipton and Jeffrey Sachs present a detailed highlights the points of disagreement, which mainly con- privatization plan, including a timetable. Based on a re- cern the role of the state and the issue of alternative mod- view of the international experience with privatization, els of corporate governance. the authors argue that a firm-by-firm approach would fail well before a substantial number of firms could be priva- Part VIII. Summing up and Overview tized. They offer a plan for rapid and massive privatiza- tion, although they caution that it will not produce rapid This section presents the comments of three panelists efficiency, productivity and managerial gains. Their plan whose role was to summarize the main findings of the has three stages. First, firms will "corporatize," i.e., be conference and suggest some directions for further re- transformed into treasury-owned joint stock companies. search. Subsequently the treasury will distribute a large compo- From the conference discussions, Stanislaw Gomulka nent of the shares to workers, the population at large and derives four phases for a typical transition. Phase one cen- financial intermediaries such as mutual funds, pension ters on macrostabilization, phase two on structural ad- funds and commercial banks, while remaining temporari- justment, phase three on the recovery of growth and ly the single largest shareholder. Finally, the treasury will phase four on sustaining growth and macroeconomic bal- sell its shares to "core investors." This plan combines fea- ance. Much of the conference discussions centered tures of the free distribution of shares with the sale of around phases one and two. In this regard, Gomulka em- shares contemplated in case-by-case schemes, and it ad- phasizes that the paths to transition are distinguished by dresses simultaneously the issue of the distribution of two main elements: the speed of price liberalization, in- wealth and the control of enterprises. cluding current account convertibility; and the speed and In "Markets and Institutions in Large-Scale Privatiza- scope of privatization. Based on the Polish experience, he tion: An Approach to Economic and Social Transforma- argues that both price liberalization and privatization tion in Eastern Europe," Roman Frydman and Andrzej should proceed very fast and notes that while rapid price Rapaczynski present a view different from that of Lipton liberalization is likely to increase the short-term costs of and Sachs mainly with respect to the role to be played by transition, fast privatization could bring about positive ef- the state. Frydman and Rapaczynski call for a drastic re- fects in the short run. duction in the role of the state, with the mutual funds ful- Johannes Linn tries to identify some of the answers ly controlled by private investors. This approach relies that emerged in the conference to the main questions re- more on privatization which would take place simulta- garding the timing and sequencing of reforms. He also neously with the creation of a private sector. It stresses emphasizes that in recent months there has been an im- the creation of the markets and institutions that are nec- portant clarification of the ultimate goals of the reforms, essary for movement toward a market economy dominat- especially as regards private ownership. Linn argues that ed by the private sector. The authors do not believe the some ambiguity is still present on the type of labor mar- state can play an active role in this process. ket these countries are aiming for. Another point relates In "A Note on the Privatization of Socialized Enterpris- to measuring and monitoring the impact and progress of 5 reforms using statististical information. As the reforms could make or break the whole process of economic most likely push activity out of the measured, mainly so- transformation in Central and Eastern Europe. cialized, sectors, the official statistics tend to overstate The type of economic transformation being attempted their costs. by the reforming socialist countries does not have prece- Jan Svejnar urges that the link between microecomic dents in recent economic history. Therefore, the road to adjustment and macroeconomic stabilization, addressed reform is unknown. The reforms that have been attempt- in some of the presentations, be stressed more forcefully. ed in developing countries have been on a much reduced He emphasizes three main areas as deserving special at- scale. Nonetheless, there are important lessons to be tention: capital market imperfections; inter-enterprise learned from the experience of these countries. Careful credit; and labor market institutions and distortions. Sve- analysis of these experiences could provide important in- jnar argues that in designing macroeconomic policies, at- sights, as well as guidelines, for the authorities who are tention should be paid to those three areas. He notes that directing the reform effort in the socialist countries of bottlenecks in the banking system are indeed a major Central and Eastern Europe. constraint on growth and the development of a private sector. As to inter-enterprise credit, it tends to weaken References considerably the effectiveness of monetary policy and to increase the risks of large-scale insolvency. Finally, on la- World Bank. 1988. Adjustment Lending: An Evaluation of Ten Years of bor market policies, Svejnar emphasizes that consider- Experience. Policy and Research Series No. 1. Washington, D.C.: ations of efficiency should lead to a predominance of wage World Bank. incentives over wage controls. He concludes that the la-1 A n g s th bor market is often treated mechanically at the macro lev- . 199o. Adjustment Lending Policies for Sustainable Crowth, eor howkever its ofuenctioning mechanincalu its institron lPolicy and Research Series No. 14. Washington, D.C.: World Bank. el; however, its functioning, including its institutions, 6 Part I Central and Eastern Europe: An Historical Background The Economies of Central and Eastern Europe: An Historical and International Perspective Andres Solimano' Current events in Central and Eastern Europe are so forms in the 1940s and the origin and stabilization of far-reaching that a look at the economic (and political) Hungary's hyper-inflation of 1945-46 are discussed. The history of the region is critical to understanding the section then jumps to the late 1980s and early 1990s, changes fully. Extreme macroeconomic instability, eco- when socialism throughout the region collapsed. Some nomic reconstruction and structural transformation, in- of the initial conditions with which the Central and East- tegration into a global monetary system, the domestic ern European countries began their transition to a mar- consequences of changes in international conditions- ket economy are noted, along with some historical and all are phenomena that Central and Eastern Europe had international comparisons of per capita income in the experienced before socialism was established after the region. The last section offers concluding observations. Second World War. Thus, a study of pre-socialist Central and Eastern Europe may provide valuable clues to the Central and Eastern Europe Before Socialism: current problems of the transition to a market (or capi- The Inter-war Period talist) system. This transition constitutes, in a sense, a rendez-vous of the region with its own history (includ- The end of World War I led to the collapse of the four ing the problems). main monarchies-German, Habsburg, Ottoman and This paper is organized as follows. The next section Czarist-that dominated Central and Eastern Europe. In focuses on the 1920s and 1930s, documenting the for- the case of the former Austro-Hungarian Empire, it gave mation of the nations in what is today (subject to some birth to three independent states: Austria, Czechoslova- territorial changes after the Second World War) Central kia and Hungary. The former Hungarian kingdom, how- and Eastern Europe. The problems of macroeconomic ever, lost nearly two-thirds of its territory (table 2-1). stabilization and the issues of economic recovery, inte- er,ost arly thirds of Serrit abl 2-a). gration into the gold exchange standard and the tenden- cy toward protectionism that developed in the second Table 2-1. Central and Eastern Europe Before and half of the 1920s are reviewed, as are the impact of the After World War I Great Depression, the collapse of the gold exchange Area Population standard and the policy response of most Central and (in square hms) (millions) Eastern European economies to those events of the thir- 1914 1921 1914 1921 ties. The effect of German re-armament on the econo- Austro-Hungarian mies of Central and Eastern Europe in the second half of Monarchy 676,433 51,390 the 1930s is also addressed. Finally, the section describes Austria 85,533 6,536 the levels of per capita income and the main features of Hungary 92,607 7,600 the economic structure in the region in the late 1930s. Czechoslovakia 140,394 13,613 The third section begins with a look at the late 1940s. Bulgaria 111,800 103,146 4,753 4,910 It analyzes the establishment of socialism in the second Rumania 137,903 304,244 7,516 17,594 half of the 1940s in the region, including the policies of Serbia 87,300 4,548 nationalization and central planning that were the basis Yugoslavia 248,987 12,017 for a socialist (Soviet-style) economy. From a macroeco- Poland 388,279 27,184 nomic perspective, the implementation of monetary re- Source: Berend and Ranki (1974). 9 A reborn Poland (the new Polish kingdom) received ter- the countries of the region. An example was the estab- ritories from the Austro-Hungarian monarchy, Germany lishment of a soviet republic in Hungary in March 1919 and Russia. Bulgaria underwent some minor territorial led by Bela Kun, which was followed by a right-wing gov- changes, and the kingdom of Romania annexed territo- ernment. Conflicts arose over the distribution of income ries from Hungary and Russia.2 as workers sought to raise the level of real wages, which All the above countries suffered significant destruc- had been very depressed during the war. However, exter- tion of their physical capital and infrastructure and siz- nal conditions left little room to accommodate these de- able losses in human capital during the war of 1914-18. mands for higher real wages at home: the reparations Thus, economic reconstruction became the first priority. payments and ensuing external transfers required an un- Poland was the most badly affected by the war, while dervalued real exchange rate. Czechoslovakia was the least affected by the destruction. The war reparations sanctioned in the Treaties of Ver- sailles, St. Germain and Trianon adversely affected the The 1920s countries born of the Austro-Hungarian monarchy, as An important economic consequence of the war was well as Bulgaria. They put a severe burden on public fi- the total disorganization of public finances and financial nances and the balance of payments. The results of the systems in most of the region. That state of chaos rapidly payments of the war reparations and large fiscal deficits led to extreme macroeconomic instability and hyper-in- financed by printing money (table 2-2) were extreme in- flation, notably in Austria, Poland and Hungary. The flation (table 2-3) and a massive depreciation of the ex- emergence of macroeconomic imbalances and high in- change rate in Austria, Hungary and Poland.3 flation in the early 1920s was connected to the financing of the war and its aftermath. Besides taxation and the is- Table 2-2. Fiscal Budgets and Their Financing after suance of debt, to a great extent World War I was n- World War I nanced by printing money. The impact of the money creation on prices was restrained temporarily by the ex- Percentage of government tensive price controls and rationing of consumption expenditures covered goods (the military had priority in the allocation of re- by issues of paper money sources). Thus, by the end of the war the population had Hungary accumulated large involuntary savings in the form of liq- 1920-21 47.9 uid assets, and this monetary overhang certainly con- 1921-22 24.1 tributed to the outbreak of inflation. Moreover, public 1922-23 21.0 finances were strained by the economic reconstruction, 1923-24 34.4 which required increased government spending at a Austria time when fiscal revenues were scarce, particularly in Auaria those countries that had lost assets at the outset of the July 1, 1919-June 30, 1920 6370 conflict. July 1, 1920-June 30,1921 58.0 Socio-political turbulence compounded the macro- January 1-December 31, 1922 40.0 economic instability. A succession of short-lived, ideo- logically diverse governments ruled during 1918-19 in Source:Sargent(1982). Table 2-3. Inflation in Central and Eastern Europe, 1921-24 Austria Hungary Poland Czechoslovakia (price index, (price index, (price index, (price index January 1921=100) July 1921=100) January 1921=100) January 1922 =100) 1921 December 942 196.4 226.9 1922 December 17,409 795.2 1,377.7 59.6 1923 December 21,849 17,000.0 566,055.5 58.8 1924 March 23,336 49,445.24 975,686.8 60.9 April 23,361 50,823.81 963,927.7 60.2 June 24,267 52,566.6 September 53,252.3 Source: Prepared on the basis of Sargent (1982). 10 The hyper-inflation in Austria, Hungary and Poland avoid a period of extreme inflation in the aftermath of in the early 1920s shared some common features: com- the First World War. As shown in table 2-3, it experi- plete demonetization and shifts of the portfolios of do- enced deflation in 1922-23 (matching that of dollar mestic residents toward foreign currency and gold; prices) following the implementation of conservative dramatic increases in the fiscal deficits (partly endoge- fiscal and monetary policies. Czechoslovakia's economic nously generated by the reduction in real tax collections reconstruction went ahead without the detour of its induced by the inflation); very rapid depreciation of the first having to stabilize as its neighbors did. exchange rate; and the complete destruction of the con- The second half of the 1920s brought relative pros- tract-structure for goods, labor and financial transac- perity to Central and Eastern Europe. With the consoli- tions (see Dornbusch and Fischer 1986; Solimano dation of stabilization, output started to grow, a trend 1990a; and Wicker 1986). that lasted until 1929 with the onset of the world Depres- The stabilization of those hyper-inflationary episodes sion. Data on the national product for that period show illustrates, vividly, the interplay between foreign assis- Czechoslovakia's economy starting to grow in the early tance and domestic fiscal and monetary reform in stop- 1920s, avoiding the slump experienced by the countries ping extreme inflation. In Austria and Hungary, the affected by hyper-inflation in that period (table 2.4). stabilization was framed within a League of Nations re- Two external factors fueled the recovery of growth in construction plan (also signed by Czechoslovakia) that the second half of the 1920s-favorable agricultural included two protocols: the first established the political and territorial sovereignty of these countries; and the Table 2-4. Economic Growth in Central and East- second set out a plan for economic stabilization that em Europe in the Inter-war Period, as Measured tied the granting of an international loan to the adop- by National Producta tion of fundamental reforms in the fiscal and monetary (annual growth rate in percent) areas. Two key conditions were incorporated that the Czechoslovakia Hungary Yugoslavia Bulgaria countries complied with. First, they were to institute 1920-24 4.7 - 3.4 - independent central banks (these banks were to replace 1925-29 3.7 5.4 3.7 2.3 the former Austro-Hungarian Bank with its two sec- 1930-33 -2.9 -2.4 -2.1 0.0 tions, the Austrian and the Hungarian). The main provi- 1934-39 4.5b. 2.5 4.0 7.5 so in the charters of the newly created central banks-- so in the charters of the newly created central banks a. For Czechoslovakia and Yugoslavia, gross domestic product; for Hungary was a prohibition on the discounting of treasury bills, a and Bulgaria, net national product. practice used so extensively that it had led to the hyper- b. 1934-37. inflation. Second, the budget was to be balanced Source: Elaborated on the basis of Mitchell (1975). through a combination of reduced expenditures and in- prices and foreign capital inflows. The favorable agricul- creased taxation. tural prices benefitted not only the agrarian countries of The League of Nations plan also lowered and clarified the Balkans, but also Hungary and Poland. The major the amount and timing of the reparations payments inflow of capital, including direct investment, came owed to the Reparations Commission. In addition, it lift- from of West, ingludind the itentate ed the lien on state assets imposed by the original treaty. from the West, with England and the United States the Those measures were undoubtedly a stabilizing factor, as most important sources. they reduced current and expected fiscal obligations Although the economies of Central and Eastern Eu- abroad.4 The result in Austria and Hungary of the plan rope maintained their trade with the rest of Europe and was an end to inflation and the stabilization of their cur- overseas, a general tendency toward protectionism de- rencies in 1924. Poland managed to eliminate the hyper- veloped in the region in the mid-1920s. In Hungary, av- inflation and stabilize its currency through policies sim- erage tariffs were raised from 20 percent, the average ilar to those followed by Austria and Hungary, namely, level prevailing under the monarchy, to 30 percent by stabilization of the exchange rate, creation of an inde- 1925. Romania raised its import duties from 30 percent pendent central bank and balancing of the budget. How- before the war to almost 40 percent by 1924. Bulgaria ever, unlike in Austria and Hungary, it achieved the hiked its tariff levels to prohibitive levels: for some activ- stabilization without a League of Nations loan. Later on, ities, they reached a level 100-300 percent higher than in 1927, when instability in the foreign exchange market that before the war (Berend and Ranki 1974, chapter 9). started to develop, Poland arranged for a large foreign Most countries in Central and Eastern Europe based loan with the help of the League. their exchange rate system on the gold exchange stan- Czechoslovakia, formed in 1918, was the only coun- dard, as proposed in the Genoa conference of 1922 (Nur- try in Central and Eastern Europe that managed to ske 1944 and Kindleberger 1984). This system allowed 11 the central banks (reformed with the assistance of the rency-the US dollar-also ceased to be eligible for re- League of Nations, as noted) to hold their reserves partly serve purposes. in the form of foreign exchange-instead of gold- against notes in circulation and sight deposits. The US The 1930s andAfter dollar and the pound sterling served as reserve curren- cies. The system worked relatively well for a while, pre- The economies of Central and Eastern Europe were venting the deflationary tendencies that a "full" gold not immune to the world Depression triggered in 1929. standard could have generated given the relative scarcity These countries suffered a fall in world demand for their of gold at the time. Nevertheless, toward the late 1920s, exports, a deterioration in their terms of trade and a cut- following the events in France in 1928, the system came off of foreign lending. Agriculture-exporting countries under pressure.5 (Table 2-5 provides details on exchange such as Bulgaria, Rumania and Yugoslavia were badly rates and prices.) hurt by the drop in agricultural world prices, particular- The response to the loss in confidence in the gold ex- ly for wheat and corn (table 2-6), a shock that had both change standard in countries such as Poland, Czechoslo- adverse income and balance-of-payments effects. The vakia and Bulgaria was the transfer of their reserve semi-agrarian countries of Hungary and Poland also suf- balances from London to Paris when a run against the fered directly from the drop in world agricultural prices, pound developed. After April 1933, the other reserve cur- as well as indirectly from the squeeze on industrial de- Table 2-5. Exchange Rates and Prices in the Inter-war Period (dollar and cents per unit of local currency) Czechoslovakia Hungary Poland Yugoslavia Romania Bulgaria Exchange Price Exchange Price Exchange Price Exchange Price Exchange Price Exchange Price rate level rate level rate level rate level rate level rate level (Crouwn) (1929=100) (Pengo) (1929=100) (Zloty) (1929=100) (Dinar) (1929=100) (Leu) (1929=100) (Lev) (1929=100) 1925 2.97 109 0.0017a 116 17.74 n.a. 1.70 n.a. 0.48 n.a. 0.73 95 1926 2.96 103 17.56 102 11.18 91 1.76 101 0.46 n.a. 0.72 87 1927 2.96 106 17.47 109 11.29 103 1.76 104 0.60 n.a. 0.72 88 1928 2.96 106 17.44 112 11.21 104 1.76 107 0.60 n.a. 0.72 94 1929 2.96 100 17.44 100 11.19 100 1.76 100 0.60 100 0.72 100 1930 2.96 89 17.49 87 11.21 89 1.76 86 0.60 78 0.72 82 1931 2.96 81 17.45 82 11.20 78 1.76 73 0.60 60 0.72 67 1932 2.96 74 17.45 82 11.18 68 1.64 65 0.60 54 0.72 59 1933 3.82 72 22.36 71 14.41 61 1.76 64 0.78 52 1.00 53 1934 4.24 74 29.57 71 18.85 58 2.27 63 1.00 52 1.29 54 1935 4.16 77 29.6 78 18.88 55 2.28 66 0.93 60 1.30 55 1936 4.01 77 19.78 80 18.87 56 2.29 68 0.74 69 1.30 56 1937 3.49 82 19.78 86 18.92 62 2.30 74 0.73 78 1.29 63 1938 3.47 81 19.73 87 18.86 58 2.31 78 0.73 78 1.24 65 1939 3.42 93 19.24 86 18.84 57 n.a. 79 0.71 88 1.21 66 n.a. Not available. a. The Pengo was introduced in 1925 (I Pengo=12,500 Crowns). Source: Svennilson (1954). Table 2-6. Price Index of Grains, Central and Eastern Europe, 1930-33 (1929=100) Bulgaria Hungary Yugoslavia Romania Wheat Corn Wheat Corn Wheat Com Wheat Corn 1930 87 59 55 31 63 46 47 51 1931 72 63 57 29 39 34 45 46 1932 77 59 67 29 54 30 34 37 1933 62 32 44 26 63 26 34 25 Source: Berend and Ranki (1974). 12 mand associated with the decline in real incomes in the and Eastern Europe had enjoyed under the system based agricultural sector. The more industrialized Czechoslo- on the gold exchange standard. vakia (and Austria) experienced a cut in production in Following the example of Germany, the economies of industry, in particular in the capital goods-producing Central and Eastern Europe initially abstained from a sectors during the great Depression. devaluation when facing the adverse external shocks of National product and the level of prices fell in the ear- the early 1930s. To counteract a loss in external compet- ly 1930s (tables 2-4 and 2-5), particularly in the period itiveness in most of these countries, the governments in- 1930 through 1933. The bankruptcies of German and stituted a "bonus system." The bonus combined a Austrian banks in 1931 compounded the financial reper- subsidy for exporters with a tax on imports, which was cussions of the crisis in real economic activity. The at- aimed at increasing the relative prices of traded goods tempts to continue servicing the short-term credits (with respect to home goods) without resorting to a from abroad caused large losses in their holdings of in- nominal devaluation.6 ternational reserves and gold. Hungary had to rescue Moreover, in 1933, Czechoslovakia, Hungary and Po- major banks from bankruptcy; in Yugoslavia, total de- land all raised their import duties. Those countries posits shrunk by half between 1930 and 1934, a problem maintained the tariff barriers throughout the thirties that caused several financial institutions to go bankrupt. and then intensified them in the Second World War with In Bulgaria the crisis in the banking system showed up the imposition of additional import controls. in the failure of the important Banque de Sofia and the Hungary suspended the payment of its external debt, merger of several other banks. both amortization and interest, between 1931 and 1933. In contrast with the other countries of the region, Romania, Yugoslavia and Bulgaria partially suspended Czechoslovakia came through the financial crisis of the foreign debt service even after the worst of the world De- early thirties in much better shape than its neighbors pression had passed. In contrast, Czechoslovakia did not did. To a large extent, the reason was that at the time of need to suspend its foreign debt obligations. Poland is a the adverse shocks in the early 1930s it had a relatively special case. Its government did not adopt restrictive low level of external debt. As table 2-7 shows, in the policies during the years of the Depression; it tried to 1930s Czechoslovakia's external debt (as a share of na- maintain the gold standard parity of the zloty and paid tional income) was roughly a third that of the other the amortization when due. However, this policy proved countries in Central and Eastern Europe. As such, unsustainable, and in 1936 the government had to re- Czechoslovakia was less vulnerable to the increases in schedule its payments obligations abroad, and it im- foreign interest rates and/or the cut-off in foreign lend- posed controls over foreign exchange. ing. An important external factor that shaped develop- The policy response to the Depression of the 1930s in ments in Central and Eastern Europe in the second half the countries being studied here included the imposi- of the thirties was the establishment in 1933, a period of tion of foreign exchange controls and limited currency peace, of a "war economy" in Nazi Germany. The aim was convertibility. In addition, a system of administrative to accelerate growth and secure agricultural products controls and approvals was initiated for foreign trade. and raw materials to support the re-armament. The That system dramatically changed the relative freedom countries of Central Europe and the Balkans initially of trade that had prevailed in the 1920s (although, as welcomed the measures, for they were suffering from de- noted, tariffs were not low in that period) and reduced pressed world demand for their products. Moreover, Ger- the degree of capital mobility the economies of Central man trade policies after 1933 were conducted on a Table 2-7. Indicators of Foreign Debt in Central and Eastern Europe, 1930 Hungary Czechoslovakia Yugoslavia Romania Foreign debt per capita 65.0 34.0 66.0 75.0 (US dollars) National income per 116.0 171.0 93.0 107.0 capita (US dollars) Foreign debt as a share 56.0 19.9 70.1 70.1 of national income (percent) Source: Berend and Ranki (1974). 13 bilateral basis, with the direct exchange of goods-bar- of Bulgaria, the poorest country in Central and Eastern ter trade-rather than the use of international means of Europe. Moreover, average per capita income in Central payments, which were scarce in Germany and in the oth- and Eastern Europe was around one-third the level in er countries of the region. Several countries of Central Western Europe. and Eastern Europe increased their trade with Germany With the exception of Czechoslovakia and Austria, substantially toward the second ha]f of the 1930s. That most of the countries of Central and Eastern Europe had trend was particularly noteworthy in Hungary, Yugosla- a predominantly agrarian economic structure by the late via and Bulgaria (table 2-8). 1930s (table 2-10). The Balkan countries (on average) In terms of the level of national per capita income, the generated nearly 55 percent of their national income structure of economic activity, and foreign trade in the from agriculture (63.3 percent in Bulgaria), Hungary region toward the late 1930s, table 2-9 shows the relative and Poland around 37 percent on average in 1938. Agri- backwardness of the countries of Central and Eastern culture in those countries was relatively backward, with Europe with respect to Western Europe. Austria and low growth in productivity, a structure of land tenure in Czechoslovakia had the highest per capita income in which tiny plots owned by peasants coexisted alongside Central Europe, but their average was still only around large unexploited farms owned by relatively wealthy 40 percent that of Great Britain, the most developed na- landlords. Surplus labor was another indicator of the low tion in Western Europe at the time. In turn, Great Brit- level of agricultural development. ain's national per capita income was nearly six times that The share of manufacturing in income was just over 53 percent in Czechoslovakia, close to 34 percent in Table 2-8. Composition of the Rhade of Central and Hungary and Poland and around 23 percent in Yugosla- Eastern European Countries with Germany, 1929 via, Romania and Bulgaria. In Czechoslovakia, the capi- and 1937 tal goods-producing sector, chemicals and other heavy Exports to Germany Imports from Germany branches dominated industry. In contrast, in the less in- percentage of total percentage of total dustrialized countries of the region, the manufacturing 1929 1937 1929 1937 sector centered around food processing, textiles and oth- Hungary 11.7 24.1 20.0 26.2 er light activities. Yugoslavia 8.5 21.7 15.6 32.4 The structure of foreign trade in these countries was Romania 27.6 19.2 24.1 28.9 a reflection of their patterns of development and special- Bulgaria 29.9 43.1 22.2 54.8 ization (table 2.11). In most, the bulk of exports consist- Source: Berend and Ranki (1974). ed of raw materials and food, whereas their imports entailed mainly manufactured goods and intermediate inputs. Czechoslovakia, however, was a net exporter of Table 2-9. Per Capita9National3Income7in Europe, manufacturing goods and a net importer of raw materi- a]s and food. (US dollars) As noted, Germany's drive toward war directly affect- Western Europe ed the economies (and politics) of Central and Eastern Great Britain 440 Europe. Here a distinction should be made among: the Sweden 400 group of countries that was (at least) partly annexed in Belgiurn 330 1938-39, namely, Czechoslovakia and Austria, or the so- Holland 306 called "Czech-Moravian Protectorate" ; the 'satellite" Holance 3065 countries-Hungary, Rumania and Bulgaria; and the France 265 cutisHnay uai n ugra n h group of countries that Germany occupied and physical- Central and Eastern Europe ly devastated-Poland and Yugoslavia. The members of Austria 190 the Czech-Moravian Protectorate, given its relatively ad- Czechoslovakia 170 vanced stage of industrialization and financial develop- Hungary 120 ment, were integrated into the German economy under Poland 100 a system of centralized state control typical of a war Romania 81 economy. Moreover, German intervention in the "pro- Yugoslavia 80 tectorate" entailed some confiscation of property, in- Bulgaria 75 cluding industry and bank assets. Source: United Nations (1949). The war economy did not involve open occupation of Hungary, Romania and Bulgaria, although German con- trol was effective. These countries supplied food, agricul- 14 Table 2-10. Structure of National Income by Economic Sectors of Selected Central and Eastern European Countries, 1938 (percentages) Agriculture Industry Construction Total and services Czechoslovakia 23.2 53.2 23.6 100.0 Hungary 36.5 35.7 27.8 100.0 Poland 39.0 32.2 28.8 100.0 Romania 53.2 28.4 18.4 100.0 Yugoslavia 53.6 22.1 24.3 100.0 Bulgaria 63.3 18.3 18.4 100.0 Source: Berend and Ranki (1974i. tural products in general and raw materials such as iron, pied and split up their territories. In addition, parts of bauxite and timber. German involvement in these coun- Yugoslavia were annexed to Hungary and Bulgaria. The tries also had some important macroeconomic conse- initial strategy of the Axis countries with respect to Po- quences. As Germany accumulated debts with these land and Yugoslavia was outright annihilation through countries for unpaid supplies, their national govern- the destruction of their communications systems, phys- ments printed bank notes to compensate domestic pro- ical infrastructure and industrial capacity. However, lat- ducers for the unpaid orders. Thus, with the German er that policy changed, as Germany made deliveries of deliveries unpaid, a monetary overhang developed. The food and raw materials the priority. governments repressed the inflationary impact of the re- By the end of World War II most of the countries of sulting excessive liquidity by rationing goods. On the Central and Eastern Europe had suffered enormous loss- supply side, they enacted a system of compulsory deliv- es. Poland and Yugoslavia lost an important part of their eries to the state of agricultural goods and raw materials population, and their railway systems were severely based on fixed quotas that they strictly enforced. damaged, as were the roads, infrastructure and commu- The third group, comprised of Poland and Yugoslavia, nications in general. The capital stock of industry and was the hardest hit during the war. Both lost their inde- agriculture was partly destroyed, with an obvious ad- pendence, as Germany, Italy and the Soviet Union occu- verse effect on productive capacity.7 Table 2-11. Structure of Foreign Trade in Central and Eastern Europe, 1938 (percentages of countries' total trade) Czechoslovakia Hungary Poland Yugoslavia Romania Bulgaria Exports Manlufactured 71.8 13.0 6.4 0.8 1.9 2.0 goods Raw materials 19.8 31.7 65.1 49.5 64.3 66.6 and intermediate goods Foodstuffs 8.4 55.3 28.5 49.7 33.8 31.4 ioo.o 100.0 100.0 10.0 10.0 o100.0 Imports Manufactured 29.6 30.2 28.4 44.8 68.3 68.0 goods Raw materials 57.5 61.5 54.1 50.1 27.3 31.5 and intermediate goods Foodstuffs 12.9 8.3 17.5 5.1 4.5 0.5 100.0 100.0 100.0 100.0 100.0 100.0 Source: Spulber (1957). 15 The Establishment of Socialism in the Second Half of The nationalization laws contemplated compensation the 1940s of the former owners of the enterprises through treasury bills, payments of money or paper issued by newly creat- The end of the Second World War brought far-reach- ed funds.9 ing political and economic changes to Central and East- The amount of actual compensation is hard to deter- ern Europe. Territories in Poland, Czechoslovakia, mine, but the widespread presumption is that it amount- Rumania and Bulgaria were redistributed.8 ed to very little. The nationalizations spurred the The Soviet Union gained territories at the expense of merging and consolidation of nationalized enterprises, a these countries, while Poland got territories from Ger- process that created monopolistic market structures. A many as retribution. As could be expected, all these few units ended up producing the total supply of given changes altered the distribution of agricultural land, goods. The nationalizations also encouraged vertical in- natural resources and industrial capital stock in the re- tegration and required centralized decision-making. gion. Poland and Czechoslovakia underwent profound Therefore central planning and full state ownership be- changes, as the alteration of their national boundaries came closely intertwined. was of a sweeping nature. In addition, the Soviet Union The nationalization of the banking and financial sys- specified in the peace treaties how German assets in the tems was facilitated in Czechoslovakia, Poland and Yu- former satellite countries of Germany (Hungary, Roma- goslavia by the fact that the major financial institutions nia and Bulgaria) were to be disposed of. At the political had no private property after the war. In general, merg- level, the governments emerging after the Second World ing and centralization of the banking system into state War were formed by broad political coalitions ranging national banks took place in Poland, Hungary, Romania, from liberal to Communist. The dynamics of the politi- Yugoslavia and Bulgaria in the late 1940s. These banks cal situation were such that the Communist parties performed several roles: they had a monopoly over the managed to acquire, in various cases through coups d'e- issuance of money and therefore acted as central banks; tat, dominant positions in the governments after 1948. they regulated the financial system; and they provided In keeping with the new political circumstances, the short-term credit to enterprises in the context of the economies of Central and Eastern Europe were reorga- planning system. Savings and investment banks (very of- nized along the lines of the Soviet economic system. The ten just branches of the state national bank) accepted first and most important move was the nationalization of savings from the population and channeled them as fi- industry, the financial system, and the trade and service nancing for investment. sectors, along with agrarian reform, the latter being an The monetary history of the region after World War II old aspiration of the political left and the peasantry. In resembled in several ways that of the post-World War I Czechoslovakia, before the Communist coup d'etat of period. Because World War II had been financed by the February 1948, the nationalization laws had brought printing of money (along with rationing), there was a nearly 57 percent of the total industrial labor force into monetary overhang when it ended. Moreover, the recon- the state sector. After 1948 that process was accelerated, struction and the need to recapitalize and expand the na- and by 1949 the private sector accounted for just 3 per- tionalized sector after 1945 led to a rapid expansion of cent of total employment in manufacturing. In Poland, the money supply in several Central and Eastern Euro- in 1946 the state sector accounted for nearly 85 percent pean economies (see table 2-12). of total employment in the manufacturing sector for From another angle, the accumulation of excessive li- units with over five workers. Industry was almost fully quidity originated in an increase in the wage bill that was nationalized by the early 1950s. In Hungary, where na- overly high relative to the output of consumer goods at tionalization had been relatively moderate early on, after fixed prices (the phenomenon of too high wages chasing 1948 the process was intensified, encompassing medi- too few goods) (table 2-13). The resulting shortages in um- and large-scale firms in manufacturing. With some Table 2-12. Currency in Circulation in Czechoslova- specific differences, the same process occurred in R *ma- Tia, Poland and Yugoslavia, 1938, 1945 and 1948 nia and Bulgaria. k1938=9, 44 By 1949 the share of the socialized sector in gross (1938=100) manufacturing output ranged from 85 percent in Roma- Czechoslovakia Poland Yugoslavia nia to 92 percent in Hungary, 93 percent in Bulgaria, 94 1938 100 100 100 percent in Czechoslovakia and 100 percent in Yugoslavia 1945 254 1,878 231 (Spulber 1957, p. 83). Thus, the private sector practically 1948 908 12,400 509 disappeared from the productive sphere in a period of Source: Spulber (1957). around five years. 16 consumption goods led to an accumulation of involun- vorable than that for money (Ames 1954). Moreover, re- tary savings in the form of liquid assets. ducing the excess liquidity was not easy, and several As the excessive liquidity threatened to become an in- countries of the region had to implement more than one flationary factor, governments implemented monetary monetary reform in the period 1945-53. reforms to soak up the large monetary balances. The Hungary suffered a (classic) case of hyper-inflation monetary reforms involved at least one of three mea- during its transition to socialism. Moreover, its bout of sures: (1) reduction of the supply of liquid assets hyper-inflation was the second one it had faced after the through the compulsory exchange of "old" money for hyper-inflation of the early 1920s. The hyper-inflation of "new" money at a given rate of conversion; (2) immobi- 1945-46 was more intense than the former episode and lization of a share of the supply of liquid assets by block- was more extreme than even the German one of 1923. As ing access to some deposits in the banking system, with in other historical episodes of extreme inflation, Hunga- the blocked funds going into special accounts (the de- ry's was associated with the severe increase of the fiscal positors still owned the full deposits); and (3) the conver- deficit. As table 2-14 shows, during most of the hyper-in- sion of wages into a new unit at a given rate of exchange. flation, tax revenues financed less than 10 percent of fis- In several countries of the region, the monetary re- cal expenditures. An important element in the forms involved differentiated rates of conversion for deterioration of fiscal finances was the burden imposed bank notes vis-a-vis bank deposits, with a more favorable by the reparations payments Hungary agreed to in the rate of conversion for deposits (intended to tax specula- armistice of January 1945. It owed those reparations tors who often operated with currency). In addition, a mainly to the Soviet Union, but also to Czechoslovakia distinction was made based on the amount of liquid as- and Yugoslavia.'2 sets held by individuals and enterprises, with small hold- The costs of the reparations and occupation account- ers favored over large ones or private enterprises. ed for between 23-50 percent of public spending.13 The actual rates of conversion of money varied from The fiscal deficit was financed through treasury bills is- country to country. Under the Hungarian monetary re- sued by the Hungarian govemment and discounted by the form of December 1945, the rate was 4 to 1 old for new central bank. In addition, indexation of the money supply currency bank notes. In Bulgaria, it was 100 to 1 in the is considered to have been a critical contributor to the very reform of March 1947. In Rumania, the rates varied from high rates of inflation (table 2-15). 100 to 1 to 400 to 1 under the monetary reform of Janu- The Hungarian stabilization plan of August 1, 1946- ary 1952. Poland blocked bank deposits beginning in De- designed by the Marxist economist Eugene Varga and cember 1944, Czechoslovakia in October 1945 and backed by the Communist partyonsisted of the fol- Bulgaria in March 1947.11 The extent to which the monetary reforms were suc- lowing measures. In the area of monetary reform, the cessful in reducing the excess liquidity and inflationary Table 2-14. Public Finances in Hungary, 1945-46 pressures in Central and Eastern Europe is debatable, as, in general, the rate of conversion for wages was more fa- Percentage of Occupation and fiscal expenditures reparation costs as Table 2-13. Consumption Spending and Wages, financed by fiscal revenues percentage of fiscal expenditure 1948-51 1945 (1948=100) 1948 1949 1950 1951 July 6.8 23.0 August 5.3 34.0 Czechoslovakia September 7.3 23.0 Consumption 100 108 118 131 October 5.7 24.0 Wage bill 100 112 142 159 November 6.6 31.0 Hungary December 7.1 40.0 Consumption 100 122 134 152 Wage bill 100 131 169 206 1946 Poland Consumption 100 101 118 126 January 14.2 26.0 Wage bill 100 143 180 256 February 14.4 32.0 Romania March 13.0 39.0 Consumption 100 90 109 111 April 9.8 50.0 Wage bill 100 154 210 n.a. May 7.3 38.0 n.a. Not available. Source: Bomberger and Makinen (1983). Source: Ames (1954). 17 Table 2-15. Inflation in Hungary, 1945-June 1946 The Collapse of Socialism in Central and (price level, January 1945=100) Eastern Europe: The Late 1980s 1945 I 104 II 124 The seemingly irreversible socialist experiment in III 290 Central and Eastern Europe came to sudden, and largely IV 1,826 unexpected, end in the late 1980s. The collapse of Soviet- style socialism has both economic and political roots. 1946 I 4,442 Central planning led to endemic shortages, slow techni- IV 185x106 cal change, low quality goods and over-expanded public Note: I-IV refer to quarters of the year. sectors. In practice, full employment and social protec- Source: Spulber (1957). tion had been bought at the price of economic backward- ness and a lack of political freedom. Attempts at partial government introduced a new unit, the forint, and ini- reform in the 1980s often worsened the macroeconomic tially it imposed a 100 percent reserve requirement on conditions, as credit and wage policies were relaxed and commercial banks. The central bank was forbidden by fiscal budgets turned into deficits. In addition, several law from making direct or indirect loans to the govern- Central and Eastern European economies borrowed ment. Discounting of treasury bills was abolished, and heavily abroad, accumulating a large external debt that the only operation allowed between the treasury and the further complicated domestic economic management. central bank was lending to finance the acquisition of At the political level, more than four decades of one- gold or foreign currencies. In the area of fiscal reform, party rule and authoritarianism generated a (non-offi- the tax system was completely revamped, with the tax cial) domestic consensus on the need for a political rates on labor income and property going up, as did the opening-up and democracy. In 1989, emerging political sales tax. The government introduced a turnover tax on parties and social movements with an agenda of political enterprises and reduced fiscal expenditures by cutting transition to a Western-style democracy and the creation the number of civil servants and shrinking the size of the of a market economy toppled the authoritarian regimes army and police. With respect to external financing, the dominated by the Communist party (or its equivalent). plan entailed external support in the form of tied US Table 2-16 gives a broad picture of both the final state loans, increased food supplies granted by the United Na- of the socialist experiment and initial results of the tran- tions and an extended payment period for the debt owed sition toward a market economy. The image is that of de- to the Soviet Union. teriorated economic conditions, as reflected (among The plan succeeded. Hyper-inflation stopped rapidly: other indicators) in the lower growth and higher infla- from August 1 to December 1946, inflation rose by only tion in the second half of the 1980s. The macro instabil- 6 percent, and the rate of inflation in 1947 was approxi- ity was more acute in Yugoslavia and Poland, where near mately 19 percent. Stabilization followed the rapid mon- hyper-inflation developed in 1989. To eradicate the infla- etization (the increase in money demand after the tion and create a more favorable macro environment for stabilization was accommodated), and the stock of mon- a market-oriented economy, both countries adopted ey increased by 221 percent between August 31 and De- sharp anti-inflationary policies in early 1990. Their plans cember 31, 1946. During 1947, the nominal money included a large real depreciation of the exchange rate, stock grew by 132 percent (Bomberger and Makinen the deregulation of most controlled prices, tight mone- 1983). The effects of the stabilization on output and em- tary and credit policies, the correction of fiscal deficits ployment are hard to determine given the sketchy data. and a drastic lowering of import tariffs.14 However, a reduction in employment along with an in- In addition, both countries are trying to privatize crease in national product seem to have taken place dur- state-owned enterprises, albeit still with only limited ing 1946 and 1947. Politics took center stage after the success. achievement of stabilization: in June 1947 a Communist Hungary and Czechoslovakia are in better shape, al- coup d'etat forced the resignation of Prime Minister though inflation in the former is considered relatively Nagy. The Communist party took over practically all the high. Czechoslovakia, following a long history of macro ministries, and the state was reorganized in a way that stability and prudence (that goes back to its pre-socialist furthered the establishment of socialism. In the eco- period), had not experienced any upsurge in open infla- nomic sphere, nationalization and the imposition of cen- tion through the end of 1990, although acceleration in tral planning were accelerated. inflation cannot be discarded for the future.15 18 Table 2-16. Inflation and GDP Growth in Selected Central and Eastern European Countries, 1986-90 (percentages) Poland Yugoslavia Hungary Czechoslovakia Bulgaria GDP Infl. GDP Infl. GDP Infi. NMP' IOnf. GDP Inf/. growth ratea growth rate growth ratea growth ratea growth ratea 1986 1.5 18.0 3.4 89.9 1.5 5.3 2.6 0.50 4.2 1.3 1987 3.4 25.2 -0.5 120.8 3.4 8.6 2.1 0.09 6.1 0.07 1988 0.1 60.0 -1.0 194.1 0.1 15.7 2.3 0.09 2.6 2.3 1989 -1.0 700.0 -2.0 2,700.0 0.5 19.6 1.9 1.44 -1.9 4.4 1990 -12.0 249.0 -3.2 118.6 -6.5 30.0 -3.5 13.9 -12.0 100.0 Source: World Bank. a. Based on the Consumer Price Index. b. Net material product. The economic outlook for Central and Eastern Eu- development of the capital goods sector, to the clear det- rope is bleak. The results of 1990 reveal large-scale con- riment of the production of manufactured consumer traction in economic activity, particularly in Poland, goods, whose quality was poor.17 Bulgaria and Yugoslavia, along with stubborn inflation. These trends are worrisome as they may undermine pub- lic support for the reform process. Table 2-17. Structure of GDP in Selected Eastern The situation does not leave much room for optimism and Central European Countnies, 1988' given the monumental task of transforming a very obso- (percentages) lete economic system into a dynamic one. Currently, the Agriculture Industry Construction Total problems of structural transformation are compounded and services by the disintegration of Comecon (it will entail a big Czechoslovakia 6.4 49.6 34.0 100.0 terms of trade deterioration for Central and Eastern Eu- Hungary 14.0 37.0 49.0 100.0 rope vis-a-vis the Soviet Union) and the reluctance of the Poland 10.6 40.4 49.0 100.0 international capital markets to commit ample credit to Yugoslavia 14.0 49.0 37.0 100.0 the region until the reform is more consolidated. Bulgariab 12.6 58.5 28.9 100.0 In looking at the structure of production in Central Ind Easterng atroe instheuatue 1980s, procti n rucedhare a. The shares for Czechoslovakia and Bulgaria correspond to net material and Eastern Europe in the late 1980s, the reduced share product. of agriculture and the relatively high share of industry in b. The shares correspond to 1989. output (between 37-60 percent) are evident (table 2- 17).16 Table 2-1& Per Capita GDP in Selected Eastern and Socialist industrialization significantly altered the Central European Economies, Compared to Western production structure of the 1930s (table 2-10). At that Europe and Selected Latin American Countries, 1988 time, Bulgaria, Yugoslavia and Romania were predomi- (in 1988 US dollars) nantly agrarian economies, with the share of agriculture Poland 1,860 in national income at over 53 percent in 1938. In the late Hungary 2.460 1980s that share was down to less than 15 percent. How- Yugoslavia 2,520 ever, as will be seen below, socialist industrialization did Czechoslovakia 3,300 not bring much prosperity. Mexico 1,760 The case of Spain provides a useful contrast. A back- Brazil 2,160 ward Western European country in the thirties, Spain Argentina 2,520 industrialized in the 1950s and 1960s around (protected) Spain 7,740 market lines and ended up in 1988 with a level of per United Kingdom 12,810 capita GDP around three times higher than the average Italy 13,330 for Central and Eastern Europe (table 2-18). In Central Austria 15,470 and Eastern Europe, socialist industrialization began France 16,090 Germany, F.R. 18,480 with an "easy phase," as idle resources could be pulled United States 19,840 from the agricultural sector. However, the industrializa- tion turned very capital-intensive, with a heavy focus on Source: World Bank (1990). 19 In practice, the industrialization of Central and East- Concluding Remarks and Summary ern Europe was based on a wall of tariffs and prohibitions on imports from the West. The Soviet Union dictated the This paper reviewed first the main features of the eco- pattern of specialization, and production was not located nomic history of Central and Eastern Europe in the in- on the basis of comparative advantage. ter-war period and the establishment of socialism in the The end of the centrally planned economy in Central forties. It then looked at the initial results of the transi- and Eastern Europe leaves as a legacy the almost com- tion to a market economy after the collapse of socialism plete dominance of the public sector in the generation of in the late 1980s. A summary of the main findings fol- output (table 2-19). After 18 months of transition from lows: socialism, it is clear that privatization is crucial to creat- * The economic history of Central and Eastern Eu- ing a market economy, although the process faces mul- rope shows episodes of extreme inflation and extreme tiple obstacles of both an operational and political macroeconomic instability following World War I and in nature. the early 1920s. An example of the economic situation Finally, a look at recent levels of per capita GDP in was the hyper-inflation of Poland, Hungary and Austria. several countries of Central and Eastern Europe is re- Its stabilization in Poland, Hungary and other Eastern vealing. The historical differences between Eastern and and Central European economies required stabilization Western Europe in the late thirties (table 2-9) were not of the exchange rate, fiscal reform, the establishment of any narrower after 40 years of socialism (table 2.19).18 independent central banks and (conditional) foreign fi- Quite the opposite, the data suggest (with the usual nancing provided by the League of Nations. Of the coun- caveats pertinent to comparisons of real income over tries in the region, Czechoslovakia, which was formed time and across countries) that the gap in per capita in- out of territories left over from the Austro-Hungarian come has widened spectacularly. For example, while in Empire, is unique for having carried out its reconstruc- 1937 Czechoslovakia had a national per capita income tion after World War I without inflation and with macro- just 10 percent below that of Austria, in 1988 per capita economic stability. GDP in Czechoslovakia was nearly one-fifth that of Aus- & The second half of the 1920s in Central and East- tria (US$3,300 for Czechoslovakia versus US$15,470 for ern Europe was a period of relatively high growth, low Austria). In 1988 the per capita GDP of the most devel- inflation, moderate protectionism and increasing finan- oped countries in Western Europe (for example, the Fed- cial integration in the context of the gold exchange stan- eral Republic of Germany) was 5.6 times larger than that dard system. London and New York were the main of Czechoslovakia, the most developed country of Cen- financial centers at the time, and the British pound and tral and Eastern Europe. In contrast, in 1937 the nation- US dollar were the reserve currencies, in addition to al per capita income of Great Britain was just 2.6 times gold. that of Czechoslovakia.19 * The great Depression of 1929-33 hit Bulgaria, Ru- Currently, the average per capita GDP (World Bank mania and Yugoslavia particularly hard. The drop in the data) in Central and Eastern European countries is clos- international prices of agricultural products was an ad- er to the levels in Latin American countries such as Mex- verse external shock for these agrarian economies. ico, Brazil and Argentina. Moreover, the disruption in the flows of foreign financ- Table 2-19. Size of the State Sector in Central and ing also hurt the Balkan states, Hungary Poland and Eastern Europe even Czechoslovakia, the most industrialized country. (percentage of output and employment of the public sector in total out- The response of the countries of the region to the ad- put and employment) verse external shocks and the collapse of the gold ex- Output Employment change standard in the early 1930s included the imposition of foreign exchange controls, the rise in im- Czechoslovakia 97.0 port duties and the suspension of debt payments abroad. (1986) * The initial recovery of most economies in Central Dem. Rep. Germany 96.5 94.2 and Eastern Europe in the second half of the 1930s was (1982) Poland 81.7 71.5 basically driven by Germany. Its war preparations re- (1985) quired an abundant supply of agricultural products and Hungary 65.2 69.9 raw materials that the Balkan countries, Hungary and (1984) Poland could provide. Czechoslovakia's large industrial Source: Lipton and Sachs, Brookings Papers on Economic Activities (1990), base was also instrumental to German re-armament. To 2:293-341. enhance its control, before the war Germany annexed 20 Czechoslovakia and Austria (the "Czech-Moravian Pro- end of the old regimes is being replaced by a less enthu- tectorate" ). siastic public attitude toward the hardships of transition. * Poland and Yugoslavia suffered the greatest de- Fragile and changing political coalitions and the temp- struction of productive capacity, infrastructure and tation of populism clearly reflect these tendencies. housing and the most extensive human losses as a result * Finally, an important finding of the paper is that of the occupation and war. Czechoslovakia suffered com- the differences in per capita income between Eastern paratively less after the conflict. and Western Europe widened under the socialist re- * After 1945 the nationalization of industrial firms gimes. In 1937, before the Second World War and social- and financial systems and the launching of agrarian re- ism, the per capita income of Great Britain, which was form were the main features of the economic policies the highest in Western Europe at the time, was 2.6 times implemented by the coalition governments that greater than that of Czechoslovakia, which had the high- emerged in Central and Eastern Europe. The drive to- est per capita income in Central and Eastern Europe. In ward full nationalization and the creation of a system of 1988, the per capita income of the Federal Republic of central planning were fully consolidated after a series of Germany (which is now greater than that of Great Brit- coups d'etat led by the Communist parties in 1948. ain) was 5.6 times greater than the per capita GDP of * The second half of the 1940s provides an interest- Czechoslovakia. The average per capita income of Cen- ing laboratory of monetary reform aimed at reducing ex- tral and Eastern Europe is now closer to the levels in Lat- cess liquidity. In Hungary, Rumania, Poland and in American countries such as Mexico, Brazil and Bulgaria the reforms involved a reduction of the money Argentina. supply through (differential) rates of exchange between "old" and "new" money. Poland, Czechoslovakia and Bul- garia instituted the blocking of deposits in the banking Notes system between 1945 and 1955. Western European countries such as France, Austria and Belgium imple- 1. The comments by Bela Balassa, Rudi Dornbusch, Raimundo mented some of the same reforms over the same period. Soto and Steven Webb on a previous draft are appreciated. mented some ~~~~~~~~~~~~~~2. See Berend and Ranki (1974) for a full analysis of the economic - Hungary in 1945-46 provides a good example of hy- history of Central and Eastern Europe before and after World War 1. per-inflation during the transition to socialism. It later 3. Germany's hyper-inflation is not discussed in this paper. For ref- achieved economic stabilization by bringing public fi- erences in the literature see Graham (1930), Bresciani-Turroni (1937), nances into balance, stabilizing the currency and arrang- Dornbusch (1987), Sargent (1982), Solimano (1990a), and Webb ing for a package of foreign financing, debt and relief from (1989). war forea packageiofnforeignsfinancing.debt and relief from 4. The League of Nations loans were not only necessary for the re- war reparations. construction of the economies of the recipient countries but also en- * The collapse of socialism in Central and Eastern hanced the functioning of the system of international payments that Europe in the late 1980s made evident the structural emerged after the First World War. In fact, the United States required weakness of central planning in improving the standards that the Allies-England, France and Italy-honor their war debts of living of the population. The economic hardship and with the United States. However, to honor those debts, the Allies re- quired that the losers of the war-Germany, the countries formned from closed political systems could not last forever. In several the collapse of the Austro-Hungarian Empire and Bulgaria-transfer cases, systemic reform has been implemented alongside to them resources in the form of war reparations. After a few years it the macroeconomic stabilization. The initial results became evident that without external support the loser countries have been a massive collapse in economic activity and would be unable to meet their obligations. Thus, the League of Nations stubborn inflation. Economic transformation looms as a loans involved a recycling of the war reparations payments. long and complicated process given the difficult initial 5. Nurske (1944) describes the situation: The fate of the gold ex- long and complicated process given the difficult initial change standard was sealed when France decided in 1928 to take noth- conditions for the transition to a market economy. The ing but gold in settlement of the enormous surplus accruing to her economies of Central and Eastern Europe are character- from the repatriation of capital and the current balance of payments. ized by obsolete and uncompetitive productive capaci- The French gold imports certainly aggravated the pressure of deflation ties, macroeconomic imbalances, lack of modern in the rest of the world and specially in London. In London the pres- infrastructure and factor markets, and weak institutions, sure became unbearable in the end, and the gold parity of the Pound was abandoned. as well as other problems. In addition, the external envi- 6. The bonus system was administratively cumbersome, as the size ronment in the East is not very supportive: the disinte- of the subsidy tax varied by currency and trading partner. gration of Comecon will entail large terms-of-trade 7. The estimates of the destruction are rather imprecise. Among losses for Central and Eastern Europe vis-a-vis the Sovi- the numbers are: Yugoslavia lost 10 percent of its population and half et Union, and a massive influx of Western capital is un- its industrial engines and around 40-50 percent of its agricultural ma- et Union, ~~~~~~~~~~~~~~chinery and equipment. In Poland, 40 percent of the railways were de- likely to happen in the short to medium run. On the stroyed, 85 percent of the housing in Warsaw was devastated, and there political side, the initial euphoria associated with the was massive destruction of industry and agriculture. See Berend and 21 Ranki (1974, chapter 13). .Macroeconomics and Finance: Essays in Honor of Franco 8. Hungary was the only country that returned to its pre-war ter- .Modigliani. Cambridge, Mass.: MIT Press. ritorial status (see Spulber 1954). 9. The exception was property confiscated from collaborators with Dombusch, R., and H. Wolf. 1990. "Monetary Overhangs and Reforms the Germans. The compensation of foreigners, when it occurred, in in the 1940s." Mimeo. Massachusetts Institute of Technology, Cam- general took an ad hoc form and was tied to trade negotiations and oth- bridge, Mass. er deals. 10. See Gurley (1953) for an interesting discussion and documen- Dombusch, R., and S. Fischer. 1986. "Stopping Hyperinflation: Past tation of the monetary reforms in Western and Eastern Europe. A more and Present." Weltwirtschaffliches Archiv [Review of World Eco- recent analysis of the topic is found in Dornbusch and Wolf (1990). nomics] (Kiel Institute). 11. The blocking of bank deposits was also part of the monetary re- forms in Western Europe after World War II. Examples are Belgium in Graham, F. 1930. Exchange Rates, Prices and Production in Hyperin- October 1944, France in June 1945 and January 1948, and Austria, flation: Germang 1920-1 923. Princeton, N.J.: Princeton University Denmark, Norway and the Netherlands in 1945 (see Gurley 1953). Press. 12. The reparations due were calculated to be US$300 million, pay- able within a period of six years. The national income of Hungary in 1945-46 is estimated to have been around US$1 billion (see Bomberger Gurley, J. 1953. "Excess Liquidity and European Monetary Reforms, and Makinen 1983). 1944-1952."American EconomicReview XLIII (March). 13. The worsening of the fiscal situation in Hungary is also attrib- utable to lower tax revenues, the result of administrative mismanage- Kindleberger, C. 1984. A Financial History of Western Europe. Lon- ment after the war, the destruction of records on income and assets don: George Allen & Unwin. during the war, the fall in the level of economic activity and the accel- eration in inflation (see Bomberger and Makinen 1983). Lipton, M., and J. Sachs. 1990a. "The Creation of a Market Economy in 14. For a description of the stabilization policies in Poland, see Eastern Europe: The Case of Poland," Brookings Papers on Eco- Lipton and Sachs (1990a). A comparative analysis of the Polish and Yu- nomicActivity 1:75-147. goslav stabilization plans is found in chapter 9, "Stabilization Pro- grams In Eastern Europe: A Comparative Analysis of the Polish and Yugoslav Programs of 1990" by Fabrizio Coricelli and Roberto de ]a 1990b. "Privatization in Eastern Europe: The Case of Po Rezende Rocha in this volume, For an analysis of privatization in East- land." Brookings Papers on Economic Activity 2:293-341. ern Europe, see Lipton and Sachs (1990b). 15. See Solimano (1990c) for an analysis of the Bulgarian case. Mitchell, B.R. 1975. European Historical Statistics, 1750-1970. New 16. This number tends to be a bit overstated in Czechoslovakia, York: Columbia University Press, where the ratio involves net material product, which excludes several services such as education, health and other "non-productive" activi- Nurske, R. 1944. "The Gold Exchange Standard." In B. Eichengreen, ties. ed., The Gold Standard in Theory and History. New York: Methuen, 17. See Solimano (1990b) for a discussion of the patterns of pro- 1985. ductivity growth in socialist economies. 18. The comparison is based on World Bank estimates of per capita Sargent, T. 1982. "The Ends of Four Big Inflations." In R. Hall, ed., In- GDP in dollars for 1988. Other sources, such as the C.I.A. and Plan flation Causes and Effects. Chicago: University of Chicago Press. Econ, yield higher estimates of per capita income for Central and East- ern Europe. Solimano, A. 1990a. "Inflation and the Costs of Stabilization: Historical 19. Great Britain had the highest per capita income in Europe in and Recent Experiences and Policy Lessons." The World Bank Ob- the 1930s. server 5 (2)(July). References . 1990b. "Macroeconomic Adjustment, Stabilization and Growth in Reforming Socialist Economies. Analytical and Policy Ames, E. 1954 "Soviet Block Currency Conversions." The American Issues." Policy Research and External Affairs (PRE) Working Paper Economic Review XIV (June). Series #399. World Bank. Washington, D.C. Berend, I., and G. Ranki. 1974. Economic Development in Central and . (1990c) "Inflation and Growth in the Transition from So- Eastern Europe in the 19th and20th Centuries. New York and Lon- cialism: The Case of Bulgaria." Policy, Research and External Af- don: Columbia University Press. fairs (PRE) Working Paper Series #659. World Bank. Washington, D.C. Bomberger, B., and G. Makinen. 1983. "The Hungarian Hyperinflation and Stabilization of 1945-1946." Journal of Political Economy 91 Spulber, N. 1957. The Economics of Eastern Europe. Cambridge, (5)(October). Mass.: MIT Press and New York: J. Wiley & Sons, Inc. Bresciani-Turroni, C. 1937. The Economics of Inflation. London: Allen Svennilson, 1. 1954. Growth and Stagnation in the European Econo- & Unwin. my. Geneva: United Nations Economic Commission for Europe. Dornbusch, R. 1987. "Lessons from the German Inflation Experience United Nations. 1949. Economic Survey of Europe, 1948. Geneva: of the 1920s," In R. Dornbusch, S. Fischer and J. Bossons, eds., United Nations. 22 Webb, S.B. 1989. Hyperinflation and Stabilization in Weimar Germa- World Bank. 1990. World Development Report 1990. Baltimore: Johns ny. New York: Oxford University Press. Hopkins University Press. Wicker, E. 1986. "Terminating Hyperinflation in the Dismembered Hapsburg Monarchy." American Economic Review 76 (3)(June). 23 Part II The World Bank Experience with Adjustment Lending i I I .3 World Bank -Supported Adjustment Programs: Lessons for Central and Eastern Europe' Vittorio Corbo The World Bank introduced adjustment lending in As some countries were adjusting to the second oil 1979. It grew rapidly in the first half of the 1980s and shock and others were planning the transition from the then leveled off in 1986-88, when it averaged 24 percent initial stage of financing the deterioration in their bal- of Bank Group lending commitments. In 1989 it reached ance of payments to a second stage of adjustment, they 27 percent of total lending commitments. also had to contend with the effects of major policy The main differences between adjustment and tradi- changes in the industrial countries. Following the fron- tional project lending are that adjustment lending is tal attack on inflation that most industrial countries ini- quick-disbursing and is made conditional on policy re- tiated in late 1979 and the early 1980s, nominal and real forms. The two principal instruments of this program interest rates reached their highest level in 50 years. The are structural adjustment loans (SALs), which support industrial countries entered into a major recession, and economy-wide reforms and institution-building, and the prices for primary commodities collapsed. To make sectoral adjustment loans (SECALs), which have more of matters worse, following Mexico's difficulty servicing its a sector focus.2 Another distinction-that between foreign debt in late 1982, commercial lending to devel- structural adjustment programs and adjustment lend- oping countries all but disappeared. Current account ing-is also worth noting. Countries have on their own deficits that were financed without much trouble in been implementing structural adjustment programs for 1981 suddenly could not be financed at all. a long time, and the Bank has been supporting the prep- For most of these countries, adjustment became ever aration and implementation of adjustment programs more urgent. In those where large distortions and insti- since at least the late 1950s. The quick-disbursing bal- tutional weakness were preventing the economy from ance-of-payments support for adjustment programs pro- producing as much as it might, removal of the distor- vided under adjustment lending was, however, a new tions was seen as a less costly option than reducing the initiative. current account deficit by lowering expenditures. It was The Bank established adjustment lending in response recognized tha. many developing countries would bene- to the difficulty a large number of countries were having fit tremendously from reducing the inefficiency with by the late 1970s in financing their balance-of-payments which they used resources, in particular, by eliminating deficits. They were also experiencing very weak growth. the excessive regulation and moderating the anti-export When the second oil shock hit in 1979, they found them- bias of their policies. Stabilization programs accompa- selves ill-prepared to absorb the increasingly higher nied by reforms toward market-based allocation were prices without radical changes in their economic poli- seen as ways to lessen the cost of the oil shock. Not un- cies and institutions. The common causes of these prob- expectedly, countries had to initiate adjustment pro- lems were many years of distortionary trade, regulatory grams, including Chile, Thailand, Uruguay, Ghana and and exchange rate policies and an expansion of the Mexico. Given these developments in the international state's role in the allocation of resources. Ultimately, economy, it is not surprising that adjustment lending these factors resulted in the inefficient use of resources, grew rapidly in the first half of the 1980s. discouraged exports and created an internationally un- The main purpose of adjustment lending is to help competitive structure of production. Countries facing member countries restructure their economies to create these types of problems were spread throughout the dif- the conditions for equitable growth while maintaining a ferent geographical regions. sustainable balance of payments. Structural adjustment 27 programs include measures to achieve or consolidate inflation, restrictions on factor mobility, ill-defined macroeconomic stabilization and the structural trans- property rights and a large public sector sheltered from formation of the economy. Adjustment lending facili- efficiency considerations, a reform effort that does not tates the phased reduction of the current account deficit address these barriers has a low chance of success. Fur- while a country is introducing reforms, and thereby re- thermore, if these barriers are not removed, the reforms duces the short-run costs of adjustment on output, em- will have low credibility and could unleash responses ployment and consumption. that cause a deterioration in the economic situation in- The risk of this type of lending is that, while it pro- stead of an improvement (Calvo 1989). vides financing to help sustain expenditure levels, it can In countries that are experiencing acute macroeco- result in the postponement of necessary reforms. There- nomic imbalances manifested in high open or repressed fore the Bank attaches conditionalities to ensure that inflation, large fiscal deficits and major balance-of-pay- disbursement of the loan is linked to implementation of ments crises, the structural reforms should start by at- an agreed-upon program. tacking the fundamental causes of the macroeconomic Periodically the Bank evaluates the design and imple- crisis (Fischer 1986; Corbo and de Melo 1987; Sachs 1987; mentation of the adjustment programs it has supported. Corbo and Fischer 1990; and Rodrik 1990). Given that the For example, it recently carried out two comprehensive success rate of stabilization programs in countries that analyses, presented in the Report on Adjustment Lend- have experienced a prolonged period of inflation is very ing-1 (RAL-1) and Report on Adjustment Lending-2 poor, it is dangerous to proceed simultaneously with re- (RAL-2) (World Bank 1988a and 1990b, respectively). forms whose ultimate success depends on the control of The Bank also carries out a research program on devel- inflation (i.e., major trade and financial liberalization). opment policies that has a heavy focus on the current Successful stabilization attempts have involved major fis- and prospective problems of developing countries. The cal adjustment through cuts in government spending, re- findings of this research and of research carried out else- ductions in large subsidies, lessening of the losses of where provide important underpinnings for the Bank's public enterprises and a drastic reduction in the losses of advice on the design of adjustment lending programs the central bank (Dombusch 1989; and Kiguel and Livia- and operations. tan 1988). In the period immediately after the Second The rest of this paper is divided into five sections. The World War, countries that had accumulated a large mon- next section looks at the emerging consensus on the op- ey overhang during the war years used monetary reform timal design of adjustment programs. The following sec- to avoid an explosion in inflation (Dombusch and Wolf tion presents the main findings of RAL-2 on the 1990). Reforms to restore macroeconomic stability could effectiveness of adjustment programs and on the experi- require major structural reforms in the operation of pub- ence with implementation. The findings of RAL-2 on lic enterprises, the tax system and the financial system (in policies for the recovery of growth are reviewed in the the latter to establish at least the capacity to evaluate subsequent section. The next to last section offers some loans on a commercial basis). In countries with a large lessons for the design of adjustment programs in Central public sector that contains public enterprises experienc- and Eastern Europe. Some overall conclusions are pre- ing major losses and where govemment subsidies are sented in the final section. widespread, stabilization could require a major overhaul of the public sector (as in Chile in 1973-76, Mexico in The Emerging Consensus on Program Design 1982-89 and Argentina starting in 1989). Significant in- creases in the mobility of labor and flexibility in hiring and Some important lessons have emerged from analyses firing should be introduced early on to facilitate stabiliza- of the economic reforms, both successful and unsuccess- tion. Parallel programs that provide a safety net for the ful, in developing countries. These lessons, which are temporarily unemployed can increase the political accep- summarized below, are the basis of an emerging profes- tance of these programs and make them more equitable. sional consensus on the design of adjustment programs, Once enough progress has been achieved in reducing although the design of an individual country's program inflation and the balance-of-payments deficit-in a cred- needs to take into account its specific economic and po- ible way-other structural reforms aimed at improving litical conditions. the allocation of resources and achieving sustainable For reforms to be successful, they should have a good and equitable growth should be attempted. These re- chance of resulting in major progress toward removing forms would address such areas as the public sector, the obstacles to maximum output from existing resourc- trade, the financial sector, the regulatory framework and es. In countries with significant impediments to re- the labor market. source reallocation in the form of high and variable 28 The Main Findings of RAL-2: The Effectiveness Effectiveness of Adjustment Programs- of Adjustment Programs and Their Macroeconomic Indicators Implementation3 To assess the contribution of adjustment lending to sustainable growth, RAL-2 examined performance in For countries experiencing acute macroeconomic terms of several intermediate indicators of structural imbalances and suffering from deep-rooted structural transformation-saving, investment and export ratios problems that impede a better use of resources and that and the rate of growth of output. Countries were limit growth, the benefits of a structural reform pro- grouped into three categories for the analysis: early in- gram will take time to develop. Therefore, to evaluate tensive-adjustment-lending (EIAL); other adjustment the effectiveness of adjustment programs adequately, lending (OAL); and non-adjustment-lending (NAL) (ta- the assessment cannot be started until enough time has ble 3-1).4 elapsed since initiation of the program. Intermediate The observed performance of an adjusting country re- evaluations, however, can be conducted using indicators sults from (1) the policies that would have been in place of progress with the adjustment. The analyses underly- in the absence of adjustment lending from the Bank, (2) ing RAL-1 and RAL-2 followed that route, considering world economic conditions, (3) the effects of a Bank- both macroeconomic and social welfare indicators of supported program and (4) other shocks to the economy performance. (droughts, earthquakes, etc.). To isolate the net contri- Table 3-1. Country Classification I. EIAL (Early Intensive-Adjustment-Lending Countries) (25) Bolivia * Cote d'Ivoire Madagascar * Morocco Tanzania * Brazil Ghana * Malawi * Nigeria * Thailand Chile Jamaica Mauritania * Pakistan * Togo * Colombia Kenya * Mauritius Philippines Turkey Costa Rica Korea, Rep. of Mexico Senegal * Zambia * II. OAL (Other Adjustment-Lending Countries) (25) Argentina Guinea-Bissau * Sierra Leone * Bangladesh * Guyana * Somalia * Burkina Faso * Honduras Sudan * Burundi * Hungary Tunisia Central African Rep. * Indonesia Uruguay China * Mali * Yugoslavia Congo, People's Rep. of Niger * Zaire * Ecuador Panama Zimbabwe Guinea * III. NAL (Non-Adjustment-Lending Countries)(28) Algeria (NA) Haiti * (NA) Paraguay (NA) Benin * (NA) India * (NN) Peru (NA) Botswana (NN) Jordan (NA) Portugal (NN) Cameroon (NA) Liberia * (NA) Rwanda * (NA) Dominican Republic (NA) Malaysia (NN) Sri Lanka * (NA) Egypt, Arab Rep. of (NA) Myanmar * (NA) Syrian Arab Republic (NN) El Salvador (NN) Nicaragua (NA) Trinidad and Tobago (NA) Ethiopia * (NA) Oman (NN) Venezuela (NA) Greece (NN) Papua New Guinea (NA) Yemen Arab Republic * (NN) Guatemala (NN) Notes: ELAL = countries that have received two SALs or three or more adjustment operations, with the first adjustment operation in 1985 or before; OAL = other countries receiving adjustment lending; NAL = countries that did not receive adjustment lending in the period 1980 to 1988; NA = countries that did not adjust although it was necessary for them to do so; NN = other NAL countries. * IDA low-income countries. Middle income countries are non-IDA countries. Source: World Bank (1990b), annex table 5.5, pp. 72-78. 29 bution of the Bank-supported program, a counterfactual export prices, lower import prices, lower interest rates scenario was created by estimating the effects on perfor- on external debt and higher external financing. There- mance of: fore, to assess the contribution of an adjustment pro- gram to these intermediate objectives, it is necessary to * External shocks (in terms of interest rates, terms control for these non-program factors. In the report, the of trade and non- official lending) contribution of an adjustment program to macroeco- * The economic policies in the pre-program period nomic performance is assessed by comparing the actual (indicated by the real exchange rate, ratio of the fiscal performance of a country in the period after adjustment deficit to GDP and the annual rate of inflation) was initiated (1985-88) with an estimated counterfactual * The initial values of the four selected indicators of scenario of what would have happened in that period in macroeconomic adjustment-real GDP growth and the the absence of the program but with the same exogenous ratios of investment, saving and exports to GDP. influences. These results are shown in table 3-3 (table 3- 4 gives the leading and lagging performers for each indi- Performance in 1985-88, the period after adjustment cator, after adjusting for other factors). was initiated, as measured by the four indicators, was After explicitly adjusting for the external shocks, ini- compared with that in 1970-80 and 1981-84. Because tial conditions, levels of external financing and policies some countries started to receive adjustment lending in followed in the pre-program period, the change in the the early 1980s, the base period 1970-80 corresponds annual average rate of GDP growth in the EIAL coun- more closely to the period before adjustment lending.5 tries was higher, although it was not statistically differ- Before-and-after comparisons of the performance in- ent, from that of all the other countries when measuring dicators were also made, as they provided useful back- changes between 1970-80 and 1985-88. Between 1981- ground for the counterfactual analysis (whose results 84 and 1985-88, however, the adjustment programs are are presented later). Although a program's effectiveness estimated to have boosted the rate of GDP growth by cannot be judged on the basis of before and after com- close to 2 percentage points. (These results have been parisons, they are likely to be important to the political adjusted for the effects of International Monetary Fund viability of adjustment programs. (Before and after re- [IMF] programs.) Typically, the successful aajustment sults can be obtained directly from table 3.2). programs improved the growth rate as a result of greater It is possible that exogenous influences that had exports, which more than offset the short-term contrac- nothing to do with the Bank-supported adjustment pro- tionary effects of the reform policies. In contrast, the less grams could have led to the observed improvement in successful programs did not shift resources rapidly GDP growth, saving rate and export rate of the EIAL enough from non-tradable to tradable activities so as to countries in 1985-88, after the Bank-supported adjust- increase growth, probably because of market distortions ment programs started. These influences include higher and institutional weaknesses. In some countries with se- Table 3-2. Country Performance Real GDP growth Investment to GDP Domestic saving to GDP Exports to GDP (% per year) (% of GDP) (% of GDP) (% of GDP) Country groups' 1970-80 1981-84 1985-88 1970-80 1981-84 1985-88 1970-80 1981-84 1985-88 1970-80 1981-84 1985-88 EIAL 4.6 (2) 1.3 (3) 4.2 (1) 22.5 (2) 19.9 (3) 18.6 (3) 17.4 (1) 14.8 (1) 17.2 (1) 24.2 (2) 25.1 (2) 28.1 (1) LIC 4.0 0.1 3.9 22.8 18.4 16.6 16.9 11.8 13.2 29.1 27.3 29.6 MIC 5.5 2.8 4.7 22.1 21.9 21.2 17.9 18.6 22.4 17.9 22.4 26.1 OAL 3.9 (3) 2.3 (2) 3.0 (2) 21.3 (3) 22.0 (2) 20.1 (1) 14.0 (3) 12.7 (3) 13.3 (3) 22.2 (3) 24.4 (3) 23.6 (3) LIC 3.2 2.1 4.1 18.3 19.3 19.3 8.1 3.3 6.8 18.0 19.4 19.1 MIC 4.8 2.5 1.7 25.0 25.2 21.1 21.4 23.8 21.6 27.5 30.2 29.3 NAL 5.5 (1) 3.1 (1) 2.7 (3) 23.4 (1) 24.1 (1) 20.0 (2) 16.7 (2) 14.0 (2) 14.4 (2) 26.3 (1) 26.1 (1) 24.6 (2) LIC 4.1 3.1 2.7 17.5 19.1 15.5 9.7 6.0 7.7 19.0 18.1 15.5 MIC 6.2 3.1 2.7 26.2 26.4 22.1 20.1 17.7 17.6 29.7 29.9 28.9 NA 4.7 2.2 1.7 22.2 23.5 19.6 15.7 13.1 13.3 25.7 25.2 22.5 NN 7.0 4.7 4.5 25.6 25.1 20.7 18.6 15.4 16.4 27.3 27.8 28.5 a. LIC = Low-income countries, MIC = Medium-income countries. For the other categories see table 3.1. b. The rate of growth is calculated from constant price local currency data. The ratios are calculated from current price local currency data. The figures in parentheses indicate the ranking for the variable, with (1) indicating the best performance. Source: World Bank data. 30 Table 3-3. The Effect of Being an EIAL Country Periodl Change in rate Change in Change in Change in dependent of growth of GDP" investmentlGDP domestic saving! export/GDP variable (%) (%) GDP (%) (%) Current prices 1985-88 with 1.3 -4.1** 4.0* 6.4** 1970-80 1985-88 with 2.0* 0.5 4.2** 5.0* 1981-84 Constant prices 1985-88 with 1.0 -5.6** 2.0 1.2 1970-80 1985-88 with 1.9* -0.1 5.8** 2.3 1981-84 a. The rate of growth of GDP is measured at constant prices in both cases, but the estimation procedure requires the use of lagged values of all the performance indicators, the reason for the slightly different estimation of the effect of programs on the rate of growth of GDP in the top and bottom of the table. * Statistically significant at the 10 percent level. ** Statistically significant at the 5 percent level. Source: Author's calculations. vere macroeconomic instability, the programs support- However, such reductions in public investment in infra- ed by the initial adjustment loans broke down, a structure and human capital seriously jeopardized the situation that depressed growth in 1985-88. resumption of private investment and the ultimate suc- In the case of investment, after adjusting for other cess of the adjustment programs. At the same time, how- factors, the adjustment programs appear to have led to a ever, expansion of efficient public investment enhances drop in investment's share in GDP (in current prices) of the supply response to the reformed incentive structure 4.1 percentage points between 1970-80 and 1985-88. by increasing the credibility of the adjustment programs The changes between 1981-84 and 1985-88 were small and thus contributing to the expansion of private invest- and not statistically significant. Further, because most ment. EIAL countries carried out a real depreciation in 1985- The decline in investment is a matter for concern, 88, the relative price of investment goods rose. As a re- since in most countries sustainable higher growth is sult, the effect of the programs was an even larger aver- likely, later on in the adjustment effort, to require an in- age reduction-5.6 percentage points of GDP-in the crease in investment above the average levels of the constant price ratio of investment to GDP between 1970- 1980s. The hoped-for recovery of investment to sustain 80 and 1985-88. This decline resulted not only from low- future growth did not occur in most EIAL countries, al- er public investment, but also from lower private invest- though their experience varies (table 3-4). At the same ment, probably caused in part by the greater economic time, the impact of the programs on investment should uncertainty at the start of the adjustment program. It be viewed with caution. Since adjustment is not estimat- should also be mentioned that where an integral aim of ed to have reduced growth, it must have increased the the adjustment program was to curtail public (and pri- average efficiency of capital. Indeed, the most efficient vate) investment programs whose efficiency was low, a way of getting higher growth is just from policy reforms decrease in the rate of investment was part of the adjust- that raise growth without the cost of increased invest- ment. ment. The reduction in the rate of private investment may With respect to the domestic saving rate in current have been unavoidable in the initial phase of the adjust- prices, after adjusting for the effects of other factors, it ment programs (see the next section). Under pressure to increased more in the EIAL countries than it did in other reduce public sector deficits, many governments sub- countries (table 3-3). (The domestic saving rate is more stantially reduced their investment programs (and cur- appropriate than the national saving rate for measuring rent expenditures for the maintenance of infrastructure) the impact of adjustment on resource mobilization, be- because of their inability to reduce other expenditures. cause net factor payments abroad are not deducted and 31 Table 3-4. Country Differences in Performance, about 6.4 percentage points of GDP between 1970-80 1985-88 Compared with 1970-80 and 1985-88 and 5.0 percentage points between 1985-88, although with large differences across countries. The Leading performersa Lagging underperformersa strong positive effect of the programs on the ratio of ex- Change in annual average rate of GDP growth ports to GDP in current prices could in part be the result Korea Nigeria of the accounting effect of the real devaluations by the Mauritius Philippines EIAL countries in the third period. As to the ratio of ex- Morocco Malawi ports to GDP in constant prices after controlling for oth- Ghana Cote d'lvoire er factors, the adjustment programs on average had a Thailand Mexico positive, although not statistically significant, effect (ta- ble 3-3). Saving rate in current prices The low and statistically insignificant effect on the ra- Korea Nigeria tio of exports to GDP in constant prices raises concern Chile Zambia about the speed of the supply response of exports to the amaica Philippines changed incentives brought about by a real devaluation. Mauritius Senegal The small and slow average response may have been Investment rate in current prices caused by the absence of the investment needed to in- Costa Rica Cote d'Ivoire crease supplies and by uncertainties about the stability Korea Malawi of the improved incentives for exports. For countries Jamaica Nigeria with a long history of macroeconomic instability, dis- Chile Zambia crimination against exports and unstable real exchange Kenya Philippines rates, the export response would be low.7 In terms of the ratio of imports to GDP in current Investment rate in constant prices prices, the structural adjustment programs in the EIAL Korea Malawi countries had a small negative effect of 1.7 percentage Mauritania Zambia points from 1970-80 to 1985-88 and a small positive ef- Mauritius Nigeria fect of 1.3 percentage points from 1981-84 to 1985-88.8 Togo Cote dPlvoire Given that in some countries output has been con- Madagascar Philippines strained by a lack of imports, the improved access to im- Export-to-GDP ratio in current prices ported inputs may in part have led to the increase in the Jamaica Kenya efficiency of investment between 1981-84 and 1985-88 Mauritius Senegal by permitting fuller use of productive capacity. Chile Malawi The macroeconomic performance of intensive-ad- Korea Zambia justment-lending countries has thus been at least ade- Mauritania Brazil quate along the dimensions of GDP growth, saving, Export-to- GDP ratio in constant price-s exports and imports, with the very strong performriance Jamaica Kenya of some countries more than making up for the declines Mauritania Zambia experienced by others. In the area of investment, howev- Korea Nigeria er, aggregate performance has been poor, although with Togo Madagascar important differences across countries. Mauritius Malawi a. The designation of leading and lagging performers is based on the perfor- The Effectiveness ofAdjustment Programs-Social mance ranking of each country after adjusting for the effects of initial conditions Welfare Indicators and the external environment. Source: Author's calculations. Aggregate economic performance does not tell the net foreign transfers are not included.)6 As a share of whole story of the effect of structural adjustment. It is net, oei tasers are not n shrceno important also to consider what happened to the indica- Gpoithe domestic shIAv ungtraterose thabot 4id prntage . tors of consumption, especially by the poorest levels of points more in ther unt than it did in th- society, and what happened in the education and health er countes, whether usg 1970-80 or 1981-84 as the sectors. base period. In countries with an unsustainable current account With respect to the ratio of exports to GDP in current cit,the mnom icacompon ent adjust pie, afe adutn fo ote,atr h aksp deficit, the macroeconomic component of the adjust- porices,atedro s hadj f postver efact,ors the Ban-u ment program encourages a reduction in aggregate de- ported pormhaapstvefetbomand, generally through monetary and fiscal restraint. 32 It also encourages a switch in production from the non- od, as noted, interim evaluations such as those here are tradable to the tradable sectors, generally through a real still necessary.9 In general, the limited cross-sectional devaluation. The reductions in aggregate demand are data on changes in poverty do not suggest that adjust- likely to have negative short-run effects on the growth of ment lending has on average increased it. Other aggre- output and employment. On the other hand, the struc- gate data support similar conclusions: tural reforms (economy-wide and sector-specific) im- prove the efficiency of the economy, have a longer term * On average, the rate of growth of private consump- positive impact on the growth of output and are likely to tion in the EIAL countries recovered in the late 1980s to reduce poverty in the medium to longer run. That is, the rate achieved in the 1970s-and the rate in the late while the reduction in aggregate demand and the struc- 1980s was higher in total and on a per capita basis in tural reforms are bound to have distributional conse- comparison with other groups of countries. quences, the potential adverse short-run effects must be * The socioeconomic indicators of the status of the weighed against the longer run benefits. poor in developing countries or in the EIAL group of In countries with high inflation, policies that perma- countries did not deteriorate in the 1980s on average. nently reduce the fiscal deficit make a major contribu- The indicators of nutrition improved, and average pro- tion to the reduction of inflation. Lower inflation should tein intake continued to rise from 1983-84 to 1986 in all help the poor, who are less able to protect the real value categories of countries, with and without adjustment of their assets and incomes from inflation. Lowering the lending. Infant and child mortality, both indicators of fiscal deficit requires a combination of revenue increases the longer run health status of the poor, continued to and expenditure cuts. On the revenue side, higher in- improve on average for country categories with and come taxes generally do not affect the poor, and the without adjustment lending. goods they consume can be exempted from excise taxes. * The shares of central government expenditures for As to expenditures, whether the decreases will affect the health and education declined on average in the EIAL poor depends on the incidence of the cuts. For example, countries having data. Some of the decline may have oc- reductions in health spending could have a negative im- curred because better targeting of public expenditures pact. If, however, the composition of health expenditures left middle- and upper-income groups paying for more of were to switch toward preventive medicine and away the provision of these services or because other levels of from curative medicine, which goes mainly to the mid- government took responsibility for some of these expen- dle class, the impact on the poor could be softened. ditures. In education, there were declining rates of pri- As to structural reforms, changes in relative prices mary school enrollment for the EIAL countries. This that remove the biases against labor should help reduce trend is inconsistent with restoring sustainable long- poverty in the long run. Devaluation of the exchange term growth, which requires strengthening the human rate will help the poor if they produce tradable goods and capital base, an important input into growth. In health, will hurt them if they consume tradable goods, such as the coverage of immunization generally increased in all imported necessities. Removal of the ceilings on agricul- country groups, a trend that probably accounts for much tural prices will benefit the rural poor, who are net pro- of the continuing decline in the rates of infant and child ducers of food, but hurt the urban poor, who are net mortality. consumers of food. It is difficult to assess the effects of adjustment pro- Program Implementation grams on the poor for three reasons. First, it is inherent- ly difficult to establish causality-to isolate the effects of Progress with implementation is measured by the adjustment programs from other factors-and particu- share of conditions in the loan agreements that the larly to determine whether alternative policies would countries have implemented by the time the final have done better or worse. Second, socioeconomic data tranche is released. The data on conditionality and im- on the living conditions of the poor are scarce and often plementation (note that relative to RAL-1, RAL-2 used a of dubious quality. Although many of the poor work in much expanded sample of loans approved in FY79-88) the informal sector, data on the output of that sector and show most of the same patterns found in the first Report on other variables are usually not included in official sta- as well as some new ones (Webb and Shariff 1990). Coun- tistics. Third, the adjustment programs are relatively tries began implementing their structural adjustment new, and their long-run positive effects probably take programs before the adjustment loans became effective longer than the experience with adjustment so far. and frequently continued implementation after the dis- While complete analysis of the adjustment programs bursements ended. Of the conditions in the loan agree- must await the conclusion of the entire adjustment peri- ments in the sample, 84 percent had been implemented 33 at least substantially-a higher rate than that found with Main Results of RAL-2: Policies That Promote RAL-1-and 66 percent had been implemented fully or Growth more than fully by the time the final tranche was re- leased. RAL-2 not only evaluated the effectiveness of adjust- The rate of implementation of loan conditions in- ment lending but also provided a summary of the re- creased during the 1980s, both for countries that had re- search on what policies promote sustainable growth. ceived adjustment loans since the early 1980s and for The lessons from this research are relevant for the de- countries that started more recently. For the loans in the sign of growth- oriented adjustment programs. In theo- sample whose final tranche was released in FY89, i.e., ry, adjustment measures to achieve an adequate supply since the first Report, 99 percent of the conditions were response should boost investment, particularly in the implemented at least substantially and 80 percent of the tradable goods sector. An increase in investment will conditions as originally written were implemented fully provide a connection between adjustment, growth and or more. In the rare cases where a condition as originally external balance that will ensure the sustainability of the written came to be seen as unnecessary, the Bank waived adjustment effort. The reason is that structural adjust- the condition, with approval from the Board. If the one ment programs change an economy's incentives to in- loan in the sample for which this eventuality occurred in crease efficiency and encourage growth. Given the FY89 is excluded, the rate of full implementation of con- pivotal role of increased investment in the recovery of ditions rises from 80 percent to 88 percent. In short, the sustained growth, adjusting countries will need to insti- final tranches were released only when all the conditions tute further policy reforms in the later phases of adjust- in the loans were at least substantially fulfilled. An im- ment to remove the impediments to the increase in portant finding is that the implementation rates in investment. countries with higher rates of inflation were lower. This This section looks at why adjustment programs have finding once again illustrates the importance of macro- constrained investment and then suggests further policy economic stability. reforms adjusting countries will need to make to achieve Governments were more frequently able to develop sustainable growth. Specifically, they must address the and maintain political support for the structural adjust- uncertainty and lack of credibility that often accompany ment when they designed the program with this condi- adjustment programs and deter investors, they must fos- tion in mind and were active in explaining the source of ter higher rates of saving by both the public and private the problems addressed by the program, how they sectors, and they must increase the efficiency of invest- planned to tackle them, why that approach was the best ment by removing distortions. option and how people would benefit from the new poli- cy environment. Typically this openness mobilized ben- Increasing Investment eficiaries to become political supporters. While technical considerations sometimes caused unavoidable delays in To understand what policies are needed to increase program implementation, more prompt implementa- investment, it is useful to understand why it declined so tion almost always increased the chances of political frequently during adjustment. One reason for the slow support. Awareness of the economic problems that mo- investment response to structural adjustment programs tivated the initial decision for reform was greatest at the was investor uncertainty about the governments' com- beginning, giving the authorities maximum latitude for mitment to carry out the programs. That is, the adjust- reform. Support to sustain the new status quo then de- ment programs lacked credibility. Private agents may veloped as the structural reform paid off in growth and also have been receiving mixed incentive signals-some higher living standards. associated with the previous policy rules, some with the Although adjustment programs often call for a reduc- stabilization package and some with the structural re- tion in the resources going to the public sector, it is forms aimed at restoring medium-term growth. This un- equally important to strengthen public institutions in certainty about the future economic environment, terms of improved policies, organization and manage- particularly the incentive structure, left investors reluc- ment. Institutional development is essential for both the tant to make fixed investments, as for the most part implementation and ultimate success of many of the re- those investments are irreversible: capital, once in- forms the Bank supports. Not surprisingly, actions to stalled, can seldom be put to productive use in a different improve the efficiency of the public sector, including the activity, at least not without incurring a substantial cost. design and implementation of public policies, are inte- Thus, a trade reform that is suspected of being only tem- gral parts of most adjustment operations. porary can reduce investment, as the economic agents 34 postpone their investment decisions while waiting to see cases it was unsustainably high and of dubious produc- whether the reform lasts (Serven and Solimano 1990). tivity. Other reasons were the deterioration in fiscal con- the opposite is also true, however: a stable and pre- ditions as a result of the cut in foreign lending and the dictable incentive structure and macroeconomic poli- lack of adjustment in other fiscal expenditures, the rise cies can further investment, even more so than tax in international and domestic interest rates, and the incentives or subsidized interest rates might. Under con- sharp acceleration of inflation, which eroded real tax re- ditions of great economic and political uncertainty, it is ceipts. In the case of private investment, slower or even usually prohibitively expensive to make tax and related negative growth discouraged it in several countries, as incentives high enough to have any significant impact did the adverse external shocks, the high real interest on investment (Pindyck 1989). In Argentina, Honduras, rates, the uncertainty about the new configuration of Morocco and even Turkey, the investment incentives relative prices and incentives, and the inability of gov- substantially enlarged government deficits without ap- ernments to stabilize their economies. In addition, the preciably increasing investment. debt overhang may have discouraged investment not Based on a first look, investment declined because of only through the uncertainty it created but also through the reduced availability of financing. Lower external fi- its implicit "tax" on future output and the resultant nancing forced an important decline in the deficit in the credit rationing in the international capital markets. resource balance-defined as the difference between do- Uncertainty and lack of credibility often undermine mestic investment and domestic saving-following the the effectiveness of macroeconomic policy. There are debt crisis in 1982 (table 3-5). Because this decline was two aspects to credibility: the internal consistency of the not matched by a sufficient increase in domestic saving, adjustment program; and the government's commit- the deficit was almost entirely reflected in reduced in- ment to carry it out despite possible short-run costs. vestment. The factors accounting for the drop in invest- When credibility is low and the investment response is ment should be looked at more carefully. therefore insufficient to restore growth, a structural ad- The demand for investment fell for several reasons. justment program may entail larger than anticipated so- For one, public investment contracted because in some cial and economic costs.10 A persistent slump may Table 3-5. Gross Domestic and National Saving Ratios in 83 Developing Countries Indicator Group 1970-80 1981-82 1983-84 1985-88 Gross domestic All 22.4 24.0 20.2 19.6 investment Highly indebted 22.8 23.0 18.0 18.4 (percentage of GDP Middle income 25.5 28.6 24.4 21.9 at current prices) Low income 19.7 20.3 17.0 17.4 Gross domestic saving All 16.1 13.7 13.9 14.9 (percentage of GDP Highly indebted 20.3 20.1 19.8 20.2 at current prices) Middle income 18.3 17.5 17.7 17.8 Low income 12.5 7.6 8.0 9.9 Resource balance All 6.4 10.3 6.2 4.6 deficit Highly indebted 2.5 2.9 -1.7 -1.8 (percentage of GDP Middle income 7.2 11.1 6.7 4.0 at constant prices) Low income 7.2 12.7 8.9 7.5 Gross domestic All 23.4 24.1 20.6 19.6 investment Highly indebted 23.1 22.3 17.1 16.8 (percentage of GDP Middle income 25.7 28.6 24.9 22.1 at constant prices) Low income 21.5 20.7 17.8 18.0 Rate of growth of All 4.7 2.7 1.8 3.3 real GDP Highly indebted 5.0 -0.3 -0.4 2.7 (percentage per Middle income 6.1 4.5 3.9 3.2 year) Low income 3.5 2.5 0.5 3.5 Source: World Bank Economic and Social Database (BESD). 35 develop before investors become confident that the ad- enterprises have contributed to the high public deficits. justment measures will be maintained. In Bolivia, for example, public enterprise deficits Resolving the problem of a low investment response reached 5 percent of GDP before the 1987 stabilization. may be particularly critical for economies with a history In Argentina, they have fluctuated between 2 percent of frequent policy swings or failed stabilization attempts. and 7 percent since the early 1980s. In Zimbabwe, they As the recent experience of Bolivia and Mexico shows, were reduced from 9 percent of GDP in 1982 to 4 percent while establishing the right economic incentives is a in 1988. In some countries, the losses of the central bank precondition for investment and growth, it cannot guar- arising from fiscal operations have been even more im- antee them (Dornbusch 1989). portant than the deficit of the general government or the An assessment of credibility and uncertainty should public enterprises. Often these losses have resulted from influence the choice between gradual and abrupt adjust- operations such as emergency loans at subsidized rates ment. Under gradual adjustment, the initial objectives to failing domestic financial systems and from foreign are modest ones that can be achieved with near certain- exchange subsidies to domestic holders of foreign debt. ty, so that the government can build its reputation. In Argentina the central bank's losses have fluctuated Abrupt adjustment starts with an over-adjustment-say, between 2 percent and 6 percent of GDP since 1982, an over-depreciation of the exchange rate through large while in Chile they were 7 percent of GDP in 1985 and in cuts in tariffs-the aim being to frontload the incentives Venezuela 6 percent of GDP in 1987 (Corbo and for resource reallocation. However, this approach also Schmidt-Hebbel forthcoming). concentrates the costs of the adjustment."' The choice Changes in the public sector deficit and in public sav- appears to depend largely on the specifics of each coun- ing result not only from the direct effects of tax and ex- try, with the social distribution of the adjustment costs penditure policies but also from the interaction of fiscal and policy experience likely to be important factors. policy with other policies and with foreign economic Sufficient external support for the adjustment effort shocks.12 For example, depreciation of the exchange rate of a committed government can raise the confidence of affects all budget items that are fixed in foreign curren- investors in the sustainability of the adjustment and cies or indexed to world prices. A real depreciation in- thus enhance the take-off of investment. Implementa- creases the budget deficit (relative to GDP) when the tion of well-targeted public investment projects that at- public sector has more expenses than revenues denomi- tract private investment may be another important nated in foreign currencies-as in countries where the element in getting growth under way. The removal of foreign debt service is a large part of public spending impediments to resource allocation, the development of (e.g., Brazil, the Philippines and Turkey). In countries financial markets, and a more transparent legal and reg- where the public sector obtains much revenue from im- ulatory framework could also enhance the investment port taxes or commodity exports-oil in Mexico, phos- response. phates in Morocco and copper in Zambia-a real depreciation tends to decrease the budget deficit. Increasing Saving In most countries the bulk of saving is accounted for by the private sector. While government policy can To sustain investment for a desirable rate of growth, readily alter the disposable income of the private sector, the adjusting countries have to increase their rate of sav- in a market economy it has only limited influence on the ing. This need is greatest in the highly indebted coun- share of disposable income that the private sector saves. tries, mostly in Latin America and Sub-Saharan Africa, Key policies that affect private saving are the rate of re- whose saving rates fell significantly in the 1980s (table turn on saving, the level and form of taxation, the rate of 3.5). inflation, the real exchange rate, the flight of capital, the Although public policies can affect public saving di- business cycle, the inflows of foreign capital and the rate rectly, there are limits on their effect on private saving. of growth (table 3-6). Public saving and the way it is financed affect the eco- Credit rationing and controls on interest rates dis- nomic environment-GDP growth and inflation-and courage saving in many countries. The low or negative this environment in turn affects decisions on private sav- real interest rates on deposits and targeted credits re- ing. That is, public and private saving, although analyzed duce the supply of loanable funds and hence effectively separately here, are closely linked. ration investment.'3 Financial reforms that raise real in- To measure public saving properly requires defining terest rates to market clearing levels are justified be- the public sector comprehensively as encompassing the cause they improve the efficiency of resource use, which central and local governments, financial and non-finan- boosts long-run growth. 14 The effect of higher real inter- cial public enterprises and the quasi-fiscal operations of est rates on the level of private saving is ambiguous, the central bank. In many countries, the losses of public however. The reason is that an increase in the real return 36 on saving has two offsetting effects. First, a higher real rates to rebound from very negative to near-zero levels interest rate decreases the present cost of future con- often has a positive impact on measured private saving sumption, so that it is attractive to consume less now in financially unstable, high-inflation countries. This ef- and more in the future and thus to save more today. Sec- fect is attributable to reduced flight into consumer dura- ond, it is no longer necessary to save as much to achieve bles and foreign capital after the interest rate has risen. a target level of future consumption. A higher real inter- Public policies to raise public saving and hence na- est rate therefore allows greater consumption both today tional saving have a key role in structural adjustment. and tomorrow and reduces the need to save today. The evidence shows that the private sector does not re- Given this theoretical ambiguity, the effect of the real duce its saving one-for-one with tax increases or public interest rate on saving becomes an empirical issue. A spending cuts; it follows that reductions in budget defi- large body of evidence for both industrial and developing cits increase national saving. This fact has profound im- countries shows that, on average, real interest rates or plications for Bank-supported adjustment lending. It after-tax real rates of return do not have a significant ef- was concluded earlier that after controlling for the effect fect on the share of private income that is saved.15 How- of other factors, the domestic saving rate rose 4 percent- ever, financial liberalization that allows real interest age points more in EIAL countries than in any other Table 3-6. Public Policy and Private Saving: Effects of Intervening Variables After-tax Inflation real rate Per capita disposable income Foreign Foreign and relative of return on Deviation resource income Capital Total effect price instability saving Growth Trend from trend constraint concentrat. flight on saving Effect of an increase in intervening variable - + + + + - + on private saving Policy Financial liberalization ? + + +? (L.run) (L.run) Fiscal/monetary + ? + + - + stabilization (L.run) L.run) (S.run) (L.run) Selective tax incentives +? + on particular financial assets Shift of taxation from + + corporations to households Shift of taxation from + + higher to lower income households Shift from income to ? + + consumption tax Real exchange rate + + + depreciation (L.run) (S.run) (L.run) Foreign capital inflows + (L.run) Notes: L.run = long run; S.run =short run. The signs in the first line indicate the effect of the intervening variable on private saving, while the signs in the remainder of the table denote the fiscal effect of each policy through the corresponding intervening variable. Source: World Bank (1990b), table 7.2, p. 95. 37 group of countries. Thus, adjustment programs support- much to foster higher growth in the long run, a conclu- ed by the Bank have played a significant role in raising sion that is based on an examination of country experi- aggregate domestic saving and changing its composition ence and a simulation of a structural model of growth. in favor of public saving. Since policy-makers have only a limited amount of polit- ical capital for correcting distortions, they should con- Increasing Investment Efficiency centrate their efforts on the changes that have the largest pay-off in terms of increased growth rates. If Because resources for investment are scarce, they are more than one major distortion exists, all should be re- costly, and it is therefore vital that the efficiency of in- duced together. vestment be increased. Further, because more efficient investment has a higher rate of return, increasing the ef- Lessons for Central and Eastern Europe ficiency of investment encourages savings to stay in the developing economy and not to go into capital flight. This section presents some lessons for the design of Investment is more efficient with relatively non-dis- programs in Central and Eastern Europe from the expe- tortionary policies, and therefore policy changes that re- rience, successful and unsuccessful, with adjustment duce the distortions in resource allocation not only raise programs elsewhere. the baseline level of efficiency but also tend to raise Although initial conditions are not the same in all growth in the long term. The size of the initial distor- Central and Eastern European countries, some common tions and the magnitude of the reduction affect how elements exist. At the cost of simplification, the general much growth will respond. Growth-enhancing policies characteristics are the iollowing. There is large excess include lowering tariffs, relaxing import quotas, raising demand at existing prices. This situation holds particu- or decontrolling domestic interest rates, reducing re- larly in Bulgaria, Romania and, until recently, Poland. serve requirements or mandatory government bond- Excess demand is less severe in Czechoslovakia, Hunga- holdings, reducing government subsidies for the ry and Yugoslavia. When the external debt is large and consumption or production of particular goods, and re- the balance of payments unsustainable-conditions forming taxes to reduce or eliminate differential treat- found in all Central and Eastern European countries ex- ment of sectors or inputs. cept Romania and Czechoslovakia-a large part of for- Reforms to improve incentives can proceed in several eign trade takes place on non-market terms. At the steps. For example, tariffs can initially be substituted for structural level, there are several problems. The struc- quotas, both to increase the transparency of incentives ture of industry in most Central and Eastern European and to raise public revenue. Later, tariffs can be reduced countries is dominated by large state enterprises that are as other revenue sources are expanded. Institutional de- uncompetitive at international prices. There is a lack of velopment may be necessary to strengthen the private labor, land, financial markets and the whole set of insti- responses to such changes in incentives. In low-income tutions and arrangements needed to support a market countries, the lack of well-developed public and private economy (especially appropriate accounting procedures, institutions may be an important hindrance to growth property rights, bankruptcy laws and commercial law). even if trade and financial incentives are not distorted. A At the same time, Central and Eastern European coun- stable system of civil liberties, well-defined property and tries have some big advantages-they have very good contractual rights, and predictable and equitable regula- human capital bases, are next to Western Europe, one of tion are widely believed to be particularly important in the most dynamic international markets, and have a harnessing the energies of entrepreneurs in Africa privileged political relation with the Western European (World Bank 1989b). A comparison of the long-run governments.'6 growth experience among many developing countries For Central and Eastern European countries that concluded that the administrative competence of gov- start with acute macroeconomic imbalances in the form ernments was the single most important factor explain- of high open or repressed inflation and/or unsustainable ing the differences in growth (Reynolds 1985). Political current account deficits, the adjustment program needs stability and the safeguarding of civil liberties have also to start with stabilization and give priority to reforms been found to increase growth (Kormendi and Mcguire that aim to restore macroeconomic balances. Otherwise 1985; Scully 1988). the success of other reforms will be put in jeopardy. For An important finding of the research is that the larg- countries with a history of increasingly binding price est pay-off comes from changing high distortions into controls, stabilization must embody two components: low ones. Neither a small reduction in high distortions first, elimination of the monetary overhang (a stock nor the complete removal of small distortions does problem); and, second, elimination of the public sector 38 deficit (a flow problem). Elimination of the monetary functioning of a market economy (i.e., property rights, overhang can be achieved through a once-and-for-all in- bankruptcy laws, appropriate accounting procedures, crease in the price level or through monetary reform. etc.). Therefore, early on the authorities need to face the Coming out of World War II, most European countries issue of clarifying property rights and establishing the used monetary reform (Dornbusch and Wolf 1990). In basic conditions for the functioning of the labor market. contrast, in the post-1950s most countries have sought For a country that is about to initiate a major reform to eliminate the monetary overhang by increasing the effort, the question it has to face from the beginning is level of prices. the sequence and speed of other reforms. For one, any Countries with a large monetary overhang that follow reform that calls for a major reallocation of resources the increase in the price level route run the risk that the will need to deal with appropriate and credible relative initial increase in the price level could result in a pro- prices early on. For another, trade reforms have a very tracted period of high inflation. Even after taking care of high priority in countries that have much to gain from the fiscal adjustment, some countries have found it very integration into the world economy, such as was the case difficult to contain the inflation dynamic that usually de- in Mexico in 1982 and Chile in 1973 and is the case in the velops following the initial increase in prices. For exam- Central and Eastern European countries today. The ini- ple, when Chile liberalized prices in 1973, even though tial stages of a trade reform, such as the replacement of it reached a public sector surplus by 1975, it experienced managed trade by open trade, the replacement of quan- three-digit inflation until 1977 (Corbo and Solimano titative restrictions for tariffs and the reduction of ex- 1990). Most of the reduction in expenditures has to be treme tariffs, should be attempted early on. However, done through fiscal policy, as monetary policy does not major trade liberalization should be attempted only have much of a role to play in countries without capital when clear and credible progress has been achieved in markets and where the main borrowers are public enter- reducing inflation or when there is a clear perceived prises without a hard budget constraint (the case in commitment from the authorities that the anti-infla- Chile in 1973, Mexico in 1983 and Egypt in 1990). tionary program is a very high priority. In countries with In countries where there is limited flexibility in the a very uncompetitive domestic economy, trade liberal- labor market, to break the inflation inertia that can fol- ization may help the stabilization effort, as may have low the liberalization of prices, the drastic reduction in been the case in Mexico in 1988. In countries with a long the public sector deficit needs to be followed by some period of a very isolated domestic economy, trade liber- kind of income policy.1'7 Israel followed this route in alization could also help in providing a market-based 1985, Mexico in 1987 and Poland in 1990. To avoid a pro- structure for the relative prices of tradable goods, as was longed period of inflation, a permanent reduction in the the case in Poland in 1990. However, a major fiscal ad- public sector deficit should be achieved early on in the justment is necessary to generate a suitable reduction in reform. A permanent reduction in the public sector def- spending and to avoid a balance-of-payments crisis. icit requires the imposition of a hard budget constraint Trade reforms aim to shift investment and labor from on public enterprises, a drastic reduction in public sec- non-tradable and highly protected import-competing tor subsidies and the creation of an efficient tax system. activities toward export-oriented and efficient import- Another lesson is to avoid the temptation to use only the competing activities. However, unpredictable relative nominal exchange rate to bring stubborn inflation under prices (as is usually the case in high inflation econo- control. The real appreciation that could develop could mies), the lack of labor mobility, the absence of financial put the full reform effort in jeopardy (as in Chile in 1978- markets, and impediments to the creation of enterprises 82, Mexico in 1988-89, Israel in 1986-88, Argentina in are major roadblocks to successful trade reform. There- 1978-80 and Uruguay in 1980-82). fore, in many of the Central and Eastern European coun- Socialist countries in transition to a market economy tries privatization and the development of the labor and and heavy regulated economies in Africa also face, as financial markets could be an important component of countries such as Turkey did in 1980, Chile in 1973 and the economic restructuring program. Mexico in 1982, the problem of how to reduce both the A major study of trade liberalization (Papageorgiou, large distortions in relative prices and the distortions re- Choksi and Michaely 1990) found that countries that sulting from heavily regulated or almost non-existent la- carried out sustained trade reforms usually had lower bor and financial markets. On top of these distortions, fiscal deficits and inflation than did countries where lib- there is the additional burden of a large public enterprise eralization failed. In some cases, successful trade liberal- sector with low responsiveness to price incentives and ization was carried out while stabilization was still under the lack of institutions to manage macroeconomic poli- way-as in Chile in 1974-79 and Turkey in 1980-84. In cies and establish a minimum set of rules for the normal more typical cases, either severe macroeconomic insta- 39 bility contributed to the failure of stabilization (Argenti- is even worse. The road is made even more difficult by na, Brazil and Sri Lanka in the 1960s and Peru, the the lack of much precedent in recent economic history Philippines, Portugal, Turkey and Uruguay in the 1970s), for reforms such as those being initiated today in Central or stability contributed to successful liberalization and Eastern Europe. Nevertheless, the evidence re- (Greece, Korea and Spain and all of Western Europe). viewed here suggests that output levels will most likely Internal reforms have also been important determi- suffer in the early years of a massive economic restruc- nants of the success or failure of trade liberalization. The turing. It is very important that governments be aware lack of labor mobility (including restrictions on labor of these adjustment costs as they are the equivalent of an movement within a firm and the requirements for high investment in a better economic system that will emerge severance payments), restrictions on the entry and ex- as the reforms are implemented. To make this invest- pansion of firms, and restrictions on the exit of firms (in- ment highly profitable, governments will need to stick to cluding distress financing for firms that will never be their policies. As the credibility of the reforms grows, the profitable at the new undistorted relative prices) can se- investment and output responses will materialize. verely reduce the benefits of trade liberalization and The road to reform in Latin America is littered with jeopardize the whole reform effort. A lack of incentives failures that arose from governments' incapacity to and regulations that slow or make it costly for firms to achieve and maintain macroeconomic balances and/or restructure or close have been important factors in the their abandonment of well-intended reform efforts in failed or costly liberalization attempts in Poland, Hunga- the face of short-term, unavoidable costs. It was a long ry, Turkey (in the 1970s) and Yugoslavia. In contrast, de- time before the reforms in Chile in the 1970s and in Mex- regulation of the labor market played an important role ico in the 1980s produced sustainable growth paths. The in the success of the trade reforms in Chile (World Bank record of these two countries could have been improved 1990a). Domestic regulatory policies in Mexico and Mo- by addressing some of the issues raised in this paper. Re- rocco that restricted factor and output mobility, includ- form-oriented governments should be prepared to sell to ing restrictions on the entry and exit of firms, raised the the public and to maintain programs that will take two adjustment costs of the trade reforms. In this area, Cen- or three years before output levels begin to rise. In the tral and Eastern European countries need to make major short run, access to external financing to support the ad- efforts. justment effort can help achieve higher consumption levels and finance part of the investment in activities tar- Conclusions geted for expansion as a consequence of the reforms. Based on growing evidence from failed and successful Notes reforms, it can be concluded that: (1) High open or repressed inflation and other mani- 1. The views expressed in this paper are the author's and do not festations of severe macroeconomic imbalances, such as necessarily reflect the views of the World Bank or its affiliated organi- unsustainable current account deficits or very large pos- zations. unsustainable current account.deficits or very large pos- 2. Both types of loans generally disburse against general imports, itive real interest rates, need to be tackled at the begin- with a small list of excluded items. ning of an adjustment program. 3. This and the next section draw heavily on World Bank (1990b). (2) Restrictions on labor mobility and on the exit and 4. The groups of countries and periods used here differ from those entry of firms should be removed at roughly the same in RAL-1. Here the EIAL countries include all the intensive-adjust- pace as trade is being liberalized so that the reforms can ment-lending countries of RAL-1 plus 13 more. Because another year of performance could be looked at, the RAL-2 intensive adjusters in- produce an increase in output early on rather than cause clude additional countries whose second or third adjustment loans unemployment. came after 1985. Within each group, a breakdown has been made be- (3) The creation of a full-fledged financial system tween low- and middle-income countries. Low-income countries are should wait until the stabilization is well-consolidated; defined as all IDA countries (including those receiving a blend of IDA however, markets for working capital financing with ap- and IBRD loans) and middle-income countries as all the rest. RAL 1 concluded that by the end of 1987 the 30 countries receiving propriate mechanisms to assess credit risks should be SALs before '985 performed better on average than did the developing created early on to facilitate the economic restructuring. countries not receiving such loans. This conclusion was based on two (4) Institution-building, including a capacity to for- comparisons: the performance of countries before and after receiving mulate and execute macroeconomic policies and the adjustment loans; and the average performance of countries receiving regulatory framework that support the function of mar- adjustment loans before 1985 and of countries not receiving such loans. The 30 countries receiving loans had modest improvements in kets, are important complements to successful reforms. performance as compared with the other group of countries, despite a The road to reforms is difficult, but the alternative of more unfavorable external environment. The 12 countries that re- perpetual stagnation and deteriorating living standards ceived three or more adjustment loans before 1987 had more pro- nounced improvements. 40 Corbo, V., and K. Schmidt-Hebbel. Forthcoming. "Public Policies and 5. The base period 1970-80 was chosen because it preceded the Saving in Developing Countries." Joumal of Development Eco- major shocks of the early 1980s and was not dominated by conditions nomics. in a particular year or two. Performance in 1985-88 is also compared with that in 1981-84. Corbo, V., and A. Solimano. 1990. "Chile's Experiences with Stabiliza- 6. The effect of the programs on the national saving rate is also tion Revisited." In M. Bruno, et al., eds., Experiences with Stabili- positive but is statistically significant at the 5 percent level only when zation. Cambridge, Mass.: MIT Press. comparing performance in 1981-84 with 1985-88. 7. On the effect of incentives and uncertainty on exports, see Ca- Dornbusch, R. 1989. "Credibility and Stabilization." Massachusetts In- ballero and Corbo (1989). stitute of Technology. Cambridge, Mass. 8. This result comes from the identity that investment minus sav- ing equals imports minus exports. 9. The central topic of the World Bank's World Development Re- Dornbusch, R., and H. Wolf. 1990. "Monetary Overhang and Reforms port 1990 (World Bank 1990c) is poverty. This report includes addi- in the 1940s." Massachusetts Institute of Technology. Cambridge, tional analyses of the impact of adjustment programs on poverty. Mass. 10. Credibility introduces an externality that creates a wedge be- tween the social and private returns on investment. In fact, higher ag- Fischer, S. 1986. "Issues in Medium-Term Macroeconomic Adjust- gregate investment helps sustain the adjustment and therefore results ment. ' World Bank Research Observer 1(2):163-82. in higher returns on investment. However, the individual investor will ignore this mechanism. Fry, M.J. 1988. Money, Interest and Banking in Economic Develop- 11. On these topics, see Kiguel and Liviatan (1988) and Solimano ment. Baltimore, Md.: Johns Hopkins University Press. (1990). 12. Public saving and the deficit are directly linked: the public def- Giovannini, A. 1985. "Saving and the Real Interest Rate in LDCs." Jour- icit is defined as public investment minus public saving. nal of Development Economics 18. 13. For a recent review of financial systems and financial liberal- ization in developing countries, see World Bank (1989b). Kiguel, M., and N. Liviatan. 1988. "Inflationary Rigidities and Orthodox 14. This policy prescription is taken from McKinnon (1973) and Stabilization Policies: Lessons from Latin America." World Bank Shaw (1973). Economic Review 2:273-98. 15. Among the studies presenting growing evidence of the insen- sitivity of saving in developing countries to interest rates, see, for in- stance, Giovannini (1985). For an alternative view, see Fry (1988) and Kormendi, R.C., and RG. Mcguire. 1985. "Macroeconomic Determi- the survey by Balassa (1989). nants of Growth: Cross-Country Evidence." Journal of Monetary 16. For a good description of initial conditions in the Central and Economics 16:141-63. Eastern European countries, including the social and political factors, see Lipton and Sachs (1990). Lipton, D., and J. Sachs. 1990. "Creating a Market Economy in Eastern 17. Inertia can also result from a lack of credibility or from lagged Europe: The Case of Poland." Brookings Papers on Economic Ac- indexation schemes. The latter form of inertia played a central role in tivity 1:75-147. the slow deceleration of inflation in Chile in the period 1978-82 (Corbo 1985). McKinnon, R. 1973. Money and Capital in Economic Development. Washington, D.C.: Brookings Institution. References Meier, G.M., and W.F. Steel, eds. 1989. Industrial Adjustment in Sub- Balassa, B. 1989. "The Effects of Interest Rates on Saving in Developing Saharan Africa. : Oxford University Press. Countries." World Bank. Washington, D.C. Papageorgiou, D., A.M. Choksi and M. Michaely. 1990. Liberalizing Caballero, R., and V. Corbo. 1989. "The Effect of Real Exchange Rate Foreign Thade in Developing Countries. Washington, D.C.: World Uncertainty on Exports: Empirical Evidence." World Bank Eco- Bank. nomic Revview 3(2):263-78. Pindyck, R. 1989. "Irreversibility, Uncertainty and Investment." World Calvo, G. 1989. "Incredible Reforms." In G. Calvo, et al., eds., Debt, Sta- Bank, Policy, Planning and Research Department (PPR) WorKing bilization and Development. Oxford and Cambridge, Mass.: Basil Paper No. 183. Washington, D.C. Blackwell. Reynolds, Lloyd G. 1985. Economic Growth in the Third World, 1880- Corbo, V. 1985. "Reforms and Macroeconomic Adjustment in Chile: 1980. New Haven, Conn.: Yale University Press. 1974-82." World Development (August). Rodrik, D. 1990. "How Should Structural Adjustment Programs Be De- Corbo, V., and J. de Melo. 1987. "Lessons from the Southern Cone Pol- signed?" World Development 18(7):933-48. icy Reforms." World Bank Research Observer 1(2):111-42. Sachs, J. 1987. "Trade and Exchange Rate Policies in Growth-Oriented Corbo, V., and S. Fischer. 1990. "Adjustment Programs and Bank Sup- Adjustment Programs." In V. Corbo, et al., eds., Growth-Oriented port: Rationale and Main Results." World Bank. Washington, D.C. Adjustment Programs. Washington, D.C.: International Monetary Fund-World Bank. 41 Scully, G. 1988. "The Institutional Framework and Economic Develop- . 1989a. Sub-Saharan Africa: From Crisis to Sustainable ment."Joumal ofPolitical Economy 96(3). Growth. Washington, D.C.: World Bank. Serven, L., and A. Solimano. 1990. "Private Investment and Macroeco- . 1989b. World Development Report 1989. Washington, D.C.: nomic Adjustment in LDCs: Theory, Country Experiences and Pol- World Bank. icy Implications." World Bank. Washington, D.C. -_____ 199ua. Lessons in Trade Policy Reform. Policy and Research Shaw, E.S. 1973. FinancialDeepening in EconomicDevelopment. New Series #10. Washington, D.C.: World Bank. York and London: Oxford University Press. -_____ 1990b. Adjustment Lending: Policies for Sustainable Solimano, A. 1990. "Inflation and the Costs of Stabilization: Historical Growth. Policy and Research Series #14. Washington, D.C.: World and Recent Experiences and Policy Lessons." WorldBank Research Bank. Observer 5(2):167-85. _._____- 1990c. World Development Report 1990. New York: Oxford Webb, S., and K. Shariff. 1990. "Designing and Implementing Adjust- University Press for the World Bank. ment Programs." World Bank. Washington, D.C. World Bank. 1988. Adjustment Lending:An Evaluation of Ten Years of Experience. Policy and Research Series #1. Washington, D.C.: World Bank. 42 4 World Bank Adjustment Lending in Central and Eastern Europe Ulrich R. W. Thumm The World Bank has been active in adjustment lend- and chapter 3, 'World Bank-Supported Adjustment Pro- ing in a large number of countries for about 10 years. grams: Lessons For Central and Eastern Europe," by Vit- However, Central and Eastern European countries are torio Corbo, in this volume). relative newcomers to the World Bank, and their partic- The most important lesson from the World Bank's ex- ipation has been largely limited to project lending. perience is that the largest problems impeding the effi- In dealing with adjustment lending in Eastern Eu- cient allocation of resources and limiting growth should rope, an obvious question is whether there is anything be addressed first. In many if not most cases, the largest special about adjustment lending in this region as com- problems result from acute macroeconomic imbalances pared with that in other regions and countries, whose that translate into high inflation and unsustainable bal- experience is summarized in the next section. The third ance-of-payments deficits. In other cases, the most se- and fourth sections address the above question following vere distortions may stem from a interventionist trade an overview of the typical macroeconomic and structur- regime, pervasive price controls or inefficient financial al issues facing the Central and Eastern European econ- intermediation. If the largest distortions are not ad- omies. Based on that information, a set of key elements dressed first, the effectiveness of other policy reforms that ideally should be included in meaningful adjust- will be greatly hampered if not nullified. In a second-best ment programs is postulated. situation, it is even conceivable that a partial reform pro- The fifth section reviews the recent adjustment pro- gram might make the overall situation worse rather grams in Yugoslavia, Hungary and Poland, which the than better. World Bank has been supporting with structural adjust- In most cases, effective stabilization and assurance of ment loans (SALs). In addition, the results of prior adjust- a minimum level of functioning of the goods and factor ment lending in Yugoslavia and Hungary are discussed markets have to precede structural adjustment. Chang- briefly, particularly because their experience is of great ing relative prices and allowing resource allocation to importance in the design and implementation of current proceed in line with such changes are the keys to struc- programs at what is a historically crucial juncture. A few tural adjustment and greater efficiency of resource use. words on the incipient adjustment program in Czechoslo- In the presence of major macroeconomic imbalances vakia are also provided. The key findings are summarized such as high inflation, price relations tend to be highly in the concluding section. volatile and greatly distorted. This environment pre- cludes a clear perspective on profitability and under- A Brief Summary of the Lessons from World mines investor confidence, and thereby impairs greater Bank Adjustment Lending factor mobility and structural adjustment. While the need for macroeconomic stability as a pre- The focus of this paper is on Central and Eastern Eu- condition for successful structural adjustment can hardly rope and the World Bank's support through adjustment be overemphasized, this emphasis should not distract lending for country-specific reform or transformation from other grave distortions and priority areas for correc- programs. Thus, discussion of the Bank's general experi- tive action. In Central and Eastern Europe, the establish- ence with adjustment lending and related lessons is kept ment of functioning goods and factor markets is a high to a minimum (comprehensive coverage is provided else- priority. It will require the liberalization of trade and pric- where-see World Bank 1990a; Corbo and Fischer 1990; es, the promotion of competition and the dismantling of 43 monopolies, development of a financial system with inde- Finally, it should also be highlighted that "govern- pendent financial intermediaries, flexibility in the labor ment ownership" of the adjustment program and thus market, and, last but not least, clarification of ownership the government's full commitment are critical to the rights and a critical minimum participation by the private successful implementation of a program. While the con- sector in production and distribution to make the decen- ditionality the World Bank attaches to its financing may tralized, market-based approach to economic manage- help spur full implementation, it is not a substitute for ment credible. government commitment. In the World Bank's experi- Institutional reforms, particularly in the public sec- ence, many programs have failed for lack of government tor, have also proven to be extremely important. Nor- ownership and commitment. A case in point is the Yugo- mally these reforms include a thorough overhaul of the slavia program of the early 1980s. tax system (aimed at non-distortionary taxes that yield sufficient revenues and that are relatively easy to collect Crucial Economic Problems in Central and at a reasonably low administrative cost) and public en- Eastern Europe terprise reform (aimed at improving efficiency and ulti- mately at privatization). In Central and Eastern Europe, As a result of the dramatic political developments in in the broadest context of a thorough overhaul of the 1989, economic reform reached a critical, historically role of the state in the economy, government expendi- unprecedented stage in all Central and Eastern Europe- tures should be curtailed and rationalized, with the ulti- an countries during the course of 1989-90. These coun- mate goal of lowering excessive tax burdens. tries had earlier attempted different reforms of varying Another area of great importance for the lasting suc- scope at different times; by no means could Central and cess of adjustment is the recovery of investment and a Eastern Europe be considered a monolithic bloc with re- concomitant increase in domestic savings to keep the gard to their mode of economic organization. Hungary balance of payments sustainable over the longer run. and Poland experimented the most (particularly Hunga- The recovery of investment, particularly private invest- ry, with its adoption of the New Economic Mechanism in ment, depends most critically on a stable macroeconom- 1968) but failed to address until recently the root causes ic environment as well as on the functioning of markets. of their well-known and acknowledged economic prob- Those conditions in turn require removal of the impedi- lems. For most of the post-World War II period Yugosla- ments to domestic factor mobility and international via followed its own economic model, although not with trade. Without functioning markets, any recovery of in- any greater success than in other reforming countries. vestment after stabilization will result in a continued Other countries made other attempts at reform. misallocation of resources, with negative implications As they entered the critical reform stage in 1989/90, for longer term growth and the sustainability of the bal- each of these countries faced its own specific economic ance of payments. problems. The degree of macroeconomic imbalance var- In many countries, stabilization and structural ad- ied significantly from one to another. So did, among oth- justment have proven to be controversial because of er factors, the rate of inflation, the acuteness of the their perceived social cost. The friction tends to be the shortages, external indebtedness, domestic price distor- greatest the lower the degree of flexibility in the factor tions, the rate of economic growth and the degree of cen- markets is. Any adjustment program should explicitly tralization of economic decision-making. consider the social and distributional implications and, Despite these differences, there is sufficient common at a minimum, include compensatory measures for the ground to warrant a stylized description of the typical poorest people who might be the worst affected by the economic problems in the region that need to be ad- structural changes. Ideally, the social mitigation compo- dressed through reforms to achieve the common overall nent should be part of a more comprehensive and long- objective of a transition to a market economy in order to term effort to establish a social safety net that provides restore growth and achieve greater social and economic minimum support in cases of unexpected social hard- welfare with freedom of consumer choice. To varying de- ship, that maintains the incentives to work and that is grees, in 1989 all countries in the region faced major in- fiscally affordable. In the reforming economies of Cen- ternal and external imbalances: excess demand resulted tral and Eastern Europe, where the distribution of in- from unsustainable fiscal and/or social enterprise defi- come is relatively even, the greatest emphasis needs to cits, which spilled over into sizable balance-of-payments be placed on the creation of employment in the private deficits and high external indebtedness. (Table 4-1 pre- sector, since it is perhaps the most effective means to sents a number of macroeconomic indicators.) In a few mitigate the costs of transition. countries, most notably Poland and Yugoslavia, the im- balances led to high inflation or, most recently, to hyper- 44 Table 4-1. Economic Performance in Selected Central and Eastern European Countries Indicatora Years Hungary Poland Yugoslavia Czech. Inflation 1980 9.1 9.4 30.9 3.7 1985 7.0 15.1 72.3 2.3 1989 17.0 244.5 1,239.9 1.4 1980-89 8.1 53.1 74.9 1.6 1984-89 8.9 62.9 106.3 0.9 GDP growth 1980 0.2 -10.0 2.3 2.4 1985 0.3 3.6 0.5 2.9 1989 -1.5 0.5 0.8 1.3 1980-89 2.1 1.4 0.5 2.0 1984-89 2.5 4.0 0.3 2.5 Export growth 1980 19.7 21.2 18.6 23.7 1985 -14.8 -2.3 -1.1 -4.0 1989 15.0 4.5 9.8 8.5 1980-89 2.9 -0.6 7.3 6.3 1984-89 5.2 6.7 10.0 2.5 ExportIGDP 1980 21.7 14.0 9.1 10.6 1985 20.3 7.5 15.6 9.8 1989 22.6 11.4 15.3 11.2 Current account/GDP 1980 -1.7 -1.4 -3.0 -3.5 1985 -4.1 -0.9 0.7 2.3 1989 -5.2 -2.8 3.0 0.8 Currency debt 1980 9,090 25,500 18,486 6,850 (mil. US$) 1985 13,955 33,100 18,407 4,608 1989 20,605 40,578 17,320 7,915 DOD/GDPb 1980 40.6 44.9 24.5 16.6 1985 67.6 46.6 42.1 11.7 1989 73.8 61.2 28.3 16.3 Debt/XGNFSC 1980 151.0 286.5 158.2 85.0 1985 228.4 508.4 177.9 87.0 1989 246.1 465.1 105.8 104.0 Debt service/XGNFSC 1980 13.5 95.9 32.9 17.2 1985 60.4 86.0 28.9 17.4 1989 40.8 75.3 29.4 19.4 a. The current account, export and debt data are in convertible currencies. The aggregate economic activity index is gross social product for Yugoslavia and net ma- terial product for Czechoslovakia. The ratios are in percentages of GDP for Yugoslavia. b. DOD=debt outstanding and disbursed. c. XGNFS=exports of goods and non-factor services. Source: Country data, World Bank (1990c); Yugoslavia (various issues). inflation, mainly as a result of the monetization of enter- exchange rates aggravated the external imbalances. The prise deficits. In other countries, most notably perhaps preferred tool to keep the imbalances within manageable Bulgaria, the former German Democratic Republic proportions was the centrally controlled allocation of (GDR) and the USSR, the governments repressed infla- foreign exchange. tion with price controls, and the underlying macroeco- The pervasive state intervention carried down to the nomic imbalances translated into shortages and microeconomic level. Direct price controls, subsidies consequently into forced savings and an ever-increasing and taxes produced grave price distortions, while highly monetary overhang. Multiple and generally overvalued differentiated foreign exchange and trade regimes led to 45 a large number of effective exchange rates. Directed 1990; Kornai 1990). This set of structural or systemic credit allocations at highly negative real interest rates features makes the design and implementation of adjust- and direct control of investment caused further distor- ment programs in Central and Eastern Europe unique. tions, and the de facto guarantee of full employment re- The standard tools of stabilization and adjustment with sulted in overstaffing. which the international financial institutions have The results of this approach to economic management gained ample experience over a long period, most nota- are well-known. The level of efficiency was low, examples bly during the 1980s, are not quite sufficient for the task being an energy intensity 2 to 3 times higher than in the in Central and Eastern Europe. The task goes well be- Organisation for Economic Co-operation and Develop- yond mere reform and adjustment. Rather, it requires a ment (OECD) countries; labor productivity that was 2 to 5 systemic transformation. While state-owned enterprises times lower; transport intensity that was 4 to 10 times are important in many Latin American countries, and higher; excessive inventories, perhaps 10 to 20 times their poor performance has unleashed severe crises for higher in relation to GDP than in the OECD countries; entire economies, at the same time these enterprises low capital productivity as reflected in high incremental have been operating in some sort of market environ- capital/output ratios (ICORs); severe environmental prob- ment. Such systems lend themselves to reform. In con- lems; a growing technology gap; a lack of competitiveness trast, in Central and Eastern Europe many of the in the open international markets despite technology im- fundamental rules of a market-based economy have to be ports from OECD countries; and lagging growth and stan- designed and established from scratch, a process for dards of living. (Table 4-2 presents a number of structural which there is little historical guidance. and efficiency indicators.) While the standard of living in Based on theoretical considerations of the efficiency a country such as Czechoslovakia was about at par with of different modes of organizing production and the con- the West European countries before World War II, it is crete experience with reform in a large number of coun- now only one-third to one-half that of Western Europe. tries, it can be concluded that, ultimately, excessive Over the years Czechoslovakia lost market shares in the direct state intervention and the critical degree to which OECD countries and moved toward a "regressive special- private ownership is lacking largely explain the dismal ization" that replaced its former technology-intensive ex- performance of the Central and Eastern European econ- ports such as machinery with material- and energy- omies (Hinds 1990; Milanovic 1989). The numerous at- intensive goods such as steel, organic chemicals, fertiliz- tempts in these economies to decentralize and introduce ers and textiles. competition or quasi-competition have not included Although these stylized results may appear dramatic, measures to deal with the crucial issue of ownership and the underlying causes as summarized in the preceding concomitant management of the means of production. It paragraphs read like familiar stories from other World was once thought that market socialism, i.e., decentral- Bank client countries. An important question to be ization of decision-making along the lines postulated by asked, then, is whether there are any other critical fac- neoclassical economic theory while maintaining state tors that set Central and Eastern Europe apart from the ownership of enterprises, was a workable system, em- World Bank's typical developing country clients and that bodying, as it did, a compromise aimed at retaining the constitute some sort of quantum leap in underlying efficiency of a market economy without incurring the structural problems (and not just the same problems to undesirable aspects of a capitalist system such as unem- a greater degree). ployment and social inequality. To date, all attempts A large number of such factors have been at work in have failed. Central and Eastern Europe simultaneously for pro- Pure neoclassical economic theory pays little atten- longed periods of time: the preponderant importance of tion to the importance of ownership in achieving maxi- political decisions; central planning (albeit to varying mum efficiency. However, the growing literature on degrees across the countries); the total lack of a financial ownership rights focuses on their critical implications system; the absence of functioning factor markets; the for management and decision-making and thus for the almost complete state ownership of enterprises (or other functioning of markets. The implication is that a divorce forms of ownership and management with confused of the ownership of capital and decision-making over its ownership rights that have had severe implications for use will result in inefficiencies (Milanovic 1989). The the functioning of the factor markets); overemphasis on problem is even more acute if the relations of ownership an even distribution of income and thus a lack of perfor- and management are confused, as is the case in many mance-related compensation systems; and the over- Central and Eastern European countries that have work- whelming role of barter trade arrangements within the er self-managed firms with social ownership or state- Council for Mutual Economic Assistance (CMEA) (Hinds 46 Table 4-2. Structural Indicators A. Energy i. AnnualAverage Percentage Changes Years Countries GDP Energy Oil Energy use consumption elasticity 1976-79 OECD 3.9 3.0 2.7 0.8 1976-79 Hungary 4.1 3.9 2.14 1.0 1975-79 Poland 3.8 3.7 4.1 1.0 1975-79 Yugoslavia 6.6 2.6 5.6 0.4 1980-82 OECD 0.9 -2.9 -6.6 n.a. 1980-82 Hungary 2.9 0.2 -5.7 0.1 1980-82 Poland -3.2 -1.4 -5.0 n.a. 1980-82 Yugoslavia -0.6 3.6 -2.5 n.a. 1983-85 OECD 3.2 1.2 -0.4 0.4 1983-87 Hungary 0.5 0.7 -0.3 1.4 19683-87 Poland 4.6 1.9 1.6 0.4 1983-87 Yugoslavia -2.0 1.2 -0.1 n.a. n.a. Not available. Source: Country data,World Bank; and World Bank (1990b). ii. Intensity of Use ([TOE] of primary energy per US$1 million of GDP [1975 US$]; purchasing power parity [PPP]-based) Country group in 1965 1975 1982 OECD-15 682.1 631.7 487.3 CMEA-7 1,312.3 1,145.5 1,138.9 CMEA-3a 1,169.6 1,974.1 1,132.5 a. CMEA-3 consists of Hungary, Poland and Yugoslavia. Source: Gomulka and Rostowski (1988). B. Transport: Surface Transport in East Europe and OECDa (ton-km per US$1 of GNP) Countries 7bn-km per n $ of GNP Czechoslovakia 1.6 Hungary 1.8 Poland 2.3 Yugoslavia 0.24 a. The data generally relate to 1988 and do not precede 1984. Source: EMENA, Terhnical Department for the infrastructure, World Bank, from International Road Federation, WorldRoad Statistics 1984-88. 47 C Inventories: Changes in Stock as a Percentage of GDP (current prices) Hungary Poland Yugoslaviaa Czech. OECDb 1980 1.9 1.6 9.6 6.1 0.28 1981 3.2 -0.2 11.6 2.2 0.45 1982 3.3 7.8 9.3 2.4 -0.41 1983 1.9 4.9 12.6 2.4 -0.15 1984 2.7 5.6 16.3 1.9 1.22 1985 2.5 6.5 18.8 2.1 0.31 1986 2.9 7.0 7.7 2.6 0.36 1987 2.8 6.3 7.0 3.3 0.58 1988 3.9 10.0 5.0 1.4 0.57 1989 5.2 10.0 5.0 2.1 0.53 a. Because of inflationary distortions, the change in stock in Yugoslavia is not reliable. b. The unweighted average includes France, Germany, Japan and the United States. Source: IMF,InternationalFinancialStatistics;World Bank (1990c);Yugoslavia (various issues). D. ICOR Hungary Poland Yugoslavia 1980-85 16.5 23.1 37.4 1985-89 27.7 22.2 n.a. n.a. Not available. Source: Country data, World Bank. E. Staffing (percent) Disguised unemployment Unemployment (1990) Hungary 10-30 2 Poland 8-10 in manufacturing 7 20-30 in service Yugoslavia 20 11 Czechoslovakia n.a. 3 n.a. Not available. Source: Adam (1984); Mencinger (1989); and World Bank. F Tar Revenue as a % of GDP, 1985 Countries % % Czechoslovakia 50.7 OECD total 37.2 Hungary 49.9 OECD Europe 38.9 Poland 37.0 EEC 39.5 Yugoslaviaa 33.7 Romania 11.5 Average 35.9 a. Implies gross social product and a broad case. Source: Coricelli and Rocha (1990). 48 G. Money Stock as a Percentage of GDP Hungary Poland Yu oslavia MIIGDP M21GDP M1/GDP M2/GDpa MIIGSP° M2/GSP 1982 23.2 47.8 22.7 57.4 20.0 32.6 1983 22.7 47.0 22.6 52.1 17.6 30.0 1984 21.8 45.4 20.6 47.3 14.2 26.3 1985 23.5 47.2 20.0 44.8 10.5 22.6 1986 26.6 51.0 18.5 46.2 9.8 20.4 1987 25.1 48.8 14.9 51.4 9.1 17.4 1988 21.9 43.0 15.7 56.4 7.4 14.3 1989 n.a. 38.3 n.a. 36.8 n.a. n.a. n.a. Not available a. Implies gross social product. b. The M2/GDP for Poland includes foreign currency deposits valued at black market rates. The ratio for 1989 is at December 31, 1989 exchange rate. Source: Country sources, World Bank. a H. Importance of Public Enterprises in Selected Countries (percent of output) OECD France (1982) 16.5 (1973: 11.7;1979: 13.0) Austria (1978-79) 14.5 Italy (1982) 14.0 Germany, F.R. (1982) 10.7 (1977: 10.3) United Kingdom (1978 11.1 (1972: 10.2)) United States (1983) 1.3 Developing countries Sudan (1980s) 40.0 Guyana (1978-80) 37.2 Venezuela (1978-80) 27.5 Chile (1981) 24.0 India 10.3 Socialist countriesb Czechoslovakia(1986) 97.0 German Dem. Rep. (1982) 96.5 USSR (1985) 96.0 Poland (1985) 81.7 (1980: 83.4; 1970: 82.2) China (1984) 73.6 (1982: 77.8; 1980: 78.7) Hungary (1984) 65.2 (1980: 69.8; 1975: 73.3) a. Govemment/state-owned corporations. b. Including government. Source: Milanovic (1989). 49 owned firms in which workers' councils ultimately de- Fund (IMF) financial programs (e.g., standby arrange- cide on the use of resources. ments) and World Bank adjustment operations (e.g., Some of the weaknesses of market socialism, most structural or sector adjustment loans). importantly the lack of financial discipline at the enter- Perhaps the most comprehensive and successful eco- prise level, may even have been exacerbated (Kornai nomic adjustment programs in any World Bank client 1990; Brus and Laski 1989; Aslund 1990). Financial dis- countries to date are Chile and Mexico's, which have cipline and competition, including competition from in- been implemented since the mid-1970s and mid-1980s, ternational trade, can only be exercised in a credible respectively. Although the Chilean and Mexican pro- manner when there is a sufficiently high degree of de- grams may be the most comprehensive ones to date, for centralization, autonomy of enterprises and, ultimately, the most part they lack the systemic dimension clearly private ownership. Private ownership would make it needed in Central and Eastern Europe. State ownership highly unlikely and extremely difficult to resort to large- of firms was extensive in Chile and Mexico and required scale government support in the case of financial diffi- far-reaching privatization programs. At the same time, culty. Moreover, a predominantly private system has however, the economies had market elements, albeit room for other forms of ownership and management, in- with gross distortions whose correction constituted the cluding worker self-management, as financial discipline core elements of the programs. In other words, market and even application of the rules of the game to all play- behavior and private ownership of firms did not have to ers are guaranteed or at least have a high degree of cred- be "invented" in those countries. ibility. The typical adjustment or transformation program in The great emphasis in Central and Eastern European Central and Eastern Europe has to be broader in scope, countries on ownership and the effective exercise of with macroeconomic, microeconomic and systemic ele- ownership rights, and the implications for enterprise ments. To have a minimum chance of success, beyond management, is a key aspect of the quantum leap in un- addressing the gravest distortions first, the transforma- derlying structural problems referred to earlier. While in tion program has simultaneously to: an economy with mostly private ownership the exposure of state-owned enterprises to effective competition * create a stable macroeconomic environment; through trade liberalization and domestic deregulation * create a market-driven incentives system, togeth- may be more important in achieving greater efficiency in er with greater flexibility in the factor markets; and these enterprises than privatization is (Douglas 1990), in * create a critical degree of decentralization and the Central and Eastern European economies with their competition, with private ownership of firms. overwhelming state ownership this pattern does not seem to hold, as evidenced by the failure of previous re- These key elements are crucial, and there is not much form attempts. room for different sequencing. Most elements have to be present from the outset, and the government's commit- Key Elements of an Economic Transformation ment to the comprehensive transformation has to be Program for Central and Eastern European credible. At the same time, implementation of the differ- Economies ent elements will follow different time paths; in particu- lar, privatization and institutional development to support a market economy, especially the development Given the list of typical policy issues to be addressed of financial markets, greater flexibility in the labor mar- in the Central and East European economies and the ket and upgrading of the legal system, will take longer main lessons from the World Bank's experience with ad- than some other elements. It is, therefore, all the more justment lending, what should be the key elements of an important to pursue all elements simultaneously from economic transformation program that can reasonably the outset. This approach appears to be crucial to bring- be expected to succeed? ing about the necessary behavioral changes, particularly Most economies with unsustainable balance-of-pay- in enterprises (International Institute of Finance 1990; ments positions requiring adjustment typically face both Kornai 1990). major macroeconomic and microeconomic distortions A stabilization program may be required at the begin- that need to be attacked simultaneously. While the pro- ning of the transformation because of high or hyper-infla- grams would obviously vary from country to country, tion, as in the cases of Poland and Yugoslavia and to a they would follow a by now traditional pattern that en- lesser degree Hungary and the USSR. Even where infla- tails varying doses of stabilization cum adjustment, typ- tion is not a major problem at the beginning of the trans- ically supported by a series of International Monetary formation, a stabilization program involving tight fiscal 50 and monetary policies and perhaps a more radical mone- World Bank-Supported Programs in Central tary reform to control or eliminate the monetary over- and Eastern Europe hang is still needed. Significant price corrections, including in most cases a depreciation of the exchange World Bank adjustment lending in the region began rate and an upward adjustment of interest rates, are re- with the first SAL to Yugoslavia in 1983. The operation quired, to be effected through the elimination of subsidies was not successful, and much time elapsed before the and the liberalization of trade so as to align domestic with second SAL was approved in April 1990, after a prepara- international prices. These measures could easily trigger tion period of more than three years. In Hungary, adjust- high inflation if not controlled through rigorous fiscal ment lending was initiated in 1988 with the Industrial and monetary management. Sector Adjustment Loan (a sector adjustment loan, or A major realignment of relative prices through the SECAL), followed by a SAL, approved in June 1990. elimination of subsidies, tax reform and liberalization of (Quasi-adjustment lending in support of industrial re- domestic and international trade, and the imposition of structuring had actually started in 1986 with a series of financial discipline on firms through a reformed finan- sector operations-Industrial Restructuring I, II and cial system, would have to be the core elements of any III-with the latest approved in 1989.) In Poland, World program of transformation aimed at establishing a mar- Bank lending was initiated in February 1990 with two ket-based system. However, only if there is greater factor project loans. An additional three project loans followed mobility will the aforementioned measures lead to a re- before the first SAL was approved in July 1990. allocation of resources and greater efficiency. Function- This section highlights the main features of all ad- ing labor and capital markets are indispensable. justment lending operations in the region and provides Persistent rigidity in the factor markets could cause the brief assessments, mainly of program design, in the con- macroeconomic stabilization to translate into unduly text of the normative framework specified in the preced- high losses in output and loss in employment. ing section.1 A few words are also added on the World As experience in Hungary, Yugoslavia and-to a less- Bank's recent recommendations for reform in Czecho- er extent-Poland suggests, these core microeconomic slovakia. A summary assessment is provided in the con- reforms by themselves have little chance of producing cluding section. the expected increase in efficiency unless complemented by a massive transformation of ownership and manage- Yugoslavia ment that makes financial discipline credible. As to privatization, it is a long-term proposition in light of the The first SAL to Yugoslavia was made in June 1983 in sheer proportions of state ownership in Central and support of the government's adjustment program. The Eastern Europe. Nevertheless, a critical mass of private program, entitled Long-Term Program of Economic Sta- firms has to be achieved in a relatively short period. bilization, was based on the recommendations of a high- Based on World Bank experience in other countries, level Commission on Stabilization. Its focus was the these changes can be greatly facilitated by mitigating strengthening of market forces in the economy, and its any resulting social hardship through specific support main components were: measures for the most adversely affected. That is, ade- quate provision for a social safety net, in particular un- * more liberalized external trade policies and allo- employment benefits complemented by employment cation of foreign exchange; services and training programs, should be made from * improved pricing policies and enterprise deci- the outset. sion-making, with particular emphasis on the agricul- This section has focused on the requirements of do- ture and energy sectors; and mestic policy reform. However, the external dimension * adjustment of interest rates and improvement of is equally important. To succeed, adjustment programs investment planning and allocation. need to be adequately financed. As domestic savings will tend to fall short of investment requirements, external The program also included a traditional stabilization financing on adequate terms will be needed. In the cases effort that was supported by an IMF Standby Arrange- of Poland and Bulgaria, adequate external financial sup- ment. port also has to include a lasting solution to the debt The reforms faltered, particularly as the stabilization overhang. effort could not be sustained for a sufficient period and 51 inflation accelerated. During 1985-86, there was consid- Other key structural elements incorporated into this erable policy backsliding, and by mid-1987 Yugoslavia new program are: had essentially fallen back to the situation in 1982-83, when the stabilization and policy reforms were initiated. * further liberalization of the foreign exchange and In addition to the failure to maintain an adequate trade regimes and concomitant domestic price liberal- macroeconomic environment, which in itself hampered ization. the effectiveness of the structural reforms, the program * enforcement of financial discipline by enterprises supported by SAL I tried to address too many issues and through restructuring and tighter bankruptcy proce- was not well-focused on the most crucial ones, such as dures, together with an adequate interest rate policy. In reform of enterprise management, enforcement of fi- the short run, the enterprise adjustment will be backed nancial discipline, improvement of the financial system, by non-inflationary fiscal support to the enterprises-a development of a new approach to the employment of la- sharp departure from the past when enterprises were au- bor, and, ultimately, changes in enterprise ownership tomatically financed through the banking system and and the approach to enterprise management. These ultimately the central bank. shortcomings were not the result of absent or misguided * broader-based fiscal financing of the social safety analysis. Thorough analysis by the high-level commis- net to protect displaced workers without placing a fur- sion and by Bank staff had preceded adoption of the re- ther burden on productive enterprises. This measure form program. The lack of success was ultimately also will be complemented by specific programs to promote the result of the political situation in Yugoslavia, which small private sector enterprises. led to both lagging implementation and eventually to abandonment of the program. At that time Yugoslavia The program addresses the most crucial issues con- was not really committed to the transition to a market- tributing to the poor performance of the Yugoslav econ- based system, and-within the existing legal frame- omy and closely follows the normative framework set work-the federal government was severely limited in out in the previous section. Moreover, the program is be- bringing about the interregional consensus required for ing strengthened by expanding its coverage to enterprise effective implementation of major reforms. reform, attention to the financial sector issues and prep- SAL I had been conceived as the first in a series of aration of the ground for bank restructuring, as well as such loans. It took, however, about seven years to agree further rationalization of unemployment benefits and on a follow-on operation, which for all practical purpos- improvement of the functioning of the labor market, es could be considered as a new operation. This long hi- with additional support from the World Bank. atus reflects the difficulty of designing and agreeing on Key questions remain, however. Given the historical- a program that would address the key issues in a mean- ly rooted political difficulties, will program implementa- ingful way and provide for a reasonably stable macroeco- tion be effective? Can the ultimate and systemic root nomic environment after the inflation degenerated into cause of Yugoslavia's economic calamity-the lack of hyper-inflation in 1989. That time was not entirely lost, clear enterprise ownership and concomitant manage- however. In the interval the government made substan- ment-be addressed effectively, given the possibility of tial progress in amending the constitution and strength- having to liquidate and restructure enterprises and dis- ening the institutional setting for enterprise miss workers. While the federal government is pursuing management and the financial sector. It adopted key such policies, it has yet to demonstrate that the new pieces of legislation such as a new enterprise law (enact- rules spelled out in recent legislation will be applied and ed in 1988) and a new banking law (enacted in 1989), further strengthened. Liquidations and bankruptcies, al- along with several other important laws affecting invest- though obviously not the objectives of the program, are ment. the only credible threats to enterprises to enforce finan- Yugoslavia's current adjustment program, which is cial discipline and thus improve market behavior, which being supported by SAL II, was initiated along with a are key factors in securing macroeconomic stability and radical and far-reaching heterodox stabilization pro- efficient resource use. The credible imposition of finan- gram, with support from the IMF. Contrary to previous cial discipline has major implications for the manage- stabilization attempts, the current program also ad- ment of enterprises and will necessarily require a dresses structural issues such as adjustment of the en- significant "cultural" change, particularly in the labor terprise and banking sectors, which were seen as two of market: through competitive pressure in the goods mar- the root causes of the financial instability and poor eco- kets, firms will be forced to cut costs, an eventuality that nomic performance in Yugoslavia. will no longer permit maximization of the wage fund and 52 other distributable surpluses but rather will lead to the provided an easy-and from an enterprise's point of dismissal of excess labor. view, profitable-way out. Enterprises participating in the restructuring program actually increased their ex- Hungary ports to CMEA markets rather than to convertible cur- rency areas. World Bank adjustment lending in Hungary started Around the end of 1989 the interim government with the approval of an industrial sector adjustment loan strengthened the impetus for reform, as did the govern- (ISAL) in 1988 to support the industrial policy reforms ment that took power in 1990. Most importantly, a more the government planned for the period 1985-90. The rigorous approach to macroeconomic management was ISAL was actually preceded by two sector operations, the put in place based on stricter fiscal and monetary poli- Industrial Restructuring Loans I and II, approved in cies, including adjustments of interest rate and foreign 1986 and 1987, respectively, and was followed by the In- exchange policies. At the same time, the structural ele- dustrial Restructuring Loan III in 1989. All three re- ments aimed at more disciplined enterprise behavior structuring loans obviously also support the program to were strengthened. The revised and strengthened pro- reform industrial policy through specific restructuring gram is being supported by an IMF Standby Arrange- programs at the enterprise level. The first two operations ment and a World Bank SAL, approved in June 1990. The were hybrid, including quick-disbursing components for World Bank is also providing complementary support balance-of-payments support against a set of policy mea- through previously approved industrial restructuring sures that had been designed to improve macroeconom- loans and a recently approved loan for the moderniza- ic and sector performance. tion of the financial system. The industrial policy reform program has been part of The revised program's main structural features are a larger reform effort aimed at macroeconomic stabiliza- the following: tion and greater overall efficiency, centering around three key elements: * ownership reform and privatization based on the Law of Transformation enacted in June 1989 (and other * increasing domestic and external competition; laws) aimed at effective enterprise management by vig- * tightening financial discipline in the enterprise orous exercise of ownership rights; sector; and * promotion of competition through further, pro- * facilitating the mobility of capital and labor to en- gressive trade liberalization, domestic deregulation and able the restructuring of the industrial sector. price reform, greater participation of foreign invest- ment, and tightened financial discipline, to be enforced The program design, as modified to deal with the through stricter application of the bankruptcy legisla- shortcomings observed in 1985-87, stressed the links to tion and with the help of a financial sector now being macroeconomic performance by: concentrating on the strengthened for more effective intermediation; most inefficient subsectors and enterprises receiving the * eduction of the state's role in the economy, most highest budgetary support; cutting producer and con- notably through a sharp reduction in subsidies and the sumer subsidies; strengthening tax reform; and provid- limitation of public investment to core areas of the phys- ing incentives for convertible currency exports. ical and social infrastructure, in addition to privatiza- However, program implementation through the end of tion; and 1989 was not very successful. Macroeconomic manage- * major reform of social policies, including hous- ment continued to falter, and excessive current account ing, health, social security and unemployment compen- deficits in the balance of payments and increasing exter- sation. nal financing difficulties emerged. Lax macroeconomic management in the face of interest rates that had not Besides vigorous macroeconomic management reached adequate levels also had direct negative implica- aimed at lower inflation and a sustainable balance of pay- tions for enterprise behavior. Moreover, enforcement of ments, the core of the program deals with the reform of financial discipline by enterprises lagged, as the liquida- enterprise behavior. The design of the program seems to tion and restructuring plans for important enterprises be broadly in line with the norms postulated in the pre- were delayed. The lagging structural reform in turn fed vious section. The proof will be mainly in the implemen- back into the poor macroeconomic performance, which tation. The program addresses the root causes of further weakened the reform. inefficiency in Hungary, except perhaps for import liber- An additional factor explaining the lagging restruc- alization, which, after many previous attempts, is still turing has been the CMEA trade arrangements, which relatively modest in scope and phased over a number of 53 years. Current macroeconomic management has been transformation, Poland has to find a longer term solu- quite successful in improving the balance of payments tion to its severe debt problem. (which has gone from a large and increasing deficit to a Poland joined the World Bank in 1986. Because of its slight surplus in the current account in the first half of macroeconomic and debt problems, initially it was not 1990) but less successful in controlling inflation, which considered creditworthy for any lending from the Bank, has actually accelerated. With regard to enterprise be- and a credible stabilization and adjustment program was havior, the acid test will be that liquidations, bankrupt- a prerequisite for initiation of any lending. The Solidari- cies and privatization actually take place, supported by ty-led government of Prime Minister Mazowiecki pre- intensified domestic and foreign competition. The main pared the ground to receive Bank financing when it battle is still ahead, and, judging by the authorities' rel- adopted the Economic Transformation Program on Jan- atively modest assumptions about future unemploy- uary 1, 1990. That program, which is comprised of a ment, it would seem the task of restructuring has been strong heterodox stabilization component and far-reach- either substantially underestimated or political compro- ing structural adjustment measures, is being supported mises may be likely regarding the pace of the adjust- by a standby arrangement from the IMF and a SAL from ment. the World Bank, approved in July 1990, following initia- There is, as noted, no substitute for government own- tion of project lending in February. Complementary ership, i.e., the government is ultimately responsible for lending operations to support, inter alia, industrial re- the design and implementation of its reform efforts. structuring, financial sector development, agricultural However, to the extent that program design and imple- development, and strengthening of the labor market and mentation can be strengthened through loan condition- employment promotion are also under preparation. ality, the SAL builds on prior experience in Hungary and The Economic Transformation Program is Poland's general lessons learned from World Bank adjustment most comprehensive and radical attempt to date to sta- lending. The government took important actions prior bilize and set the stage for transformation into a market to loan approval, and more specific actions of a crucial economy. In its combination of macroeconomic and nature are required for the release of the loan's second structural or systemic elements, it is certainly the stron- tranche. The authorities designed the macroeconomic gest program among the three described in this paper. framework for the overall program in close consultation The stabilization component, which is similar to Yugo- with the IMF and the Bank. Monitoring of performance slavia's, aims to eradicate inflation, after prior price lib- will be based on specific indicators, and there has to be eralization, through the pursuit of tight fiscal and agreement on the future policy course, including the monetary policies, with a drastic cut in subsidies and a 1991 budget (with special emphasis on the reduction of major adjustment of interest rates, as well as through producer and consumer subsidies) prior to release of the the use of a fixed exchange rate and partially indexed second tranche. Similarly, strong and credible action on wages as nominal anchors. The structural or systemic privatization, enterprise liquidation and trade liberaliza- component goes beyond Yugoslavia's or Hungary's, how- tion is required for release of the second tranche. ever, in terms of the liberalization of imports and prices and the enforcement of financial discipline of enterpris- Poland es. Finally, contrary to the other programs, Poland's pro- gram aims at comprehensive debt relief from all To some extent, Poland is unique: in addition to the creditors, a move that appears indispensable to its viabil- macroeconomic, sector and systemic problems it shares ity and to laying the groundwork for the restoration of with its neighbors in the region, its creditworthiness is growth and creditworthiness. seriously impaired. It has not been able to service its debt More specifically, the stabilization component of the with Paris Club creditors since 1982 and has undergone Economic Transformation Program comprises: a virtu- five reschedulings, the latest one in February 1990. ally balanced budget, to be achieved mainly through a While Hungary also is highly indebted (on a per capita drastic reduction of subsidies; strict control of credit; basis even more so than Poland), it has managed to stay positive real interest rates; an almost complete liberal- current with its debt-service obligations. Poland has also ization of prices (including major adjustments of still ad- had to reschedule its commercial debt on seven occa- ministered prices, particularly in energy); radically sions; until about a year ago, it managed to pay the inter- changed foreign exchange and trade regimes, with com- est on all commercial debt but since then has suspended plete import liberalization and a fixed exchange rate af- payments on medium- and long-term debt, which ac- ter significant devaluation; and a substantial reduction counts for about 90 percent of total commercial debt. in real wages. Thus, in addition to the "standard" issues of economic 54 The structural adjustment or systemic component of (as in the other countries) largely depends on enforce- the Economic Transformation Program comprises three ment of the rules and, thus, on determined initiation of main features: liquidation and bankruptcy procedures. While there have been a number of liquidations, generally speaking * enterprise restructuring, privatization and devel- enforcement has lagged. opment of the private sector, with appropriate legisla- Credit ceilings and mandatory dividend payments are tion to impose financial discipline, establishment of an essentially short-term devices to enforce financial disci- adequate institutional framework for enterprise restruc- pline. In the longer run, a well-functioning and indepen- turing (including stricter environmental standards) and dent financial system as well as increased private preparation of the ground for privatization; ownership of enterprises are more effective guarantees. * financial sector reform, with emphasis on im- Both elements are included in the Economic Transfor- proved banking regulation and supervision, introduc- mation Program, although both are longer term propo- tion of adequate accounting and auditing standards (also sitions. for non-financial enterprises), and strengthening and Implementation of the program's structural compo- phased restructuring of the banking system; and nent lacks vigor, a situation that may also explain the * establishment of a social safety net, with emphasis stubbornness of inflation, which, after having dropped on adequate unemployment benefits, employment ser- sharply from 78 percent in January, hovered around 3-5 vices, training, a program of minimum social assistance percent per month during May-September. The econo- and improved health policy. my is highly concentrated, and monopolistic structures dominate large segments, including the distribution sys- Enterprise restructuring and privatization are the tem. Moreover, the persistence of energy subsidies and a Economic Transformation Program's most crucial ele- somewhat undervalued exchange rate (in spite of the ment. These measures presuppose a reasonable degree of real appreciation since the beginning of 1990) may have competition, to be achieved through the complete liber- provided some sectors with a relatively comfortable ini- alization of imports and demonopolization of (or en- tial cushion that diminished the effectiveness of import hancement of competition in) the domestic economy, as competition. well as a functioning financial system. The demonopoli- Again, as in the other cases, SAL conditionality was zation and financial sector development need to be ac- designed to strengthen the government's hand in imple- celerated to make enterprise restructuring and menting the Economic Transformation Program-to privatization effective tools in increasing overall efficien- the extent strengthening can be achieved through out- cy. side support or requirements. The Bank will closely The poor performance of enterprises in Poland, as in monitor macroeconomic performance based on a set of other socialist economies, was the result of lax financial key indicators, which also stress the medium-term di- discipline (or the soft budget constraint, in Kornai's rnension through an emphasis on savings and invest- words), i.e., easy access to bank credit at subsidized in- ment. Importantly, the macroeconomic conditions for terest rates, with the funds ultimately originating from release of the second tranche include agreement on the the central bank's printing press. The Economic Trans- 1991 draft budget, with particular emphasis on a further formation Program seeks to address financial indisci- reduction in the energy, housing and transportation pline in the short run in two ways: by limiting the overall subsidies, as well as on adequate investment. An effective availability of credit and raising interest rates signifi- external financing strategy-which in the case of Poland cantly; and by enforcing the payment of a mandatory div- essentially means debt and debt-service reduction-is idend on that portion of a firm's revalued capital that was another important aspect of the macroeconomic frame- originally provided by the founding organization. The work. Key structural actions to support the drive toward dividend provision, which is stronger than in the past, is a market-based system and greater efficiency include: a significant financial requirement, particularly as the vigorous action on demonopolization in key sectors, availability and cost of credit are additional constraints. such as agricultural marketing, distribution of inputs, Failure to pay the dividend to the government would transportation and trade; a review of import tariffs and trigger a number of possible actions, including restruc- the remaining export restrictions; accelerated enterprise turing or liquidation. A possible avenue to restructuring restructuring and privatization based on sound policies could also be privatization. and operating principles for the respective institutions; The institutional framework to establish financial dis- and prudential bank supervision based on new account- cipline is relatively strong. However, actual performance ing and auditing standards. Ultimately, however, it is ob- 55 viously the government's commitment to the program However, based on the experience of other countries, ul- and political support that will determine the relative timately success in transforming the economy will de- success of its implementation. pend on the decisive implementation of a critical mass of simultaneous measures from the outset. Czechoslovakia Conclusions Czechoslovakia became a member of the IMF and the Bank only in September 1990. While the previous re- Given the similarity of the macroeconomic, sector gime over the years had undertaken some economic re- and systemic problems in various Central and Eastern forms, Czechoslovakia was much closer to a centrally European countries, it is not surprising that the ap- planned system than were the other countries. The new proaches to reform and the focus of IMF and World Bank government has initiated a few reforms, such as the es- support are quite similar across countries. Obviously the tablishment of a two-tier banking system and limited magnitude of the macroeconomic adjustment and the price liberalization. However, the big push for reform is kind of approach (with or without the use of nominal an- yet to come, most likely at the beginning of 1991. chors) varies from country to country, as does the rela- As compared with other Central and Eastern Europe- tive emphasis on different structural components such an countries, Czechoslovakia has managed its economy as trade liberalization, enterprise restructuring, finan- more prudently, avoiding high inflation, severe shortag- cial sector reform and privatization. es, balance-of-payments problems and excessive indebt- Experience with previous reform attempts suggests edness. However, grave price distortions, low efficiency that a critical mass of simultaneous efforts is required and heavy dependence on the defense industries and for ultimate success. This prescription applies in partic- arms exports, as well as on CMEA trade, make a compre- ular to the stabilization efforts currently being pursued, hensive approach to economic reform equally urgent. which need to be complemented by structural adjust- Similarly, given the need for major price corrections, ment. Short-term macroeconomic stabilization turns stringent macroeconomic management is indispensable out to be relatively easy: Hungary, Poland and Yugosla- to ensure that the one-time price adjustments do not via have achieved encouraging results in terms of their trigger an inflationary spiral. balance-of-payments performance, although somewhat The World Bank recently completed an Economic less so in terms of controlling inflation. By far the more Memorandum on Czechoslovakia. In line with the gen- difficult task, however, is to maintain financial stability eral lessons from adjustment lending and the specific is- in the medium and longer term and to achieve greater sues in Eastern Europe as postulated in the previous efficiency through structural transformation and section, the Bank has recommended a program of simul- change in enterprise behavior. It is the behavioral taneous policy actions in a number of key areas within change of enterprises that in the end will determine the the context of a prudent and consistent set of macroeco- supply response. nomic policies aimed at creating a stable financial envi- In the light of experience, the enforcement of finan- ronment: cial discipline in the enterprise sector can be considered the acid test of all transformation programs in terms of * far-reaching liberalization of domestic prices; securing both macroeconomic stability (because of the * liberalization of current account transactions monetary implications of the soft budget constraint) and ("internal convertibility" ); greater efficiency of resource use. This summary assess- * changes in the rights and responsibilities of the ment of the programs therefore focused on the relative managers of state-owned enterprises, as well as initia- strength of that particular aspect of the individual pro- tion of privatization and private sector development; grams. * encouragement of private direct foreign invest- Enforcement of financial discipline is used here as a ment; and broad concept comprising a number of different fea- * provision of an adequate social safety net. tures, such as competition, a far-reaching change in ownership rights and the concomitant accountability of There has been considerable debate in Czechoslova- enterprise management, a credible threat of bankruptcy kia among various factions over the scope and pace of re- and liquidation, and ultimately large-scale privatization form. The scope and pace seem to be critical issues in to bring about a qualitative change in enterprise behav- Czechoslovakia, as the population does not appear to be ior. These features are prominent in all programs and are prepared for major sacrifices given the relatively smooth supported by World Bank SAL conditionality. However, if not impressive functioning of the economy in the past. program implementation, while still at an early stage in 56 all countries, has not been as vigorous as might have also seems to be crucial, not least for the required inflow been hoped considering the dimension and urgency of of managerial talent. the task. The relative inexperience of the administra- A critical complement enterprise reform is the devel- tions and the political controversy are obvious explana- opment of an effective and independent financial system, tions for the tardiness in implementing the reforms. with severance of the intimate relationship between Regarding competition, the Polish program has gone banks and non-financial enterprises, a link that is partic- the farthest in terms of import liberalization and the ex- ularly strong in Yugoslavia and-to a lesser extent-in posure of domestic industry to international competi- Hungary. While this proposition is long-term, important tion. Yugoslavia and particularly Hungary still have a regulatory measures can be implemented in the short long way to go. However, even in Poland, the remaining run to strengthen the system and make the banks more subsidies, particularly the energy price subsidies, and autonomous. Ultimately, however, privatization of the possibly a somewhat undervalued effective exchange rate banking system is also required. (taking tariffs into account as well) provide many enter- The experience of the former German Democratic Re- prises with a still relatively comfortable cushion. Com- public, which is undergoing the most radical and fastest bined with high concentration and a correspondingly transformation, warrants careful study. However, its high degree of monopolization that affects domestic dis- conditions are hardly replicable, as other countries lack tribution systems, a non-tradable activity, these factors external support, including direct investment and tech- have slowed adjustment. However, there are now the nical assistance, of comparable magnitude. first encouraging signs of some enterprise adaptation to Rapid adjustment and transformation have become the new market environment. even more urgent with the impending change in the The governments in all the countries have been rela- CMEA trade arrangements toward convertible currency tively slow to enforce bankruptcy and liquidation proce- and international prices and the most recent oil price dures, with Yugoslavia having gone the farthest. To some shock. Both events represent a shock on the order of 5 extent, enterprises have been operating on the back of percent of GDP for the Central and Eastern European increasing inter-enterprise credit; in addition, banks countries (excluding the USSR). While some additional have had little choice but to extend more credit to enter- external financing may become available, additional do- prises in financially precarious conditions, even if they mestic adjustment is also required. do not have a clear perspective on their own restructur- In summary, all the countries should accelerate their ing and recapitalization. In all the countries, changes in structural transformation to take advantage of the the nature of ownership rights, restructuring and priva- uniquely favorable environment at this historical mo- tization have also been proceeding slowly, mainly be- ment. The World Bank's support through SALs is built cause of political controversy over workers' acquired on strong conditionality to steer the process in the desir- management rights and expectations of partial or total able direction at the appropriate pace. It cannot, howev- ownership, adequate valuation of enterprises, lack of do- er, substitute for the governments' commitment and mestic capital and the role of foreign investment. A crit- implementation capacity. In addition, an adequate attack ical mass of private ownership of enterprises with on the root causes of the financial indiscipline and inef- effective management in a competitive market environ- ficiency is critical: failure to launch one has been a main ment seems to be the most effective guarantee for great- reason for the failure of past reform efforts. The relative er efficiency and financial discipline. While all the newcomers to economic reform such as Czechoslovakia, countries now have enabling legislation in place, actual Bulgaria, Romania and the USSR should carefully study privatization, except of very small enterprises, is lagging. failed attempts at reform in other countries and apply The task is, however, daunting, given the large number those lessons in designing and launching their own of enterprises and the lack of domestic savings and effec- transformation programs in the most comprehensive tive capital markets. Rapid sale of retail outlets and other way and at the fastest possible pace. small units that could be spun off from large conglomer- ates and a few large-scale privatizations involving for- Notes eign capital as well as free distribution of shares to the population at large, combined with effective enterprise 1. The discussion is based on information available through the control mechanisms through private mutual funds or middle of 1990. Thus, it is too early in all cases to judge program im- similar arrangements, are feasible short-run measures plementation and actual performance, although some weaknesses in similar arrangements, are feasible short-run measures program design and implementation have already surfaced and are that could make an important qualitative difference in mentioned. the transformation. Acceptance of foreign participation 57 References die East and North Africa Region (EMENA) Discussion Papers. Washington, D.C. Adam, J. 1984. Employment and Wage Policies in Poland, Czechoslo- vakia and Hungary Since 1950. New York: St. Martin's Press. Institute of International Finance. 1990. "Building Free Market Econ- omies in Central and Eastern Europe. Challenges and Realities." Aslund, Anders. 1990. "Systemic Change in Eastern Europe and East- Washington, D.C. April. West Trade." European Free Trade Association (EFTA) Occasional Paper No. 31. Geneva. June. Kornai, Janos. 1990. TheRoad to a FreeEconomy. Shifting from aSo- cialist System. The Example of Hungary. New York and London: Brus, Wlodzimierz, and Kazimierz Laski. 1989. From Marx to the Mar- Norton. ket. Socialism in Search of an Economic System. Oxford: Claren- don Press. Mencinger, J. 1989. "Economic Reform and Unemployment." Privrendna kretanja Jugoslavije 19 (March). Corbo, Vittorio, and Stanley Fischer. 1990. "Adjustment Programs and Bank Support, Rationale and Main Results." Mimeo. World Bank. Milanovic, Branko. 1989. Liberalization and Entrepreneurship. Dy- Washington, D.C. August. namics of Reform in Socialism and Capitalism. Armonk and Lon- don: Sharpe. Coricelli, F., and R. Rocha. 1990. "Tax Reform in Socialist Countries: A note on the main issues." Country Economic Department, Macro- World Bank. 1990a. Adjustment Lending: Policies for Sustainable economic and Growth Division (CECMG). World Bank. Washing- Growth. Policy and Research Series #14. Washington, D.C.: World ton, D.C. Bank. Douglas, Roger. 1990. "A Brief Account of New Zealand's Evolving . 1990b. World Development Report 1990. Baltimore, Md.: Economy." The United States-New Zealand Council Quarterly Re- Johns Hopkins University Press. port (Washington, D.C.) (Summer). _______.1990c. World Tables, 1988-89. Washington, D.C.: World Gomulka, S., and J. Rostowski. 1988. "An International Comparison of Bank. Material Intensity." Joumal of Comparative Economics 12. Yugoslavia. Various issues. Yugoslavia Statistical Survey. Belgrade: Hinds, Manuel. 1990 . "Issues in the Introduction of Market Forces in Research Department, National Bank of Yugoslavia. Eastern European Socialist Economies." World Bank, Europe, Mid- 58 5 Comments on "World Bank Adjustment Lending in Central and Eastem Europe," by Ulrich R. W Thumm Miroslav Hrncir Just one decade has passed since the World Bank intro- tensive, such as Spain or Chile several years ago. It also duced its program of structural adjustment lending. Ten requires simultaneous basic institutional and systemic years ago the growing pressures of external indebtedness changes. Market agents proper and the entire framework that resulted from the profoundly changing conditions in of market institutions need to be created in the course of the world economy and the failure of development poli- mass privatization. The result will be a corresponding cies in a number of developing countries prompted the change in behavior patterns and social values. The rele- World Bank to initiate this program as a more policy-ori- vant issue is thus a substitution of the existing system, ented and comprehensive approach to addressing the and not its reform. The concept of reform that involves evolving challenges, in addition to its traditional project- changing only some of the elements of the given system type programs. Ten years later, toward the end of the does not, therefore, seem relevant any more. 1980s, dramatic socio-political changes are taking place Experience with and evaluation of this type of transi- in the countries of Central and Eastern Europe, as they tion have been limited. Not surprisingly, there is no es- endeavor to transform their economies into market-type tablished theory. Considerable knowledge has, however, ones. gradually been accumulated that perhaps represents a What role should the World Bank play in that trans- rudimentary stage of theory. The analysis presented in formation? Are the experiences with and lessons from the paper suggests a number of relevant conclusions in adjustment lending in other parts of the world relevant that respect: and applicable? What are the common grounds and what are the basic differences in both given conditions and the * all the transition programs surveyed in the paper aims being followed? (those of Yugoslavia, Hungary and Poland) aim at imple- The paper presented by Ulrich R.W. Thumm provides menting comprehensive sets of measures that should a comprehensive analysis of the relevant issues. From guarantee achievement of the desired qualitative the perspective of an observer living in Central Europe, change. The scope of these sets of measures suggests the paper succeeds in identifying and evaluating the con- that the governments are trying to avoid the shortcom- ditions and factors calling for structural adjustment ings and constraints of the partial approaches to reform lending. pursued in the past. The comments made here focus on three issues spe- * macroeconomic stabilization is treated as aneces- cifically. The first concerns the substance of the transi- sary precondition for the implementation of institution- tion being undertaken by the previously centrally al and systemic changes and the development and planned economies. The paper makes the valid point efficient functioning of markets. that its dimensions and goals are far more complex and, consequently, more demanding than has been the case All the programs are trying to combine the traditional in other parts of the world. Their transition is unique in stabilization policies of demand management (through the sense that it requires a total change in their econom- restrictive fiscal, monetary and credit policies) with less ic and social regimes. The aim in the transition of the orthodox policies on the regulation of wages and person- economies of Central and Eastern Europe is not only to al income, as well as deliberate policies of restructuring stabilize, deregulate and liberalize, as was the case in and promotion of a supply-side response. This heterodox countries where govemment intervention had been ex- approach is expected to be more productive, particularly 59 given the conditions with which the countries are start- gained its membership in the International Monetary ing their reform (monopolistic structure, rigidities and Fund (IMF) and World Bank only in September 1990, is low transparency within the domestic economy, clear not addressed in detail. The initial macroeconomic con- property rights, which mostly are still lacking, lax finan- ditions in Czechoslovakia appear to be more favorable cial discipline and low responsiveness of enterprises to than those of the other countries examined, particularly indirect policy measures). Under the circumstances, tra- with respect to two points: so far, Czechoslovakia has ditional demand management is likely to be only partly maintained a considerable degree of domestic macroeco- relevant and/or rather costly in terms of the level of eco- nomic stability; and it has a relatively low level of foreign nomic activity, employment and real wages. debt. These factors are often claimed as Czechoslovakia's The income and employment levels at which the comparative advantages. As a result, its stabilization pol- trade-offs among the various policy objectives will be icies do not have to cope with galloping inflation, as was achieved are thus conditioned on the outcome of both the case in Yugoslavia and Poland. Rather, the task is to the restrictive measures on the demand side and the pol- avoid such inflation during the transition. icies on the supply side and on the restructuring. Achiev- On the other hand, Czechoslovakia's rate of growth has ing a proper balance between them in the reform been declining over a longer period than in the other packages is likely to be key to the success of the stabili- countries-signs of stagflation emerged in the 1980s zation efforts in the medium and longer terms. At the (when hidden and repressed inflation is accounted for), as same time, the transition to a market-type economy re- did a fall-off in its share of exports to the world markets. quires a multi-dimensional approach, with stabilization Thus, it was maintaining macroeconomic stability at the policies being just one element. The other interrelated cost of the future. The high "intemal" debt it incurred is components of a transition package are institutional and manifest in the obsolete capital stock, neglected infra- systemic changes and policies for restructuring and so- structure and environmental needs. A disproportionate cial adjustment, including the creation of social safety emphasis on heavy industry and a one-sided orientation nets and changes in the social values pursued. toward trade within the Council on Mutual Economic As- The paper shows that, despite the different initial con- sistance (CMEA) region contributed to structural rigidi- ditions in both the economic and socio-political spheres, ties and a low capacity to adjust to the changing pattem of the adjustment programs of Yugoslavia, Hungary and demand. As is rightly stated in the paper, Czechoslovakia Poland are more or less the same. More precisely, they is an example of "regressive specialization" : over time it share the same aims and components, even if the rela- replaced higher value-added exports with lower ones, tive weights attached to them vary. such as raw materials and intermediate products. The multi-dimensional character of the transition Another point is that the institutional and socio-po- implies that proper timing and sequencing of the vari- litical framework of Czechoslovakia has corresponded ous policy steps as well as differential time horizons are more closely to the traditional type of centrally planned issues. The paper does not elaborate on this point. It is economy than was true in the reforming economies of therefore important to distinguish here between stabili- Hungary and Poland, even toward the end of the 1980s. zation measures that can be introduced in one stroke Another problem specific to Czechoslovakia is that (and shock therapy is an option for the three countries) the external shocks such as the considerable increase in and those that involve institutional and systemic chang- oil prices and the collapse of the CMEA institutions and es that could become effective only over a certain time regional trade coincided with the introduction of a set of horizon. The same applies to changes in social attitudes measures aimed at deregulating and liberalizing the and values. pricing and foreign exchange regimes, reforms that The second point refers to the relation between com- should represent a turning point on the road to a market mon principles in the transition processes and country- economy (including the "internal" convertibility of cur- specific conditions and factors. In emphasizing the com- rency, i.e., current account transactions of residents). mon features of the transitions in the previously central- These factors and conditions should be reflected in ly planned economies, it is important not to generalize both the plan Czechoslovakia adopts for its transition too much. The unique conditions of each country must and the timing and sequencing of its implementation. always be borne in mind. These conditions must be iden- The third comment refers to the role of the World tified and accounted for if the economic and social poli- Bank in the transition of Central and Eastern Europe. cies for the transition are to be as feasible and efficient as The efficiency of any national economy has two key di- possible. mensions: efficiency in the allocation of resources in The paper discusses the Yugoslav, Hungarian and both static and dynamic terms; and the efficiency of re- Polish cases extensively. Czechoslovakia, because it re- source use, or technical efficiency. The two dimensions 60 are interrelated. However, it seems that what is most justment lending to the achievement of the goals and critical to the future development of the previously cen- priorities being pursued, that is, the extent to which it trally planned economies are the extent and pace of re- contributes to the pace of the reforms and to the adop- structuring, product innovation and technological tion of measures that would not otherwise be imple- change. Here is where the command-type economy mented. Such a hypothetical evaluation is, however, very proved most detrimental. ambitious and perhaps not even viable, especially in con- However important stabilization policies are to a suc- nection with the evaluation of policy and institutional cessful transition, the main task-an even more de- changes linked to structural adjustment lending. manding and prolonged one-is a successful structural A possible option is to look at the degree to which adjustment. It requires a change in the entire regime af- countries have implemented the structural adjustment fecting the behavior of economic agents, including the measures agreed to, including the time horizon of im- driving forces and incentives that underlie workable plementation. An alternative criterion is to assess the ex- competition and entrepreneurship. So far the other tent to which the Bank has proceeded efficiently, given countries in Central and Eastern Europe have achieved its mandate, that is, the amount of resources it has dis- only modest progress with structural adjustment; none bursed in light of conditions in a particular country. is an obvious success story yet. Structural adjustment is, At the same time, there is merit to the view expressed however, a necessary precondition for sustained growth in the paper that structural adjustment lending should and development. be accompanied by, or even better preceded by, elimina- The structural adjustment lending of the World Bank tion of the major distortionary factors, particularly should help the countries of Central and Eastern Europe through the liberalization of prices and foreign ex- cope with the problems of transition from the very be- change and through tax reform. The World Bank should ginning of their efforts. Given the rather distorted data support those measures. However, ultimately their im- base and low level of transparency of the economies, one plementation and the credibility of the effort are the re- issue for the Bank is what criteria it should use to assess sponsibility of the governments of Central and Eastern the success of its activities in this area. The relevant cri- Europe. terion should be the real contribution of structural ad- 61 I 6 Comments on "World Bank-Supported Adjustment Programs: Lessons for Central and Eastern Europe, " by Vittorio Corbo and "World Bank Support for Adjustment Lending in Central and Eastern Europe" by Ulrich R. W Thumm Stanislaw Gomulka The paper by Vittorio Corbo discusses the World cumstances many enterprises, which together consti- Bank's vast non-European experience with adjustment tute a sizable part of the economy, would face lending in terms of policy objectives and actual econom- bankruptcy have proved unfounded. Their profit mar- ic impact. The paper by Ulrich Thumm concentrates on gins, which were already high before 1989, increased the Bank's policies for Eastern and Central Europe, tak- considerably in the autumn of 1989 and the spring of ing into account the specific systemic features of the 1990. The financial data for the largest 500 Polish enter- post-Communist economies. Both papers provide excel- prises, which together accounted for about 65 percent of lent analyses of significant value to policy-makers. The industrial output in the years 1983-87, also suggest that comments made here are limited to selected economic most state-owned enterprises are both inefficient and points and a few policy implications. profitable. The loss-making enterprises in those years The economic competition between the capitalist and were concentrated largely in the food processing indus- socialist systems has centered around the importance of try because low (fixed) prices were set for products for private ownership. In the absence of a convincing eco- social and political reasons. nomic theory, the competition had to be settled by a It is true that financial discipline has been more lax practical test, whose outcome is now clear. However, un- under socialism than under capitalism, a systemic fea- derstanding of the precise reasons for the failure of econ- ture that must have resulted in efficiency losses. In an omies based on non-private ownership to perform well is exchange with Janos Kornai, the present discussant ar- incomplete. Thumm stresses the inability of central au- gued that these losses, and not shortages, have been the thorities to impose financial discipline on enterprises as major effect of the soft budget constraint (Gomulka the key intermediate reason.1 1986, chapter 5). However, this source of inefficiency, He attributes this poor discipline to lax bankruptcy given the evidence of good profitability, must have been procedures and subsidized interest rates, both of which weaker than the other ownership-related sources of inef- are real and distinct features of these economies. How- ficiency. Thumm himself notes the presence of ineffec- ever, Thumm himself notes the much improved enforce- tive enterprise management and insufficiently rigorous ment of financial discipline in state-owned enterprises exercise of ownership rights by the state. This point can under the Polish economic transformation program. be taken further, with emphasis on the lack of sufficient The government has achieved this discipline through motivation for bright innovators and entrepreneurial the withdrawal of tax concessions, a policy of non-nega- managers to perform exceptionally well as a key factor; tive real interest rates, a drastic reduction in other enter- their efforts would be wasted or socialized. In the ab- prise subsidies and a clear signal to enterprises that the sence of such efforts, the competitive pressures and as- central bank was withdrawing from its role as lender of sociated threats to income are low even if the number of last resort. enterprises were large, which it has not been. It there- It can be argued that the credibility of the new policy fore seems that the primary reason the socialist econo- of a hard budget constraint cannot be sustained for long mies have been and are inefficient is precisely this lack if public ownership continues. There is, however, little of strong positive and negative incentives to do well.2 evidence of a markedly improved performance by these Corbo notes that the origin of the World Bank's in- enterprises. Nor is there any need to put the policy to a creased effort to promote structural adjustment dates serious test. The original fears that under the new cir- back to the oil price shocks of 1973 and 1979. These were 63 externally imposed shocks with negative balance-of-pay- World Bank's lending policy had in mind. These differ- ments implications. The main purpose of the World ences in circumstances call for a change in the standard Bank's involvement has been to help eliminate the bal- objectives and performance criteria, and perhaps also for ance-of-payments problems through structural adjust- a lengthening of the period within which these objec- ment. The economic circumstances of Eastern and tives are to be implemented. Central Europe are, however, quite difficult. In both Po- land and Yugoslavia the shocks were internally and de- Notes liberately imposed. The World Bank's structural adjustment loans (SALs) came after the introduction of 1. Essentially Thumm says the poor performance of enterprises in the programs sponsored by the International Monetary Poland, as well as in other socialist economies is the result of lax finan- Fund (IMF), which were intended rapidly to create a sta- cial discipline, that is, easy access to subsidized bank credit that origi- Fund (IMF),,which were intended rapidly to create a sta nated at the printing presses of the central bank. Budget deficits and ble macroeconomic environment and a properly func- massive printing of money have only recently become major features tioning price system. These programs produced of socialist economies, recessions and therefore balance-of-trade surpluses. 2. For a detailed discussion of the poor environment for innova- Moreover, the World Bank's SAL negotiators typically tion under socialism see, among others, Balcerowicz (1990), Hanson suggest that governments undertake further rapid price and Pavitt (1987) and Gomulka (1990, chapter 7). liberalization and complete elimination of subsidies. The countries of Central and Eastern Europe would References like to proceed with these measures, and to do so with Balcerowicz, L. 1990. "The Soviet Type Economic System, Re- considerable speed. However, it should be recognized formed Systems and Innovativeness." Communist Economies 2 (1). that price adjustments will have an additional recession- ary and inflationary impact. The upshot is that because Gomulka, S. 1986. Innovation, Growth and Reform in Eastern Eu- in Central and Eastern Europe the SAL comes during a rope. Madison, Wisc.: Wisconsin University Press. deep recession and improved external position, it is dif- 1990. The Theory of Technological Change and Econom- ficult to justify on its own terms as a balance-of-pay- ic Growth. London and New York: Routledge. ments support program. Put simply, the structural adjustment problem in Hanson, R, and K. Pavitt. 1987. The Comparative Economics ofRe- Central and Eastern Europe is much different in origin search Development andinnovation in East and West:A Survey. Cam- and larger in size than the problems the founders of the bridge, Mass.: Harvard Academic Publishers. 64 Part III Speed of Adjustment and Sequencing of Policy Changes 7 Issues in the Reform of Socialist Economies Stanley Fischer and Alan Gelb' The formidable challenges facing the reforming Euro- However, it is noted that the important strategic choices pean socialist economies are frequently said to be unique. in the reform arise out of the interplay of economics and In fact, most of the individual requirements for their re- politics: technocratic solutions alone will not be suffi- form have been faced before in China and in Latin Ameri- cient in deciding on the path and speed of reform. can and African countries where the combination of a Because the reform is both intertwined with political weak private sector, political monopoly, extensive policy- factors and economically complicated, and because induced distortions and macroeconomic imbalance is not there are substantial differences among the reforming uncommon. There are also lessons to be drawn from the countries, no single detailed road map can guide the way earlier reform experiences of Yugoslavia and Hungary. to the new systems. Therefore the paper presents general Nonetheless, the challenge is unique in its system-wide considerations that provide a framework for reform and scope, its political and historical context and required discusses the choices that have to be made in light of speed of the reform. some initial conditions in the reforming countries. In Few socialist countries remain as unreformed planned the metaphor of Vaclav Klaus (1990), the finance minis- economies. In the last decade several-for example, Chi- ter of Czechoslovakia, undertaking reform is like playing na-have instituted some economic reforms while still chess: while one needs to know the rules and have a trying to preserve the state's monopolies over political sense of strategy, it is not possible to plan each specific power, the savings-investment process and social owner- move at the beginning. ship. However, countries such as Hungary, Poland, Although the interdependence among the reforming Czechoslovakia and Bulgaria-and increasingly the Sovi- socialist countries is also very important, the problems et Union-are seeking to institute fundamental changes posed by the January 1991 reform of the trade and pay- thatwill lead to a pluralist political system, well-defined- ments institutions of the CMEA are also not addressed at and substantially private-property rights and the market- length.3 based allocation of resources. By most criteria Yugoslavia should be included here. However, in light of its long his- Initial Conditions tory of reform and political independence, and non-mem- bership in the Council for Mutual Economic Assistance Recent data on the economies of the European social- (CMEA), it regards itself as-and is-suigeneris. This pa- ist economies and the USSR are presented in table 7.1. per looks at the reform process in countries that have de- The estimates of gross national product (GNP), aggre- cided to move from a more or a less planned socialist gate or per capita, vary widely, depending on the source system to a private market economy, in which private and the conversion method, and none can be regarded as ownership predominates and most resources are allocated accurate.4 The 1988 estimates of the US Central Intelli- through the markets. Such reform demands the creation gence Agency (CIA) are well above the rates calculated of management skills and legal, regulatory and infrastruc- through the exchange rate, which are more relevant to tural conditions whose development has been set back by the borrowing status of actual and potential members of decades of socialist development. It also requires funda- the World Bank.s mental changes in the role and capabilities of the state. In any case, income levels in the European socialist This paper does not address at any length the close in- economies appear to be above the average in Latin Amer- terrelationship between economic and political reform! ica but well below that in Western Europe. Until recent- 67 Table 7-1. Basic Data on East European Economies Albania Bulgaria Czech. Ger. Hungary Poland Romania USSR Yugoslavia US Germany, Dem. Rep. FR. Population, 1987 3 9 16 17 11 38 23 283 23 244 61 (millions) Area 29 111 128 108 93 313 238 22,402 256 9,373 249 (1,000 sq km) GNP (CIA), 1988 n.a. 68 158 207 92 276 126 2,535 154 4,881 1,206 (US$ bil.) n.a. 38 51 63 44 37 2,852 33 n.a. n.a. n.a. (% of US per capita) GNP (exchange n.a. 13 18 45 13 9 16 19 11 100 n.a. rate estimates), 1988 (% of US per capita) GDP physical n.a. 30 42 45 32 27 21 n.a. 22 100 n.a. indicators, 1980 (% of US per capita) Bal. of payments, 1989 (US$ bil.) n.a. -1.1 -0.1 0.7 -0.7 -1.3 2.1 0.1 1.5 -115 60 Hard currency debt, 1989 (US$ bil.) n.a. 5.5 4.0 10.0 15.4 35.6 0.8 30.5 17.7 n.a. n.a. Life expectancy at 72 72 71 73 70 71 70 69 71 75 75 birth, late 1980s (years) Infant mortality rate, 39 15 13 9 21 24 n.a. 25 24 14 15 late 1980s (per thousand) n.a. Not available. Source: GNP (US$ bil.)-United States, Central Intelligence Agency (1989); exchange rate estimates-tentative estimates based in IMF, IntemnationalFinancialSta- tistics data where available and staff estimates in other cases; GDP physical indicators-Erlich (1987, p. 15); and World Bank, World Development Indicators and World Tables (various issues); Wharton Economic Forecasting (1989); Fink and Havelik (1989). ly, the former German Democratic Republic (East was a relative decline in their social indicators during Germany) was thought to be the most advanced of the the past decade and some evidence of recent retrogres- European socialist economies, with the CIA (United sion. States, Central Intelligence Agency, 1989) suggesting The European socialist economies experienced simi- that per capita income was over 60 percent of the United lar growth patterns after the immediate post-World War States' level. A closer look at East Germany has persuad- II period. Growth at the extensive margin, based largely ed many that Czechoslovakia has the highest per capita on the accumulation of capital through very high invest- income, followed by Hungary, but this view could ment rates and increasing participation in the labor change on closer examination of the other countries. Po- force, was quite rapid until the seventies, although there land and Hungary are heavily indebted in convertible was probably less technical progress than in Western Eu- currencies, and Yugoslavia has borrowed intensively. rope. In the late seventies growth gave way to stagna- As measured by their reported Gini coefficients of tion, worsening shortages and, in some countries, around 24 (the same as Norway's and a little higher than worsening macro-imbalances. By contrast with Western Sweden's), income is relatively equally distributed in the Europe, there was no recovery in the first half of the European socialist economies, whereas the Gini coeffi- 1980s.' Despite the broad similarities in the structures of cients for the industrial market economies commonly the European socialist economies, important differences fall in the range 30-35.6 By international standards, so- in their initial conditions will shape their reforms, in- cial indicators such as longevity and literacy are high for cluding: the income levels of these economies, although there 68 * the e-xtent of their macroeconomic imbalances, The extent and type of macro-imbalances prevailing both internal and external; before the structural reform have an important bearing * the degree of the decentralization oftheir econom- on the potential speed and design of the reform pro- ic management and extent of the product markets pre- grams. Although it can be one of the mechanisms for vailing before reform of the system begins; and eliminating a monetary overhang, the inflation that ac- * the extent of private sector activity prior to the companies liberalization that is starting from a condi- reform. tion of repressed excess demand is likely to cause serious economic and political difficulties for the reform. Al- Figure 7.1 depicts the first two initial conditions- though elimination of the monetary overhang should in macro-imbalances and management decentralization- principle be possible with a one-time change in the level in seven countries.8 of prices, rising prices may accentuate the flew disequi- librium by worsening the fiscal balance'° -thus poten- Macro-imbalances tially turning what should be a one-time change in the price level into a more persistent increase in the rate of Domestic macro-imbalance in a socialist economy inflation. High open inflation has to be eliminated for can take various forms, from open inflation to the invol- the structural reforms to be effective. Eliminating excess untary accumulation of financial claims because of the demand is also important for microeconomic reasons: rationing of acceptable goods and/or the absence of al- studies of firms in socialist countries suggest that a shift ternative assets (the monetary overhang). Poland proba- from sellers' markets to buyers' markets is essential to bly embarked on reform from the worst macro position, modifying the efficiency and customer orientation of with an overwhelming debt problem and high inflation. firms. Yugoslavia was in better shape on the external side but Countries close to macroeconomic equilibrium have also suffered from very high inflation and enormous do- greater latitude to implement those institutional re- mestic losses in the banking system that threatened forms most needed to support the operation of markets macroeconomic stability. The USSR and Bulgaria face before switching over to market mechanisms." Those major macro-imbalances. Internal disequilibrium ap- countries needing urgent stabilization face difficult de- pears to be relatively larger in the former, but the burden cisions relative to the sequencing of the stabilization of foreign debt is much lower,9 especially if account is measures (which in the past have typically involved a taken of its substantial gold reserves. Hungary has the tightening of controls) and structural reforms leading to largest external debt per capita and persistent inflation a market system. of about 20 percent but a relatively small budget deficit. With low foreign debt and probably only moderate do- Markets and Decentralization mestic imbalances, Czechoslovakia appears to be in the best macro position. The former East Germany is a spe- Turning to the other axis in figure 7.1, at the start of cial case because it has been reformed out of existence, 1990 Czechoslovakia, East Germany, the Soviet Union with the Federal Republic of Germany underwriting its and Bulgaria had relatively centralized economies, with foreign debts and any monetary overhang, a planned system for the supply of materials playing a major role and with enterprise management subject to Figure7-1. InitialConditionsforSocialistReform: constant intervention by the ministries.'2 In contrast, Degree of Economic Centralization Yugoslavia has long been the most decentralized of the Low High socialist countries, with Hungary (since 1968) and Po- Little land progressively moving the distribution of products toward a market basis-although with distorted pric- Czechoslovakia es-and according greater autonomy to firms.'3 Internal & With only limited private ownership permitted, de- EMternal Hungary centralization has resulted in self-management of the econonic social or state capital within firms. In addition to the Imbalance USSR dominant role of the self-managed sector in Yugoslavia, Yugoslavia Bulgaria some 70 percent of Hungarian and Polish firms are self- Poland managed. Severe I "Socialism without planning" has distinctive charac- teristics that have been extensively analyzed. These in- Source: The authors. clude a lack of effective balance-sheet constraints on firms 69 because of the "absence" of effective owners and the exten- ful and previously relatively independent firms account- sive fiscal redistribution among enterprises, largely moti- able, whether to the market (by reducing the cross- vated by a commitment to full employment effected subsidies and/or offering firms the opportunity to priva- through the preservation of jobs. This lack of constraints tize themselves) or directly (by asserting the state's own- adversely affects incentives and has resulted in chronic ership rights and recentralizing decision-making). scarcities, impelled by a combination of expansionist Removing acquired rights is not easy, especially since management unconstrained by considerations of profit- the shift toward representative democracy requires ability and workers seeking higher pay without fear of lay- maintaining a reasonable degree of consensus. offs and unemployment."4 Subsidies divorce incomes Together with measures to limit serious abuses from from productivity.15 Relative prices are also divorced from spontaneous privatizations," a blend of carrot and stick world patterns, with energy, transportation and state is in practice evolving in the reforming countries to deal rents normally far below world levels, credit cheap and with decentralized firms that are not yet privatized. The staple foods subsidized. Taxes and subsidies insulate the carrots, such as exemption from the wage curbs applied domestic prices of traded commodities from their world to socially-owned firms and cheap shares for employees, levels. appeal to those in potentially profitable firms. Sticks, One symptom of the distorted initial positions from such as bankruptcy and dissolution of the worker coun- which the European socialist economies are embarking cil if debts, dividends or the annual fee for the state's cap- on market reforms is that the market socialist banking ital are not paid, threaten the loss-making ones.18 system, although superficially transformed from the Assuming that opposition from its ministries has mono-bank of the centralized system, has no basis for al- been overcome, a reforming centralized economy can locating credit according to market criteria and for iden- consider and implement a wider range of options for ad- tifying and pricing risk. In this situation, the allocation dressing the ownership and management issues."9 De- of credit is indeterminate, being set neither by plan nor spite this advantage, it is unlikely that such economies by market. Avoiding severe macroeconomic imbalances face an easier transition. On the contrary, their greater in such a system is extremely difficult, essentially be- distance from a market configuration probably implies cause the economic agents have autonomy without re- more difficulty in reshaping institutions and acquiring sponsibility. the necessary skills. Decentralization may reflect a deliberate policy deci- One important question for a still centralized econo- sion to relax central planning in an attempt to gain ad- my is whether to pass through the stage of market so- vantages from market organization without confronting cialism prior to reforming to a private market economy. the ideologically difficult issue of ownership and the As this approach would involve first giving ownership technically difficult issues involved in transformation rights to the workers and then reclaiming those rights, into a private market economy. Alternatively, it may re- it is a most undesirable path for reform. Accordingly, an sult from a power vacuum following a decline in the le- economy starting out from a centralized system should gitimacy of the government or from a change in make great efforts to prevent firms' managers and work- government.'6 In many respects, a decentralized social- ers from assuming the ownership rights in an early ist economy starts the transition to a private market phase of the reform. The government can do so by defin- economy with advantages over its more planned coun- ing the ownership rights clearly and assigning them as terpart. Agents are more familiar with the markets, and rapidly as possible to agents or institutions outside the their response to market incentives is therefore likely to enterprises.20 be faster. In addition, as indicated in table 7.1, a larger share of the exports of the less centralized countries goes The Pre-Reform Scope of the Private Sector to countries outside the CMEA area and is thus subject to global competition. Firms in these countries are more Most of the European socialist economies have for likely to have had commercial dealings with foreign some time codified certain private property rights. How- firms than are firms in centrally planned systems and are ever, their private sectors are relatively small (almost more likely to have had some exposure to the real needs zero in Czechoslovakia) and comprise largely crafts and of clients. Especially in light of the depressed condition distributive trades.2' In some cases agriculture is sub- of CMEA members, the ability to increase exports rapidly stantially private (75 percent in Poland, whereas agricul- to Western markets will be an important determinant of tural production in Bulgaria is dominated by a few the speed with which the reform can produce results. public conglomerates).22 In certain other socialist coun- On the negative side, reformers of a decentralized tries the private sector is far larger than in the European economy must address the problem of rendering power- socialist economies: for example, it accounts for about 70 one-half of the economic activity in Vietnam. Because it 7.2, which classifies the needed changes, summarizes is easier to relax restrictions on an existing private sector the discussion of these changes, which follows. The issue than to attempt to create a new one from scratch, coun- of sequencing is discussed in the following section. tries with substantial private sector activity and a private asset base enter into the reform with a notable advan- Macroeconomic Stabilization and Control tage. The strategy of first empowering the private sector and then progressively reforming the public sector and Some analysts, notably Hinds (1990), have raised the relying on the private sector to absorb laid-off employees question of whether macroeconomic stabilization is pos- is more viable the larger the initial private sector is. sible in a socialist system, or whether it requires a re- form of property rights with a clear definition of Elements of Reform ownership. Following the recent experiences of Poland and Yugoslavia (as well as considering the measures The move to a full market economy requires political used to stabilize market countries with large public sec- changes that recognize the value of diversity and individ- tors), it is reasonable to assert that a tightening of the ual initiative and that require a substantial social and po- fiscal and credit policies reduces inflation and the cur- litical consensus. This consensus seems to exist in rent account deficit in socialist as well as in market Poland, Hungary and Czechoslovakia but is not present economies. in Romania or probably in the Soviet Union. Reform of However, there are some differences, which are a re- the economic system also requires fundamental changes flection of the different microeconomic incentive struc- in the role and organization of government. It means tures in the two systems. As socialist firms know no countless changes in economic structure and behavior. bottom line, stabilization policies cannot rely on the The nature of those changes can be understood by com- same responses as those in a market economy. For exam- paring the working of the European socialist economies ple, rising interest rates may encourage households to with those of the more advanced market countries. Table accumulate financial assets, but firms may simply refi- Table 7-2. Economic Elements of System Reform 1. Macroeconomic stabilization and control Implementation of stabilization programs Creation of tools and institutions for indirect macroeconomic control, monetary and fiscal Measures to harden the budget constraints Dealing with existing problems (monetary overhang, financial system, bankruptcies) la. Social safety nets (at first on an emergency basis) 2. Institutional reforms: Human capital and administrative capacity Legal and regulatory institutions Business management, including financial sector Government decision-makers and administrators Information systems (accounting and auditing) 3. Price and market reform Domestic price reform International trade liberalization Distribution systems for products Creation of market for housing Wages Interest rates 4. Small- and large-scale enterprise restructuring and privatization Management systems Allocation of property rights Agricultural land Industrial capital Housing stock Social protection and insurance rights for individuals 5. Development of financial markets and institutions Banking systems Other financial markets Source: The authors. 71 nance the growing interest charges in a giant Ponzi levers employed in modern market economies, reform of scheme that initially delays reform and then renders it the fiscal and financial sectors, with a broader perspec- extremely costly. A range of direct controls will therefore tive than the immediate macroeconomic balance, has to be needed to reinforce the indirect measures. These can be initiated. This reform is essential to increase the effi- include: the elimination of ex post subsidies for individ- ciency of the economy and to complement the enterprise ual firms and the setting of strict cash limits on them; di- restructuring that will result in firms responding to rect central bank control over the quantity of credit; and market and interest rate signals. Such responses will not controls over public sector wages. Until bankruptcy be- be efficient unless prices are themselves rational-and comes a credible threat, the proliferation of arrears on as discussed below, it may not be possible or desirable to inter-firm payments could undermine the control of impose tight budget constraints on firms without ratio- credit, especially because socialist firms tend to be close- nal prices. This point illustrates the complementarity ly linked in oligopolistic interdependencies.23 In the face among the elements of policy and structural changes of an inefficient banking system, inter-enterprise credit and the need (see below) to undertake certain reforms si- can play a useful role in facilitating adjustment to rapid- multaneously. ly changing circumstances, but the volume and distribu- Estimates of the extent of the monetary overhang tion must be monitored, just as with bank credit.24 vary; it is difficult to evaluate in the absence of most oth- The longer run success of macroeconomic tightening er stores of value. It seems to be significant in the USSR depends on the political resolve and economic skill of (estimates place it at about one-half of the financial as- the government, which is bound to come under severe sets held by households) and may exist in some of the pressure when firms run into financial trouble. A skilled smaller countries. government will already be working out criteria and There are several broad options for eliminating the methods for the restructuring of enterprises. It will also overhang without drawing on foreign exchange reserves have to institute measures-the social safety net-to or foreign borrowing: a currency reform that freezes or protect the living standards of those adversely affected by eliminates financial balances above certain levels; reli- the economic reforms. In particular, because job losses ance on inflation, with incomplete compensation by in- are likely to prove the most traumatic consequence of terest rates; an increase in interest rates on existing reorganization in economies that have not previously assets or introduction of new, higher return assets; and known open unemployment, it is essential that a social the sale of state assets to the public. safety net for the unemployed be included in the plan- A confiscatory fiscal reform is problematic, especially ning of any stabilization cum reform program. In the be- in the context of a reform aiming to secure private prop- ginning, measures will have to be instituted on an erty rights, but can possibly be justified politically as the emergency basis because the European socialist econo- result of the mistakes of the previous governments and mies have no administrative structures for dealing with the dubious nature of the activities that allowed individ- open unemployment. Over the longer run, a more com- uals to accumulate private wealth. Poland and Yugosla- plete social safety net will have to be developed.25 via in effect took the inflationary route at the end of In addition to measures to discipline firms and address 1989. However, it is inadvisable that it be taken after the unemployment, fiscal reforms must have a high priority. adjustment has started, as it imposes high costs of its With tax revenues highly dependent on remittances of own. Raising interest rates on existing assets, such as profits from enterprises, measures to increase their finan- saving accounts, will help reduce the inflationary pres- cial autonomy and at the same time absorb the losses that sure from the current overhang (in the case of the USSR, result from restructuring are likely to have a serious im- most estimates suggest that raising interest rates to nor- pact on the budget. Revenues from privatization are likely mal levels would not be sufficient to eliminate the over- to become significant only in the medium term, even if hang). privatization itself is accelerated; increased revenues from There are strong arguments for the sale of public as- privatization and greater efficiency are likely to lead to a sets, especially since this approach can be made consis- lag in the increased costs associated with the restructur- tent with the goal of strengthening the small-scale ing. Because of the uncertain size of any particular tax private sector. The sale of small enterprises and premises base in the turbulent reform environment, there are argu- for businesses and retail shops is one example. Sale of ments for raising revenues as broadly as possible, subject the housing stock is another, although this action would to the problems posed by inadequate accounting, auditing need accompanying measures to raise the cost and re- and enforcement. duce the security of renting, and these, in turn, would Since governments will in the longer term have to require the adjustment of pay levels to reflect the higher maintain macroeconomic control through the indirect rents. Because rents are usually far below maintenance 72 costs, even giving housing away would improve the bud- Institutional and Complementary Reforms get on a flow basis. At the same time, it has to be recog- nized that privatization is likely to be slow. One possible Institutions and professions that are taken for grant- strategy-it has not yet been used in the European so- ed in market economies have to be re-created in the re- cialist economies-is to combine currency reform with forming socialist economies. Legal, regulatory and the sale of assets. Asset holdings above a certain level can information systems need to be reformed to support the be frozen and their subsequent use permitted to pur- markets. chase assets that are being privatized. A secure legal environment has to be created for the Consider now the external balance. Especially once protection of property rights and the regulation of com- measures are taken to harden the budget constraints and mercial relations. Accounting and audit systems are restrain domestic demand, changes in the exchange rate needed to organize and monitor information. Needed to should be effective in switching expenditures and output complement those reforms are investments in human in the socialist economies, provided that a price reform capital in areas such as accounting, credit and market creates a link between the world and domestic prices for analysis, and bank inspection. Management skills have tradables. However, again the distinctive institutional to be upgraded and modernized, especially in finance arrangements of the European socialist economies con- and marketing. In some areas such as financial markets, strain the likely response. Over half the exports of the reform may require a greater role for the state than be- European socialist economies go to countries in the fore. CMEA system. Many of these products are not of ade- When assessing the efforts underway to change the quate quality and appeal to be acceptable in the world legal and regulatory institutions in reforming socialist markets. Within the unreformed CMEA system, trade countries rapidly, it must be remembered that those in flows were arranged through inter-government proto- market economies have evolved over centuries. Such re- cols rather than being responsi vernme prices. forms and the creation of the human capital appropriate cols rather than being responsive torelativeprices,to a new system are lengthy processes that constrain the Except for the USSR (which exports hard currency 28 commodities), the imminent comprehensive reform of efficiency and speed of reform. They also constitute an the CMEA system toward multilateral, market-based area in which foreign assistance may be especially useful trading (which is a logical complement to the internal in the transition. reforms) raises the problem that each reforming coun- Prce and Market Reform try's exports must compete in previously sheltered mar- kets with superior products from the West. Reforms in Given the mutual interdependence of prices and their one socialist country as well as in the trading system can sheer number (some 25 million in the planned Soviet render more difficult the short-run task of maintaining economy), the process of moving to a more rational set an external balance in the other reforming countries. In- of prices could be highly complex and extended. Fortu- deed, exports from certain small European socialist nately, the rational price system needed for increased economies to the West could actually fall because of the enterprise autonomy to make sense is known: it is world need to divert world-quality products to the USSR to pay prices. The best way to introduce rational prices is to for essential imports of fuels and raw materials. In the open the economy to foreign trade. An open trade (and longer run, however, reform of the CMEA system will current payments) policy also exposes domestic produc- have a positive effect on the economies of its members- ers to foreign competition and offers the prospect of and certain countries, notably Hungary, have already more rapid increases in the quality of products through demonstrated their ability to increase exports to the a range of licensing, processing, marketing and other West quite rapidly. joint-venture arrangements with foreign firms. This In sum, stabilization measures involving a combina- route is certainly available to the European socialist tion of controls on aggregates and specific interventions economies, which will become more open as they reform can be effective. However, given their economic rigidi- and which can expect considerable growth in trade (es- ties, narrow markets and small convertible currency pecially intra-industry trade) with the market econo- trade ratios, the response of socialist systems to reduc- mies.29 tions in expenditures and switching measures is likely to The Polish government's move to current account be less rapid and more costly in terms of output than is convertibility (at a heavily depreciated exchange rate) typical in market economies. Further, the cost of reform should thus be seen not only as a way to improve the cur- is likely to be raised by the simultaneous reforms in the rent account, but more importantly as an essential and other European socialist economies as a result of the dis- immediate move to a rational price system that helps ruption and loss of protected markets. force competition. In practice, some price-distorting tar- 73 iffs will remain-uniform tariffs can be justified as a The ownership issue is a political minefield: workers source of fiscal revenue-and realistically the govern- in worker-managed firms believe they have substantial ments of the reforming European socialist economies claims on the firms; while insider sales of firms by and to are no more likely to move immediately to undistorted former managers and bureaucrats (spontaneous privati- trade than are other governments. Nonetheless, world zations) create inequities. How to respond to such sales, relative prices are the right reference point. which are likely to be the first response to liberalized Trade and domestic reforms are mutually dependent. regulations, is an important policy question for govern- Domestic deregulation of agriculture, industry and ser- ments. Political opposition to the initial bursts of spon- vices and reform of enterprises have to take place if firms taneous privatization in Hungary and Poland resulted in are to respond appropriately to price signals. Some ana- their suspension and a lull in privatization activity in lysts (Nuti 1990) seem to argue for liberalizing the inter- those countries. Sales of firms before their value can be nal markets and reforming administered prices before determined through a realistic price system are likely to liberalizing international trade. Such reforms are more be inequitable ex post, and nationalist sentiment in all likely to succeed if an appropriate price system is already countries will put some limits on the role of foreign in place, and liberalized trade can be an important ele- ownership.30 The ownership of agricultural land poses ment in producing such a price system. another set of politically charged issues. Price liberalization is also necessary for the credible Corporatization, which involves a change in the sta- long-term hardening of the budget constraints. Until tus of a firm into a joint-stock or other corporate form, prices are rational, profits and losses are not necessarily is normally the first stage of enterprise reform. However, good indicators of efficiency and are therefore not good restructuring and privatization are the two real phases. guides for decisions on which firms should be closed and Opinions on which should come first differ. Proponents of slow privatization argue that firms should be sold off which should obtain financing for increased investment, gradually after restructuring, when there is a more ra- Closing a firm whose output price is being held below its tional price system, the new rules of economic behavior social value is not necessarily the optimal response to its have begun to emerge and a real business class has had financial losses. Firms cannot easily be held accountable time to develop to exercise the ownership function. The for their profitability if the government sets their prices. prices of firms will be more realistic and the sales pro- Because of past patterns of industrial organization, cess more equitable, both because the new owners will additional measures may be required to develop compet- pay the right prices and because the government will ob- itive markets. These are likely to include active and early tain more revenue. The privatization in Great Britain is demonopolization of the transport, distribution and cited as evidence supporting this approach. Proponents trade systems. They may involve the break-up of large also point to the need for "real owners"-private agents conglomerates that evolved not in response to market with a long-term view and sufficient equity in a given imperatives, but to reduce the problems of control and firm to encourage them to play an active role and to give to secure inputs often in shortage under the socialist them a strong measure of control. They also suggest that system. Although demonopolization is needed, it should premature privatization may unnecessarily create inter- be approached from an international perspective, espe- est groups of shareholders who object to the removal of cially in a small country. As a method of enhancing com- distortions, such as subsidies, which has an adverse im- petition, the opening up of trade is likely to be preferable pact on the value of their assets. to fragmenting firms to achieve competition in the do- Supporters of fast privatization argue that more com- mestic market. prehensive and rapid ownership reform is necessary to So far, this paper has considered only product prices increase efficiency.31 A further argument is political. and markets. However, the creation of factor markets Without the power of a substantial capitalist class behind and the liberalization of factor prices are also important, it-and it is agreed that such a class will not emerge rap- and they are discussed later. idly-the reform will surely be subverted and prema- turely terminated by the interest groups representing Enterprise Reform adversely affected segments of society, such as potential- ly laid-off workers and redundant bureaucrats. Reform Enterprise reform-in all sectors, not just industry- must be radical and fast-moving to deny such groups the is the heart of the transformation. As in all areas, this opportunity to coalesce. It is widely accepted that more process requires interrelated changes: the imposition of rapid privatization reduces the expected total sales reve- bottom line discipline; definition and change of owner- nue to the state; fast privatizers argue, however, that the ship; and reform of management arrangements. benefits of a rapid and irreversible shift to private pro- 74 duction outweigh the costs of reduced state revenue and fering some seller financing. The state would have to that the higher tax revenues from a more efficient econ- handle the large weak firms, with extensive restructur- omy will offset the loss in revenue from rapid privatiza- ing needed to move them into the private sector if they tion. are not closed.35 A number of schemes have been proposed to achieve The larger firms with potential are the crux of the the goal of rapid privatization in an equitable and politi- privatization problem. Many could be privatized relative- cally acceptable fashion, and in doing so to overcome the ly rapidly, at least in part. Foreign investors are already problem posed by the uncertainty of asset valuation. The active in each country. In some cases they could either schemes typically involve creating shares in the compa- take on the role of dominant shareholder (possibly on a nies and distributing them broadly within the country. temporary basis) or could manage the firm under con- Since most proposals recognize that very dispersed own- tract to a domestic holding company. Shares in such ership does not provide a basis for effective control, they firms could be distributed under the schemes discussed also include a mechanism to create a dominant share- above. holder or group of shareholders to monitor the manage- Whatever the precise rapid method used, it is likely ment of the individual companies. Some offer employees that state holding companies will play an important preferential access to a block of shares as an efficiency ownership role in the transition. Given the scarcity of incentive and to render the process acceptable to them. domestic owners and the undesirability of a fire sale to Two approaches can be considered: foreigners, there seems to be no alternative. Some form of holding company structure would likely also be need- (1) create holding companies or unit trusts to be the ed under the slow privatization option, as a vehicle dominant shareholders, sufficient in number to ensure through which the state could assert its ownership dur- competition among them. The state can retain a block of ing the restructuring. In the longer run, the holding shares for later sale or distribution to citizens, a measure companies could be privatized; alternatively they could that would ensure that the public participates in the be converted into pension funds by transferring to them benefits of privatization. The government can also offer the appropriate liabilities. Foreign management may be citizens shares in the holding companies, so that they a useful intermediate stage; so might foreign ownership, have an opportunity for diversified portfolios. The diffi- although there are political limits to this option, and it culty is to ensure that the dominant shareholders oper- is inappropriate to sell national assets off too cheaply on ate like private owners, which they are not (unless they the basis of very short-term considerations. consist of foreign owners who have sufficient wealth to In line with European tradition, and considering the buy the shares). There is also a risk that privatization regulatory difficulties, natural monopolies should prob- may stall at some intermediate stage. ably not be privatized early on, unless perhaps this ac- (2) distribute the shares in each enterprise equally tion is particularly needed to attract large foreign across the population and initiate trading arrangements investments to facilitate restructuring and moderniza- to enable dominant ownership groups to emerge. tion. Whether an effective ownership group would actually re- sult and whether the large ex post inequities that are Financial Markets and Institutions likely to arise are acceptable because of ex ante fairness are important questions.32 Development of financial markets and private sector financial institutions, including banks, is an essential No country has yet implemented schemes of these step in the transition to a market economy and in mov- types, but they are being actively discussed, and imple- ing investment decisions away from government con- menting legislation has been passed in Czechoslovakia trol. Many reforming socialist countries have already and Poland.33 taken steps in this direction, including moving from a The reforming European socialist economies are like- mono-bank to a two-tier banking system. Several of the ly to adopt a variety of approaches in practice. Consider European socialist economies are also discussing the first firms in potentially competitive sectors, and sup- rapid creation of a stock market. However, there are im- pose that price reform and macrostabilization have pro- portant differences between enterprise reform and the gressed sufficiently so that reasonable judgments about reform of financial institutions that suggest a need to viability are possible.34 Small weak firms should be treat these latter reforms differently. closed and their assets and premises sold off; small More so than with other sectors, financial markets de- stronger firms could be privatized rapidly, possibly being pend on underlying legal and informational systems and sold to all or some of their employees, with the state of- skills that barely exist in the European socialist econo- 75 mies. Further, the problems with existing loan portfoli- changes are preconditions for others: for instance, mac- os, which are very substantial, have to be addressed roeconomic stabilization is needed if price reform is to before a sound banking system can emerge. The loan be successful. The systems and skills that have to be in losses, which originated when the central bank's portfo- place for the markets to work need to be developed. To lio was allocated to the commercial banks, will have to be give an important example, financial liberalization is ex- moved off the balance sheets and allocated either to de- tremely risky unless a sound system of accounting, au- positors or other lenders or be absorbed by the budget. diting, prudential regulation and supervision is in place In addition, the large loan losses that will appear when and unless the macroeconomy is reasonably stable. relative prices and demand patterns shift as a result of The sequencing problem has been extensively studied the domestic and trade reform will need to be handled in the analysis of the structural adjustment of develop- similarly. The banks should be recapitalized and expect- ing economies. While some rules of thumb have ed to function as market institutions only after the econ- emerged,3' the more important conclusion is that a lin- omy has settled down sufficiently to enable loan ear sequence of individual policy changes is not the right portfolios to be kept reasonably clean. Before then, the approach. The problem is better thought of as one of proliferation of banking institutions should be con- how to package the reforms: the need is to introduce strained and the emphasis on competition in financial groups of complementary policy reforms sequentially. markets moderated, although some entry of foreign The details of the reform path any country will follow banks, as participants or managers of domestic institu- depend on the state of the economy, on the tolerance of tions, can play an important role. the population for the disruptions that are sure to ac- The differences between banking reform and enterprise company the reform and on the political situation. reform exist because the condition of the firms deter- Nonetheless, figure 7.2 presents an outline of a proto- mines the state of the banks' portfolios, their earning po- type reform for a representative European socialist econ- tential and their equity (assets corrected for expected omy with initial conditions somewhere between those of loan losses less liabilities). If their equity is zero or neg- Poland and Czechoslovakia. ative, as is the case in several and possibly all European For countries with high inflation and non-sustainable socialist economies, banks face perverse incentives in balance-of-payments deficits, macroeconomic stabiliza- mobilizing and allocating credit that lead to viable bor- tion has to be the initial priority. The severity of the pro- rowers being crowded out. If the banks' budget con- gram and the range of the accompanying structural straint is hardened without restructuring their reform measures will reflect the extent of the initial im- portfolios, they will attempt to compensate for the high balances and political judgments about what the popula- share of non-performing loans by raising charges to pay- tion will accept. A new government, or a government ing clients.36 The latter will face extremely high costs of with support for radical change, has far more leeway, credit because of large banking spreads. Under such con- which it should use, as it is harder to impose tough mea- ditions it is likely that mainly firms in potential financial sures later. trouble will borrow from the banks. In the extreme case of Poland, the government pur- A truly competitive banking system cannot be expect- sued a stabilization program consistent with a move to- ed to emerge before the enterprise restructuring and ward a market system. The program included sharp cuts price reform are substantially under way. This conclu- in firm-specific subsidies and tight credit limits, as well sion raises several sequencing issues that are addressed as trade liberalization on a major scale and at a heavily below, including the question of how new firms are to be depreciated exchange rate. In high inflation countries, it financed early on in the transition. Other important may be necessary in the initial phase of stabilization to questions to be considered are the ownership of the fix the nominal exchange rate to provide a nominal an- banks (they should not be owned by client firms) and chor for the price level. It is also necessary to create an whether the banks should have an ownership role as in emergency safety net that shifts social protection off the Germany and Japan. That last point is less important at shoulders of firms so as to facilitate the allocation of re- the start of reform, because the banks have no obvious sources, which allocation means, in the case of labor, capability to exert a positive ownership role. unemployment. While initial success in reducing inflation from very The Sequencing of Reforms high levels can be achieved within a few months, stabili- zation can be assured only by following consistent poli- Administrative feasibility alone ensures that not all cies over periods of years. During that time, measures to reforms can be instituted simultaneously. The more fun- allow indirect macroeconomic control should be put in damental reason for sequencing the reforms is that some place. It may take years for the expectation of macroeco- 76 Figure 7-2. Sequencing of the Reform Intense ~ Cntr~in Macrostabilization _ I __um Emer ne v Instis tionalizal ion Sodal safety net _ ] Institutional reforms Price and market Adjust tariffs t uniformi reform Small-scale privatization and private sector development Large-scale restructuring Evalu ition and privatization Autonomous banking Prep ration an n ime n system Small scale lend g Other finandal markets 0 1 2 3 4 5 6 7 8 9 10 Years Source: The authors. nomic stability to become sufficiently ingrained to affect creasingly be left to the market as firms are restructured long-term economic decision-making. A particularly dif- and privatized. ficult stage for reforming governments comes a few Realistic interest rates are needed to attract deposi- months after success with the initial phase, when unem- tors into the banking system and to provide a reasonable ployment is rising, firms are beginning to experience fi- measure of the cost of financial resources to firms. How- nancial difficulties and closings are in sight. Because ever, it will not be possible to leave the determination of inflation is lower, the demand for reflation-meaning an interest rates to the market until far into the reform be- easing of credit constraints and reinstatement of some cause appropriate incentive structures for firms and subsidies-mounts. For this reason alone, governments banks will not be in place. Premature reliance on the fi- have to move quickly after stabilization to formulate nancial market as an instrument of indirect economic measures for the emergency restructuring of firms. management is likely to raise real interest rates to the Price reform for goods is shown taking place early as a point where they create financial difficulty even for firms prerequisite for decentralization and tightening of the that would be sound in normal times, as well as to pro- budget constraints, with trade reform an important ele- voke costly financial crises. ment. Progressive reductions in state interventions con- Measures to eliminate discriminatory regulation tinue the process. against the private sector can be taken at the start of the Wage determination presents a major challenge. In process.39 Privatization of small firms can also start im- socially-owned firms (the great majority), wage-setting mediately, with small-scale commercial bank lending cannot be left to the market early in the process because needed in support. Whether or not some holding compa- the firms are not operating under correct price and man- ny mechanism is used as an intermediate stage or some agement signals. At the same time, wages must respond shares are distributed across the population, preparation to inflation and changing relative prices, including those for the restructuring and "real"privatization of most (such as the introduction of more realistic rents) needed larger firms is likely to take several years and execution to place the housing sector on a sound basis. Wage to take much longer. Nonetheless, preparation, includ- guidelines or formulae have to be used initially in non- ing corporatization, has to begin as soon as possible. privately owned firms;35 wage determination can in- Clarification of ownership rights and the transfer of re- 77 sponsibility for the direction of firms to boards of direc- current accounts. Also ignored are the complications tors should be implemented as rapidly as possible. that will arise from the changing CMEA regime. In addi- Reform of the banking system involves several stages. tion to the issue of product quality discussed above, Preparation can begin immediately with the establish- CMEA reform is likely to exacerbate the reserve problem ment of suitable accounting and asset valuation stan- by raising the share of trade for which a normal reserve dards, reform of the banking, contract, enterprise and cover is needed. bankruptcy laws and the drafting of prudential regula- Figure 7.2 shows how much has to be done at the be- tions, as well as staff training. Next come audits of firms ginning. Virtually all the reforms, or at least their plan- and banks and asset valuations, followed by portfolio re- ning, have to start quickly, even though implementation structuring (the counterpart to enterprise restructur- may take a decade. Of course, figure 7.2 exaggerates the ing), allocation of losses and recapitalization. Only after precision of the process, and it certainly does not show this process is complete can a market-based banking sys- the setbacks that are inevitable in any economic reform tem emerge.40 program. From a purely technical perspective, equity markets Figure 7.2 also cannot show one other vital element. can probably be created quite near the start of the re- A government that starts with a clear idea of what it form. However, apart from their possible initial role in wants to achieve and with a popular mandate to move in consolidating claims as noted above, they will be only a that direction, and that resolutely pursues its goals, will facade because valuation and information deficiencies transform the economy more successfully than will a will prevent them from playing a useful role in the allo- government and society that is not sure what it wants. cation of resources.4' Evolving such a role will take For this reason, reform is more likely to succeed in those years, and they should not be developed on any signifi- countries where there is a consensus on the necessity of cant scale until the banking system has been restruc- moving to a normal private market economy. tured. It is important that the reform of the financial sector The Role of Foreigners not outrun the reform of enterprises. This caveat raises the important question of how the emerging private sec- The success of economic reform in European socialist tor is to obtain financial services in the interim. One pos- economies depends heavily on the extent to which the sibility is to establish specialized financial institutions reforming governments open their economies to foreign for the private sector. However, a better solution, given trade, capital and expertise and on the response of West- the need for the banks to acquire the capabilities needed ern economies and governments. Opening up to foreign to serve private clients, is to implement a two-track trade immediately puts in place the right price signals strategy for the banking system, with service depart- and allows countries to benefit from larger markets and ments for the private sector in all major banks, albeit on from foreign technologies. The proximity of Western Eu- a small scale initially, and the separation of assets in pri- rope makes these benefits obvious to the reforming gov- vate and publicly owned entities. ernments and to those contemplating reforms. No separate line is shown in figure 7.2 for interna- The reforming governments have welcomed foreign tional trade. The reason is that reform of trade in goods direct investment, although it is not yet taking place on is treated as an element of price reform. Current account a large scale except in the former East Germany. It will convertibility, which means essentially unrestricted ac- be necessary for the governments of the European so- cess to foreign exchange for current account transac- cialist economies to develop a systematic approach to tions, is desirable early on in the reform program.42 foreign investment and foreign ownership, recognizing It is particularly important to make provisions early that the right approach is not to create special incentives in the process for the treatment of foreign investors. for foreign investors but rather to create conditions fa- Capital account convertibility should come later, when vorable to both domestic and foreign private investment. expectations of stability have been established. Official foreign finance is also needed to ensure re- This simple treatment of the trade and payments is- serves, cushion the short-run impact of stabilization and sues ignores the need for reserves to be adequate prior to help develop the infrastructure. Bilateral support except liberalizing the current account. Poland's reserves had in the case of the former East Germany has been limited been augmented by standby agreements, and Poland en- so far. However, the multilaterals have started lending joyed a de facto moratorium on its debt service; Yugosla- on a large scale. The potential role of the new European via had accumulated reserves prior to initiating its major Bank for Reconstruction and Dvelopment (EBRD) as a reforms. Countries in the position of Bulgaria, with es- lender to the private sector is important, especially given sentially no reserves, face greater risks in opening their the reluctance of commercial banks to become involved, 78 but it will take some time to work out how public money vate market economy. The organization of such econo- should be channelled to the private sector. mies is quite well-defined in general terms, but there are Foreign expertise from both the private and official important differences across industrial countries, for ex- sectors is avidly solicited by European socialist econo- ample, in terms of the role of the state, the coverage of mies. It is essential to expand the use of management social insurance and the methods of collective bargain- contracts and other methods of using experts with the ing. Economic theory offers relatively little guidance for practical knowledge of how to run businesses and the making decisions on the distribution of wealth across other institutions that support a market economy. the population. There are also important choices to be What should Western governments do to help the re- made concerning the role of foreign ownership and con- form? They cannot intervene directly in the intense po- trol. Such differences obviously will affect the design of litical ferment that in the end will determine many specific programs. aspects of the reform. However, they can make four im- Even with consensus on the elements of reform, im- portant contributions: plementation can generate substantial debate. For ex- ample, the approach to regulating natural monopolies * Market access. With generally restrictive domestic differs between the United States and Europe; similarly, spending policies and the need to upgrade to Western the tax codes and banking regulation and supervision are standards, access to the markets of industrial countries not uniform. In some of these areas, flaws are apparent is vital for a rapid return to growth. It is extremely im- in certain potential models that have evolved over portant that the industrialized countries open their mar- time.43 More often, the most important consideration in kets to exports from the European socialist economies. shaping reform may be to ensure consistency with po- * Technical assistance. The needs of the reforming tential trading partners and sources of capital, so that countries are enormous. Because of their relatively high differential regulation does not impede economic inte- levels of literacy and numeracy, the pay-off from a period gration. of intensive and well-coordinated technical assistance The important strategic choices arise, however, out will be high. Direct firm-level assistance can be especial- of the inter-play between economics and politics. Sys- ly useful. tem reform is an intensely political process, and the im- - Financial assistance. Some countries, especially peratives of political sustainability may dominate Poland, are embarking on reform with crushing debt analysis of the reform: indeed, the major differences be- burdens. Relief is an essential ingredient of reform, as is tween the alternative proposals for reform reflect, to a continued access to capital to finance the restructuring large extent, different strategic views of what will be sus- and provide a cushion during the transformation. Assis- tainable. The time needed to reform institutions, create tance aimed directly at cushioning the impact of restruc- skills-including the re-instatement of a business turing can also play an important role. class-and establish valuations argues for a measured * Coordination. Present aid efforts are so ill-coordi- pace of reform. However, a slow pace has costs, includ- nated that they stretch the already overstretched imple- ing prolonged uncertainty and perhaps an extended pe- mentation capacities of the reforming governments. riod of poor management of assets. It runs the risk that Coordination has to be improved. opposition to the changes will coalesce to terminate the process. This risk argues in favor of a rapid approach, but Conclusion in that case the markets are apt to be liberalized prior to adequate preparatory steps. Owners will be created early, This paper presented a definition of the main compo- either in the form of various holding arrangements or nents of programs to reform the socialist economic sys- spontaneously by trading claims, distributed in some tems with the aim of transforming socialist economies across-the-board manner. This rapid approach to reform into private market economies. Initial conditions that has definite merits, especially if reserves have been built shape reform were discussed, together with the comple- up to tide the country over the period of economic dis- mentarities between the major elements of reform and ruption that is likely to follow. However, it, too, runs the need for certain reforms to precede others if the pro- high risks because of the potential for chaos. gram is to be consistent. The paper then sketched out an Especially in this situation, technocratic solutions illustrative schedule for such a program for a represen- are not sufficient. In addition to trying to define optimal, tative country. or even feasible, transition strategies, economists must In practice, any given country will face many choices have an eye for the major inconsistencies, risks and concerning its reform path. Some choices involve the dead-ends likely to result from strategies driven by polit- target ofthe reform, that is, the ultimate form of the pri- ical processes. They must seek ways to defuse such prob- 79 lems, if possible before they become serious enough to fect, in which real tax revenues decline as inflation rises, but also be- undermine the reform. At the same time, politicians cause partial price decontrol may require increasing government have to recognize that the neglect of economic factors subsidies to maintain the prices of necessities constant. 11. For examples of authors who advocate such a pattern, where has caused the present crises and that political consider- the institutions needed to support the markets are put in place prior to ations alone cannot drive the reform. each stage of liberalization, see Svejnar (1990) and the so-called se- quential approach outlined in Rosefielde (1990). Notes 12. For descriptions of the Soviet and Chinese patterns of man- agement, see Hewitt (1988) and Tidrick and Chen (1989). 13. China is a distinctive case. It adopted a two-tier approach to 1. Work on this paper was begun when Fischer was chief econo- decentralization and markets, with state firms and a material supply mist at the World Bank. The opinions expressed are those of the au- system coexisting with a growing non-state sector (largely in agricul- thors and are not the official views of the World Bank. The authors are ture and rural industry) and with trading increasingly occurring on a indebted to Bela Balassa, Olivier Blanchard, Rudiger Dornbusch and market basis but with different prices for similar commodities. In its Imre Tarafas for their helpful comments. reform, China has also experimented with regional differentiation, 2. Political reform may be needed prior to economic reform to with some provinces and special zones moving far more rapidly toward prevent the latter from being stalled by an entrenched bureaucracy. markets and pluralistic forms of ownership. (See Byrd and Lin 1990.) However, as a reflection of the complex political-economic interac- Regional decentralization has progressed a long way. (Many of the pre- tions, successful, partial economic reform can provide space for polit- rogatives that firms have gained in the European socialist economies ical reform. Providing alternative private career paths for the have been captured by the lower levels of government in China.) nomenklatura, or bureaucracy (a contentious issue in several coun- 14. Authors differ in the emphasis they place on the relative im- tries) could also ease certain aspects of the political reform. portance of the absence of a well-defined bottom line for management 3. The CMEA is an essentially bilateral system of trade relations, and the responsibility of management to workers in accounting for the with the Soviet Union the dominant trading partner with each of the performance of decentralized firms in European socialist economies. smaller Central and Eastern European countries. Yugoslavia is not a (For examples, see Kornai 1990 and Hinds 1990). member of the CMEA. Payments among CMEA countries are expected 15. China's rural economy can be said to consist of product mar- to be made in hard currencies, at world prices, beginning in 1991. kets without factor markets, as individuals cannot move easily and in- 4. The estimates shown in the row "GDP physical indicators" are terregional resource flows are constrained. However, communities close to the estimates of the purchasing power of real incomes. See face quite hard budget constraints, and fiscal mechanisms to equalize Fink and Havelik (1989) for a discussion of alternative income esti- incomes across them are limited, so that, on a group basis at least, in- mates. comes reflect the availability and productivity of local factors. 5. The exchange rate-based estimates-which use actual ex- 16. Czechoslovakia made moves in this direction, with workers in change rates for countries such as Poland and Yugoslavia that have re- some firms effectively dismissing their managers through votes of no formed their exchange arrangements and staff estimates for the other confidence, but it has also taken steps to reassert centralized control as countries-imply that all European socialist economies would be eli- a prelminary step in the move to a private market economy. gible for World Bank borrowing based on the income criterion. 17. In spontaneous privatization, which is initiated by manage- 6. Inter-regional income differences account for much of the ob- ment, the firm is sold or its assets are leased to a private corporation in served inequality in socialist countries, essentially because of limited which the managers and possibly workers have an interest. Spontane- equilibrating resource flows in the factor markets. For this reason Yu- ous privatization emerged in Hungary and Poland, and somewhat later goslavia's Gini coefficient is relatively high. For a discussion of this ef- in Yugoslavia; it is also taking place in the Soviet Union. fect in rural China, see Byrd and Gelb (1990). 18. Poland and Yugoslavia have adopted variants of the carrot and 7. As with the estimates of real income, the growth rates of the so- stick approach. In the former, failure to pay a dividend to the govern- cialist economies are subject to debate. In some cases, including East ment triggers intervention; in the latter, late payment on debts is the Germany, the high reported growth rates are known to have been the trigger. result of statistical manipulation. Nevertheless, the general picture in 19. Czechosfovakia is a case in point, with the famous voucher the European socialist countries is one of quite high growth rates that scheme being only one of many proposals under consideration (see declined sharply from the latter half of the seventies, followed by stag- Klaus 1990). nation during a period of attempts at reform. This situation is very dif- 20. This point does not argue against some employee sharehold- ferent from China's, whose economy stagnated prior to the reforms ing or representation on boards, as in co-determination. However, initiated in the 1970s and then grew rapidly thereafter. there should be an independent locus of control on the board. 8. The relative positions of the countries in figure 7.1 necessarily 21. For example, limited forms of private property were legally reflect a degree of judgment. Some countries included in table 7.1 are protected beginning in 1955 in Hungary, and their scope has been pro- omitted from figure 7-1 for lack of information. gressively enlarged, notably in 1968 and 1982. Private property in 9. In March 1990 the USSR published a list of its debtors and iden- China and the USSR, on the other hand, has lacked a formal legal un- tified overall international assets of US$145 billion, of which $74 bil- derpinning. lion were owed by CMEA members and $71 billion by developing 22. For reviews of the private sector in four European socialist countries (OxfordAnalytica 1990). It is not clear to what extent these economies, see Webster (1990, Annex IA-ID). debts are collectible, but at least on paper the USSR has developed a 23. Central planning is facilitated by minimizing the number of substantial net international credit position that could give it some firms to reduce the complexity of control. Spontaneous entry and exit leeway in adjusting its external accounts. Bulgaria has also extended processes are virtually absent under socialism. credit to developing countries, notably in the Middle East. Reform of 24. Yugoslavia has instituted a centralized reporting system cov- the CMEA system will probably be to the advantage of the USSR but ad- ering both bank and inter-enterprise credit. Late payments on either verse for Bulgaria. can trigger intervention and bankruptcy. 10. This situation could arise as a result not only of the Tanzi ef- Both the phenomena discussed in this paragraph have their coun- 80 terparts in market economies: firms facing a high risk of bankruptcy 37. For example, when inflation is rapid, macroeconomic stabili- may attempt to borrow at very high interest rates-a tendency that zation is needed before reforms to change relative prices, such as trade leads lenders to ration credit at market interest rates; and inter-firm reforms, are undertaken; the goods markets should not be liberalized credit, sometimes called-the gray market, tends to expand when bank later than the factor markets; and the capital account should generally credit is constrained. be opened after the trade account. 25. The far lower levels of productivity in the European socialist 38. In the case of Poland, wages were partially indexed. economies should be borne in mind when assessing the applicability of 39. Hungary, Poland, Czechoslovakia and Yugoslavia have all tak- Western European social insurance systems, with their relatively high en steps to place private firms onto the same legal and regulatory foot- levels of benefits. Because of the initially quite equal income distribu- ing as public firms. tion in the European socialist economies, it may be difficult for them 40. This is not to say the banks cannot be active participants in the to formulate comprehensive social insurance schemes and at the same restructurings-centers of expertise can be usefully developed in the time widen the differentials to create adequate incentives. banks. At the same time, responsibility for the process and its financing 26. See Dornbusch and Wolf (1990) on the effectiveness of curren- must lie outside the banking system. cy reforms in dealing with the post-World War II monetary overhang 41. In China, for example, equity issues can be found despite the in several countries. absence of laws defining private property rights. Liabilities combining 27. One suggestion is that a reforming economy take over and features of debt and preferred shares are also issued, but without clear adapt a complete legal and regulatory system from a market economy definition of the holders' rights and the basis for their returns. with which it has historic and cultural ties. For example, Hungary and 42. Some countries may elect to use dual exchange rates and im- Czechoslovakia might adopt and adapt the Austrian framework. The port licensing for some transitional period. In these cases, it would be former East Germany has essentially adopted the institutions of the important for firms to have access to foreign exchange and automatic Federal Republic. import licenses in the free foreign exchange market. 28. Consider, for example, that in the United States it takes a min- 43. For example, few would now recommend adopting the organi- imum of five years to train an examiner to deal with the smallest and zation of the US banking industry or the institutional fragmentation of simplest bank. its regulators. 29. While open foreign trade provides a rational price system for traded goods, it does not do so for non-traded goods. Moreover, until References the capital account of the balance of payments is opened, interest rates and the prices of assets can also diverge from world levels. In a large, Blue Ribbon Commission: Project Hungary. 1990. Hungary in Tans- closed economy such as the USSR, it is less clear that a rapid move to BlueaRion C omms n: Proset y. 190.Hna ry In Hun world prices is the most appropriate strategy. However, even in the fomtiontto Freri case of the Soviet Union, rapid liberalization of trade, combined with a Institute, April. gradual reduction in tariffs, would provide potential markets for ex- porters and an important element of potential import competition for Byrd, William, and Alan Gelb. 1990. "Why Industrialize? The Incentives domestic monopolists, conditions that would help create an appropri- for Local Community Governments."In William Byrd and Lin ate relative price structure. Qingsong, eds., China's Rural Industry. New York: Oxford Univer- 30. hina's rural reform, in which communes were in effect priva- sity Press. tized to their members, is the most impressive example of the reform of property rights among the socialist countries. Lin (1989) has shown Byrd, William, and Lin Qingsong, eds. 1990. China's Rural Industry. that the introduction of the household responsibility system was re- New York: Oxford University Press. sponsible for most of the increase in agricultural output associated with the Chinese reforms. Such reforms are probably far more difficult Dornsbusch, Rudiger, and Holger Wolf. 1990. "Monetary Overhang in an industrial setting and in a less authoritarian system. and Reforms in the 1940s." Mimeo. Massachusetts Institute of 31. Discussion of the Hungarian reform provides examples of the Technology. Cambridge, Mass. distinction between the schools of slow and fast privatization, with Ko- rnai (1990) as a slow privatizer and the Blue Ribbon Commission Erlich, Eva. 1987. "Absolute and Relative Economic Development Lev- (1990) arguing for fast privatization. The periods that have been sug- els and Their Structure 1937-1980." Mimeo. Budapest. gested in various studies as appropriate for privatization of a major part of industry range from less than 3 years to around 30 years. 32. Without an adequate information base, a stock market estab- Feige, Edgar L. 1990. "Perestroika anPdaper 1. International Center for lished at an early stage of reform would not be able to offer a useful ba- Economic Growth? Marchk sis for valuation or any indication of desirable investment patterns. The initial purpose of the exchange is, however, quite different. 33. See Lipton and Sachs (1990) for presentation of a rapid priva- Fink, G., and P. Havelik. 1989. "Alternative Measures of Growth and De- tization scheme for Poland involving the distribution of shares. velopment Levels: Comparisons and Assessments." In Joint Econom- 34. In the early stages of reform, many potentially viable firms are ic Committee, Pressures for Reform in the East European likely to experience severe liquidity, and possibly solvency, difficulties Economies. Washington, D.C.: US Government Printing Office, Oc- because of their exceptionally depressed markets, disruption of CMEA tober. trade and exceptional interest charges. These short-run difficulties greatly complicate the task of deciding whether to close or support a Hewitt, Ed A. 1988. Reforming the Soviet Economy: Equality Versus given firm. Efficiency. Washington, D.C.: The Brookings Institution. 35. It seems doubtful that the private sector could handle the re- structuring of very large, weak industries because of the potential po- Hinds, Manuel. 1990. "Issues in the Introduction of Market Forces in litical reaction. Eastern European Economies." World Bank. Washington, D.C. Jan- 36. This process is apparently taking place in Yugoslavia. uary. 81 Klaus, Vaclav. 1990. Address to the Conference on Development Eco- Rosefielde, Steven. 1990. "Market Communism at the Brink." Global nomics, World Bank. Washington, D.C., May 1990. Affairs. Kornai, Janos. 1990. "The Road to a Free Economy: Shifting from a So- Svejnar, Jan. 1990. "A Framework for the Economic Transformation of cialist System: The Case of Hungary." January. Czechoslovakia." University of Pittsburgh. Pittsburgh, Pa. January. Lin, Justin Y. 1989. "The Household Responsibility System in China's Tidrick, Gene, and Chen Jiyuan. 1989. China's IndustrialReforms. New Agricultural Reform: A Theoretical and Empirical Study." A paper York: Oxford University Press. prepared for the 20th Anniversary Conference of the Graduate Pro- gram in Economic Development, Vanderbilt University, Nashville, United States. Central Intelligence Agency. 1989. Handbook of Eco- Tenn., October 15-17, 1989. nomic Statistics. Washington, D.C.: CIA. Lipton, David, and Jeffrey Sachs. 1990. "Creating a Market Economy in Webster, Leila. 1990. "Private Sector Development in Eastern Europe." Eastern Europe: The Case of Poland." A paper presented to the Draft. Industry Division, Industry and Energy Department (IEN- Brookings Institution. Washington, D.C. April. IN), World Bank. Washington, D.C. July. Milanovic, Branko. 1990. "Income Distribution in the USSR." Mimeo. Wharton Economic Forecasting. 1989. World Economic Outlook (Oc- Socialist Economic Reform Unit, World Bank. Washington, D.C., tober). January. World Bank. Various issues. World Development Indicators. Washing- Nordhaus, William. 1990. "The Longest Road: From Hegel to Haggle." ton, D.C.: World Bank. Yale University. New Haven, Conn. Various issues. World Tables. Washington, D.C.: World Nuti, Domenico Mario. 1990. "Stabilization and Reform Sequencing in Bank. the Reform of Socialist Economies." A paper presented to the Eco- nomic Development Institute (EDI) Seminar, World Bank, Wash- Yasin, E. 1989. "Modern Market Institutions and Problems of Econom- ington, D.C., March 1990. ic Reforms in the USSR." A paper presented at the Conference on Economic Reform and Integration, Luxembourg, March 1989. OxfordAnalytica. 1990. July 26. 82 Part IV Stabilization: Design of Programs, Effects on the Economy 8 Stopping Inflation: The Experience of Latin America and Israel and the Implications for Central and Eastern Europe MiguelA. Kiguel and Nissan Liviatan This paper examines the inflation-stabilization expe- and price stability can be sustained are heavily influ- rience of the chronic high-inflation countries of Latin enced by whether the country has mechanisms that America and Israel and relates it to recent events in allow it to live with high levels of inflation. For example, Poland and Yugoslavia. The hope is that, despite the the orthodox approach has been very effective in stop- structural and institutional differences between market ping hyper-inflation in traditionally low-inflation coun- and centrally planned economies, the experience of tries, as evidenced from the experience of Bolivia in the Latin America and Israel can shed some light on the na- mid-eighties and Europe in the 1920s and after World ture of inflation and on some of the difficulties that the War 1I (see, for example, Dornbusch and Fischer 1986, Central and Eastern European countries are likely to Sargent 1982 and Sachs 1986). In addition, this approach face during stabilization. has been effective in stopping inflationary outbursts in Latin America and Israel offer one of the richest traditionally low-inflation countries.' On the other hand, laboratories for studying inflation and stabilization. the orthodox approach has by and large not been very Table 8-1 shows annual rates of inflation for a selected successful in achieving rapid and drastic reductions in group of countries (Israel is not included in the table), inflation in chronic high-inflation countries.2 which can be grouped into three categories based on This paper looks at those aspects of inflation in Latin their inflationary history. First, there are the low- America and Israel that are potentially relevant to under- inflation countries such as Colombia and Costa Rica, standing current developments in Central and Eastem where inflation has remained relatively stable and low by Europe, especially Poland and Yugoslavia. The focus is Latin American standards. Some of these countries have mainly on the following issues: experienced temporary bursts of inflation for one or two years (e.g., Costa Rica in 1982-83), but these episodes (1) what are the similarities and differences in the have usually been reversed quickly, with inflation re- underlying causes of inflation, and where in the infla- turning to low levels thereafter. Second are the chronic tion process were these countries prior to stabiliza- high-inflation countries (in the terminology of Pazos tion? It is argued that Latin America and Central and 1972), whose high rates of inflation have persisted for Eastern Europe have many common elements with long periods. Argentina, Brazil, Chile, Israel, Mexico, respect to inflation. For example, there are many sim- Peru and Uruguay are in this group. The third category ilarities in the reasons for the permanent increase in includes a number of countries that have experienced inflation during the debt crisis in Yugoslavia on the hyper-inflation. Within this group are countries that pri- one hand and in Brazil and Mexico on the other and or to hyper-inflation had had a tradition of low inflation in the characteristics of the inflation that emerged (such as Bolivia and Nicaragua) and others that had ex- thereafter. Likewise, Poland and Yugoslavia experi- perienced chronic high inflation (such as Argentina, enced some of the same mechanisms that fueled in- Brazil and Peru). flation in Argentina and Brazil, especially the A central reason for differentiating among the coun- difficulty of enforcing hard budget constraints on tries' experiences with inflation is that the effectiveness public sector enterprises and of limiting the amount of alternative stabilization strategies largely depends on of credit granted by the banking system. a country's inflationary history. As will be discussed, the (2) the outcomes of the programs in those two so- speed and ease with which inflation can be brought down cialist countries are compared with similar programs 85 Table 8.1. Inflation (As Measured by the Consumer Price Index) in Latin American and Eastern European Countries (percent in annual terms) Year Argentina Bolivia Brazil Chile Colombia Costa Rica Mexico Peru Uruguay Czech. Poland Hungary Romania Yugoslavia 1960 27.30 11.53 29.53 n.a. 3.87 0.79 4.93 8.66 38.50 n.a. n.a. n.a. n.a. 9.30 1961 13.39 7.58 33.42 n.a. 8.67 2.43 1.61 5.92 22.75 n.a. n.a. n.a. n.a. 8.72 1962 28.32 5.88 51.84 n.a. 2.49 2.68 1.20 6.64 10.91 n.a. n.a. n.a. n.a. 10.14 1963 23.90 -0.71 70.08 n.a. 31.98 2.93 0.59 6.07 21.25 n.a. n.a. n.a. n.a. 5.88 1964 22.20 10.18 91.88 45.98 17.64 3.32 2.34 9.79 42.37 n.a. n.a. n.a. n.a. 11.35 1965 28.63 2.86 65.69 28.84 3.51 -0.66 3.57 16.39 56.56 n.a. n.a. n.a. n.a. 33.26 1966 31.91 6.95 41.30 23.08 19.85 0.18 4.22 8.84 73.46 n.a. n.a. n.a. n.a. 25.52 1967 29.20 11.20 30.46 18.75 8.15 1.21 3.02 9.78 89.28 n.a. n.a. n.a. n.a. 6.63 1968 16.21 5.49 22.01 26.32 5.84 4.09 2.33 19.09 125.34 n.a. n.a. n.a. n.a. 5.13 1969 7.57 2.22 22.66 30.44 10.13 2.63 3.37 6.24 20.97 n.a. n.a. n.a. n.a. 9.13 1970 13.59 3.84 22.36 32.48 6.85 4.65 5.21 5.03 16.31 n.a. n.a. n.a. n.a. 9.53 1971 34.73 3.69 20.14 20.03 9.05 3.08 5.26 6.79 23.95 n.a. 1.10 n.a. 0.60 15.70 1972 58.45 6.51 16.56 74.84 13.45 4.60 5.00 7.22 76.48 n.a. -0.10 n.a. 0.00 15.93 1973 61.25 31.49 12.68 361.53 20.76 15.21 12.04 9.49 97.00 n.a. 2.48 3.39 0.70 19.48 1974 23.47 62.83 27.60 504.73 24.28 30.07 23.75 16.89 77.21 n.a. 7.05 1.80 1.09 21.98 1975 182.93 7.98 28.97 374.73 22.93 17.37 15.15 23.62 81.41 n.a. 2.26 3.84 0.20 23.49 1976 443.97 4.50 42.04 211.81 20.23 3.49 15.79 33.48 50.62 n.a. 4.41 5.23 0.50 11.15 1977 176.00 8.10 43.68 91.94 33.05 4.17 29.00 38.05 58.20 n.a. 4.90 3.91 0.60 14.66 1978 175.52 10.36 38.70 40.12 17.79 6.01 17.46 57.85 44.55 n.a. 8.10 4.69 1.98 14.09 1979 159.51 19.73 52.70 33.36 24.70 9.18 18.17 66.69 66.84 n.a. 7.03 8.98 1.84 20.68 1980 100.76 47.23 82.79 35.14 26.54 18.13 26.36 59.15 63.48 n.a. 9.41 9.29 1.52 30.88 1981 104.48 28.57 105.57 19.69 27.48 37.06 27.93 75.43 34.05 0.82 21.20 4.51 2.20 39.83 1982 164.78 133.33 97.78 9.94 24.55 90.12 58.92 64.45 18.99 5.07 100.82 7.02 16.93 31.51 1983 343.82 269.05 142.14 27.26 19.76 32.62 101.76 111.15 49.20 0.95 22.10 6.40 5.19 40.23 1984 626.72 1,281.40 196.98 19.86 16.14 11.95 65.54 110.21 55.30 0.95 15.01 8.65 1.11 54.71 1985 672.15 11,749.60 226.86 30.70 24.04 15.05 57.75 163.40 72.22 2.28 15.10 7.01 -0.39 72.27 1986 90.10 276.34 145.24 19.48 18.88 11.84 86.23 77.92 76.38 0.50 17.69 5.27 -0.08 89.77 1987 131.33 14.58 229.66 19.87 23.30 16.85 131.83 85.85 63.57 0.08 25.21 8.73 n.a. 120.80 1988 342.96 15.99 682.30 14.69 28.11 20.83 114.16 667.03 62.19 0.16 60.00 15.80 n.a. 194.09 1989 3,079.35 15.01 1,286.98 17.03 25.84 16.51 20.01 3,398.59 80.45 1.40 244.55 n.a. n.a. 1,239.87 n.a. Not available Source: International Monetary Fund (various issues). in Latin America and Israel. The working assumption way. Nevertheless, the nature of these programs is ex- is that Poland and Yugoslavia used the exchange rate plored and related to the results of similar ones in Latin as the nominal anchor for their programs, and in this America. (The analysis in the second and third sections sense they pursued exchange rate-based stabilization. draws heavily on the authors' previous work on this In addition, the paper evaluates the positive initial re- topic, in particular, on Kiguel and Liviatan 1988, 1989 sults of Poland and Yugoslavia's programs in reduc- and 1990a).3 The paper wraps up with a discussion of ing inflation and whether the impact the programs policy issues. Special attention is paid to the issue of se- have had on output and the balance of payments is quencing. It is argued that stabilization is difficult and comparable to that under similar programs in Latin costly to maintain unless the privatization issue is ad- America and Israel. dressed from the start. (3) what difficulties have Latin American countries had in sustaining low rates of inflation, maintaining Inflation and Stabilization an external balance and coping with unemployment under their alternative strategies, and what lessons Chronic High-Inflation Countries can be drawn for countries in Central and Eastern Europe? A number of features characterize chronic high- (4) in what ways can the specific institutional char- inflation countries. First, inflation in these countries has acteristics of Central and Eastern European coun- typically gone up over the years. In Brazil and Israel, for tries, especially regarding ownership, affect the example, inflation was around 20 percent in the early evolution of macroeconomic variables during stabili- seventies, 40 percent in the mid-seventies, 100 percent zation? It is argued that a delay in establishing rules in the early eighties and 200 percent toward the mid- for privatization is likely to raise savings (and hence eighties. Although there were more fluctuations in improve the current account), as agents increase Argentina's inflation, the average rate increased from their holdings of liquid assets to take advantage of the the sixties to the seventies and from the seventies to the unique opportunities that privatization of public sec- eighties. tor enterprises will create. A second characteristic is that temporary shocks tend to have a permanent effect on inflation. In Brazil and The paper is organized as follows. The next section Israel the oil shocks caused the increases in inflation describes the main features of chronic high-inflation during the seventies. Likewise, the permanent increase countries and discusses some of the reasons that make in inflation in Brazil and Mexico during 1982-83 was as- disinflation difficult in these types of countries. Also dis- sociated with the maxi-devaluations carried out in re- cussed are the reasons why stabilization in chronic high- sponse to the beginning of the debt crisis. In contrast, in inflation countries is different from that in hyper- low-inflation countries (e.g., Costa Rica and most coun- inflationary ones. It is postulated that Yugoslavia has the tries in the Organisation for Economic Cooperation and basic features of a chronic high-inflation country, while Development [OECD], adverse external shocks have the evidence on Poland is mixed, although in many re- only temporary effects. spects Poland resembles a chronic high-inflation coun- A third characteristic is the asymmetry observed in try. The third section presents a discussion of the inflation: it ratchets up in response to adverse shocks but experience with alternative stabilization strategies in does not fall in the same manner when the economy ex- chronic high-inflation countries. In particular, it exam- periences a favorable shock. This pattern is clear in ines the experience with orthodox and heterodox stabili- Brazil and Israel, where inflation did not drop in the pe- zation programs. While both types of programs use fiscal riods when the terms of trade improved. In practice, in- adjustment as a central component of stabilization, the flation tends to display downward rigidity in these heterodox programs also use price and wage controls economies. Finally, chronic high-inflation countries (usually in the form of a freeze) and a fixed exchange tend to institute mechanisms (such as indexation of rate. Among the orthodox programs, some use a fixed or wages and tax revenues) that allow them to live with in- pre-announced exchange rate as a way to "anchor" the flation. These mechanisms help avert the redistribution price level (exchange rate-based stabilization), while of income and wealth that is usually associated with in- others use money (or domestic credit) as the nominal flation in low-inflation countries, while at the same time anchor (money-based stabilization). The current stabili- reducing the disruption in production that would occur zation efforts in both Poland and Yugoslavia incorporate as a result of nominal shocks. In Brazil, for example, it elements of the two types of stabilization programs and has been argued many times that, because of the index- hence are difficult to characterize in an unambiguous ation of wages, prices and the exchange rate, relative 87 prices have remained stable, and inflation has not ad- Fiure 8-1: Inflation (As measured by the Consumer versely affected production and exports.4 Price Index) and Devaluation, Brazil and Israel, and Paradoxically, this ability to cope with inflation is why Fiscal Deficit and Inflation, Argentina stabilization tends to be difficult and lengthy in these % inAnualTerms A. Brazil countries. Because indexation lessens many of the costs 250 of inflation, the pressures for stabilization are weaker D u than in low-inflation countries, and hence stabilization 200Devaluation--A can be postponed. A lack of credibility about the success . of a stabilization program is usually a more formidable . obstacle in chronic high-inflation countries because 150 - Inflation ' there is no compelling reason in the short run (especially , from a political-economic perspective) to undertake a 1W_u costly process of disinflation when the economy can con- tinue to grow and trade can expand in an inflationary en- vironment.' Unfortunately, if a program is not credible, 50- inflation is likely to display downward rigidity (some- times referred to as inertia) as firms and workers continue o to set prices and wages on the expectation that the pro- 70 72 74 76 78 80 82 84 86 gram will fail. This situation creates an inflationary bias Source. Intemational Monetary Fund. in the economy that not only slows the speed with which inflation comes down but also leads to an overvaluation % in Annual Terms of the currency, a deterioration in the current account and an increase in real wages. In the end, all these factors Devaluation might lead to the abandonment of the stabilization effort. 400 D A second reason that makes disinflation difficult in chronic high-inflation countries is that the causes of in- w flation are not always clear. In fact, the changes in the government's budget deficit do not usually explain the changes in inflation (contrary to what is widely be- 200 lieved). The links between the budget deficit, seigniorage Inflation (or revenue from money creation) and inflation are typ- 1_. ically very weak in these countries. For example, infla- tion in Brazil and Israel has risen in steps over the years 7 1 1 1 1 1 I I despite relatively stable budget deficits and levels of 70 72 74 76 78 80 82 84 86 seigniorage. Liviatan and Piterman (1986) discussed the Source: Intenational Monetary Fund. Israeli case, but Brazil is its mirror image. The pattern of their inflation is peculiar (as illustrated in figures 8-1-A C. Agentina and 1-B): inflation stays relatively stable at a plateau for % of GDP % in Annual Tenms a number of years, it is then destabilized, usually by a 20 700 balance-of-payments crisis, and it subsequently rises to 600 a new, higher plateau. Interestingly, the upward move- I :5' . ment to the new plateau in these episodes is not linked 15 -500 to increases in the budget deficit as predicted by the fis- / cal view, but instead to maxi-devaluations in response 10 - to external shocks. On the other hand, in Argentina . 300 there has been a close relationship between budget /3 deficits (and seigniorage) and inflation (as shown in - / ' 2 figure 8-1-C). The overall evidence shows that the causes 1W of inflation are indeed diverse in chronic high-inflation i -- %** / Inflation countries. There is no single explanation that holds for 0 I. , I I I , , I , , all places at all times. Instead, judgment coupled with 70 72 74 76 78 80 82 84 86 88 good economic analysis are essential to understanding Source: De Pblo and Interinaa Financi Statistks. the various episodes. Note.- Inflation rate in 1989 is 3,080 percent 88 Hyper-inflation was an isolated episode. In Bolivia, the rate of inflation in the sixties and seventies was not much different from Stopping hyper-inflation in countries that otherwise that in many industrialized countries. have no tradition of inflation is in many ways easier than This third point is particularly important because it stopping chronic-high inflation. This is confirmed by the suggests that once hyper-inflation is stopped, inflation recent stabilization program in Bolivia (basic data on returns to its previous level (which in Bolivia was low). Bolivia are included in table 8-2).6 The characteristics of It also suggests that restoring price stability after hyper- its stabilization resembled the pattern discussed in inflation in a chronic high-inflation country could be Yeager (1981), Sargent (1982) and Dornbusch and very different from stopping it in a low-inflation econo- Fischer (1986): the government stopped the inflation my. To the extent that hyper-inflation does not remove abruptly with an orthodox program that included elimi- the perception that the economy can function in a high- nation of the budget deficit, monetary discipline and a inflation environment, then firms and workers might ex- fixed exchange rate. pect that once the hyper-inflation is eliminated, inflation There are at least three reasons why stabilization is would go back to previous levels. In addition, the gov- easier in hyper-inflationary as compared with chronic ernment might be willing to tolerate a high, but stable, high-inflation countries. First, with hyper-inflation the rate of inflation, which after all is much lower and less causes of the inflation are clear: large budget deficits fi- disruptive than hyper-inflation. Thus, restoring price nanced by money creation. In Bolivia, for instance, the stability after hyper-inflation in a chronic high-inflation level of seigniorage rose significantly (exceeding 12 per- country is likely to be much more difficult and costly cent of gross domestic product [GDP]) during the hyper- than the evidence indicates. inflation, compared with an average of 2 percent of GDP There have not yet been any cases where hyper- in the late seventies and early eighties. These levels of inflation has been stopped in chronic high-inflation seigniorage are extraordinarily high by international countries. However, there are a few instances in Latin standards and cannot be sustained for prolonged periods America of chronic high-inflation countries experienc- at a stable rate of inflation.7 Second, hyper-inflation is ing hyper-inflation for the first time (namely, Argentina, extremely costly and hence is not sustainable. In Bolivia, Brazil and Peru). The failure of recent attempts to stop growth collapsed and tensions mounted during the hy- hyper-inflation in Argentina and Brazil suggest that the per-inflation, conditions that generated support for a se- challenge could be more formidable than the findings of rious stabilization program. Hyper-inflation eliminates Sargent (1982) suggest. many of the credibility problems typical of chronic high- A quick review of these episodes is potentially rele- inflation countries. vant for current developments in Poland and Yugoslavia. A third, and important, reason for the sudden fall in The recent history of inflation in Argentina and Brazil is inflation in Bolivia and in other well-known cases is that summarized in figure 8-2.8 For the first time (in 1989) by and large the countries traditionally had low rates of these two high-inflation countries experienced hyper- inflation for prolonged periods. Their hyper-inflation inflation. In Argentina the acceleration in inflation that Table 8.2. Bolivia: AnnualIndicators Budget Terms deficit Seigniorage Inflation GDP of (% of GDP) (% of GDP) average growth Unemployment trade (1) (2) (3) (4) (5) (6) 1980 9.0 3.2 47.2 1.2 5.8 100.0 1981 7.8 1.6 28.6 -0.4 9.7 99.7 1982 14.7 12.2 133.3 -5.6 10.9 98.1 1983 19.1 10.0 269.0 -7.2 13.0 99.3 1984 27.4 15.9 1,281.4 -4.0 15.5 104.1 1985 11.7 8.8 11,749.6 -4.0 18.0 104.3 1986 1.3 2.3 276.3 -2.9 20.0 81.9 1987 4.1 1.6 14.6 2.2 21.5 n.a. 1988 3.1 1.6 16.0 2.8 18.0 n.a. n.a. Not available. Source: Column (1), 1980-84 from Morales (1988) and 1985-88 from International Monetary Fund (1989), other columns from Morales (1988), columns (2) and (3) from World Bank (various issues). 89 Figure 8-2. Inflation (As measured by the In spite of the similarities in their approaches to stabili- Consumer Price Index), Argentina and Brazil zation, neither Argentina nor Brazil eradicated their in- flation. Instead, it remained around 10 percent a month A. Argentina for a few months and then crept up above 20 percent a % in Amnual Terms month some months later. While it could be argued that 200 the inflation has not receded because the governments could not fully commit to future deficits and hence the 150 R fiscal adjustment was not credible in the longer term, this problem is in fact common to hyper-inflation else- where. Sachs (1986), for example, forcefully argues that I0 I Athe Bolivian program was not fully credible either, and yet low inflation was sustained. 50 - The view here is that the difficulty of stopping infla- tion in chronic high-inflation countries is not fully relat- 1 5 ed to the inability of governments to deal with the fiscal 1985 1986 1987 1988 1989 1990 deficit. Perhaps as important a reason is the generalized perception among agents that these economies can function with high inflation and that the government B. Brazil will not fight inflation if it stabilizes at the old rates. % in Montly Terms In summary, hyper-inflation cannot be stopped al- 9, ways and everywhere in the same way. The inflationary 80 history of the country matters. Stopping hyper-inflation 70- in chronic high-inflation countries is different from 60- _ |stopping it in low-inflation ones. The rate of inflation 50o that prevails once hyper-inflation is stopped most likely will relate to the rate of inflation in place prior to the 40- hyper-inflation. 30- 20 Inflation in Central and Eastern Europe o0 , , , ,Inflationary History 0 -10 While the inflationary history of Central and Eastern 1986 1987 1988 1989 1990 European countries is certainly different from that in Latin American and in Israel, there are many parallels worth pointing out. The comparison is more applicable eventually led to the hyper-inflation started around to the high-inflation countries (namely, Poland and February 1989, coinciding with the collapse of the Plan Yugoslavia). The inflation in Yugoslavia has many fea- Primavera (or Spring Plan); it continued for around a tures in common with the chronic high-inflation coun- year (with a temporary period of price stability in the tries in Latin America. The rate has continuously second half of 1989). In Brazil, the hyper-inflation was exceeded 30 percent a year since 1981, with a tendency shorter, lasting around three months, and the inflation to accelerate over time. In addition, many of the institu- rates were much lower than in Argentina. tional arrangements characteristic of chronic high- The interesting feature is that, although both inflation countries, such as wage indexation and foreign Argentina and Brazil adopted very similar, orthodox sta- exchange deposits, are also prevalent there. bilization programs in March 1990 to deal with the hy- Poland is difficult to classify precisely. Inflation re- per-inflation, neither was successful in achieving price mained in the 20 percent range between 1980 and 1987 stability (of the type achieved in Bolivia).9 The stabiliza- (with the exception of 1982), but in 1988-89 it increased tion programs in Argentina and Brazil had many fea- sharply. It could be argued that Poland is essentially a tures in common with the Bolivian program: a balanced low-inflation country that underwent a temporary in- budget on a cash basis (or even a surplus in some crease in inflation in 1988-89. Alternatively, it could be months); very tight money; and a stable exchange rate. argued that the recent inflationary experience in Poland so already shows some of the signs observed in chronic It is worth pointing out that during this period the high-inflation countries. policy-makers did not use any instrument to anchor the The latter view seems more correct for two reasons. price level. The supply of money was endogenously de- First, although Poland had low rates of inflation for a termined by the need to finance public sector enterpris- number of years, it promulgated wage indexation es; the exchange rate was continuously devalued to and other institutions characteristic of chronic high- maintain an undervalued exchange rate, while wages inflation countries. In fact, the rapid acceleration in in- were indexed. Without a nominal anchor in place, infla- flation that took place in 1988-89 can in part be ex- tion could be easily destabilized even in the absence of plained by the fact that the inflationary bias observed in any fiscal imbalance. This type of problem was also ob- many chronic high-inflation countries was also present served in the chronic high-inflation countries in Latin in Poland. Second, the pattern of inflation after the America (for example, in Brazil in the late seventies and January 1990 stabilization program resembled the re- early eighties)."1 cent experiences of Argentina and Brazil as opposed to The causes of the recent inflation in Poland are some- that of Bolivia, in the sense that inflation remained at what different. The 1980 blip in inflation has many sim- levels that are high by international standards, and cer- ilarities with the experience of other low-inflation tainly above what can be acceptable on a long-term basis. countries attempting to improve the external sector. A policy of devaluations succeeded in reducing the real The Causes for Inflation wage and improving the current account. The high level A separate issue is understanding in what respects the of inflation only lasted a short while, mainly because at causes of inflation in Poland and Yugoslavia are similar that time Poland was a low-inflation country and many to those of Latin America (table 8-3 gives basic macro- of the mechanisms that tend to perpetuate inflation in economic indicators for the two countries). In Yugosla- chronic high-inflation countries were absent. On the other hand, the more recent inflationary episode was dif- via, as was the case in Brazil and Israel, the increase in inflation in the early eighties was associated with the ferent. According to Lipton and Sachs (1990), an auster- maxi-devaluations undertaken to improve the external ity program in early 1988 based on cuts in subsidies and sector and reduce the real wage."0 Because formal and sharp increases in consumer prices caused the accelera- informal indexation mechanisms were in place, the tion in inflation. They also argue that because the pro- period of devaluations led to an increase in inflation. gram lacked popular support, it rapidly confronted a tide This interpretation of events is consistent with Rocha of wage demands. By early 1989 the price spiral acceler- (1990), who argues that the accelerations in inflation in ated further as a result of the adoption of wage index- 1983-85 and 1987-89 were largely driven by exchange ation and the introduction of foreign exchange- rate shocks. As in Brazil, the devaluations that triggered denominated deposits, a situation that was compounded the increase in inflation were used to reduce the real later that year by the liberalization of food and agricul- wage in order to restore external balance. tural prices."2 While there was an increase in seigniorage in recent These developments indicate that, as in Yugoslavia, years coinciding with the large acceleration in inflation there was no nominal anchor for the price level. Wages (see table 8-3), the causality between seigniorage and were indexed, the exchange rate followed a crawling peg inflation is difficult to establish unambiguously in and the money supply was accommodating. One differ- Yugoslavia. It is conceivable that, unlike the case of ence, however, is that in contrast to other stabilization Bolivia, seigniorage increased as a result of a policy of ag- programs discussed in this paper, the Polish program did gressive devaluations, whose main objective was to not attempt to improve the current account (since it had improve export performance. In the end, however, they already done so in the early eighties). A second related, led to an increase in inflation. The data presented in and perhaps more important, difference is that real wag- Rocha (1990) broadly support this view. The argument es increased during this period, as opposed to the fall ob- could be as follows: because the devaluations increased served in programs aimed at improving the current the costs of servicing the foreign-denominated debt and account. The implication is that increases in the nomi- raised the cost of imported inputs, they led to larger def- nal wages were the main force behind the inflation. As in icits of public sector enterprises. Ultimately the National Yugoslavia, seigniorage also increased significantly in Bank of Yugoslavia, as lender of last resort, financed 1989, coinciding with the time when inflation reached these losses, printing money to keep the enterprises hyper-inflationary levels, although the causality between functioning, a measure that was reflected in a higher lev- seigniorage and inflation remains unclear, as in many el of seigniorage. other instances. 91 Table 8-3.A. Macroeconomic Indicators, Yugoslavia, 1980-89 N Indicators 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Inflation (%) 30.9 39.8 31.5 40.2 54.7 72.3 89.8 120.8 194.1 1,240.2 Devaluation (%) 29.9 41.9 43.8 84.7 64.6 76.8 40.4 94.4 242.2 1,041.3 Real wage (index) 119.6 117.6 113.7 103.7 97.3 100.0 108.1 100.3 92.5 n.a. Real exchange rate (index) n.a. 98.2 85.8 80.2 97.0 99.9 100.7 82.1 91.2 108.6 GMPa growth (%) 2.3 1.4 0.5 -1.0 2.0 0.5 3.6 -1.0 -2.0 n.a. Industrial production (index) 87.7 91.9 90.6 91.7 96.9 100.0 103.7 103.1 103.5 105.0 Industrial employment (index) 85.5 88.7 91.5 93.7 96.7 100.0 103.8 107.0 107.4 107.4 Consumption as percent of GMPI 52.7 51.8 51.5 51.1 48.2 47.4 44.7 65.0c 67.3c 69.4c Investment as percent of GMPa 35.1 31.0 29.6 25.2 21.9 21.8 20.1 22.6c 20.9c 19.9c Balance of payments (millions of US dollars) Exports 8,978.0 10,929.0 10,241.0 9,914.0 10,254.0 10,622.0 11,084.0 11,803.0 12,779.0 13,560.0 k4 Imports 15,064.0 15,757.0 13,334.0 12,154.0 11,996.0 12,223.0 13,096.0 13,010.0 13,329.0 15,002.0 Trade balance -6,086.0 -4,828.0 -3,093.0 -2,240.0 -1,742.0 -1,601.0 -2,012.0 -1,207.0 -550.0 -1,442.0 Current account -2,291.0 -635.0 -464.0 274.0 559.0 833.0 1,100.0 1,219.0 2,487.0 2,427.0 Public sector (1% of GSP)b Revenues (broad def.) n.a. 38.7 37.2 36.7 34.8 33.7 38.2 34.5 33.5 34.7 Expenditures (narrow def.) n.a. 31.9 30.2 29.4 27.2 26.1 30.3 29.4 25.8 26.3 Deficit Enterprise losses 1.6 1.3 2.2 2.9 2.1 2.8 3.0 6.6 5.7 15.0 Seigniorage Currency n.a. n.a. 1.6 1.3 1.2 2.0 2.8 2.0 2.5 5.9 Ml n.a. n.a. 4.1 3.9 4.0 3.4 5.7 5.2 5.7 5.7 Reserve money n.a. n.a. 8.0 12.6 12.0 11.6 13.9 25.8 33.5 78.1 n,a. Not available. a. GMP. Gross material product. b. GSP. Gross social product. c. Based on changes in % of GSP. Source: World Bank and International Monetary Fund data. Table 8-3.B. Macroeconomic Indicators, Poland, 1980-89 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Inflation (%) 9.4 21.2 100.8 22.1 15.0 15.1 17.7 25.2 60.0 244.5 Devaluation (%) 10.1 15.7 65.8 7.9 23.7 29.9 19.1 51.2 62.4 234.3 Real wage (index) 118.8 124.8 94.0 95.8 96.9 100.0 102.3 99.0 112.6 126.0 Real exchange rate (index) 70.9 78.5 97.9 118.9 116.3 100.0 78.2 56.5 51.6 57.2 GDPa growth (index) n.a. -9.98 -4.8 5.6 5.6 3.6 4.2 2.0 4.5 109.6 Industrial production (index) 101.11 87.4 85.8 91.4 96.4 100.0 104.3 107.5 112.7 96.9 Industrial employment (index) 108.2 107.7 102.2 101.2 100.5 100.0 100.2 99.7 98.1 94.8 Consumption as percent of GDPa 66.9 74.1 62.7 64.8 63.3 61.7 61.1 60.0 57.1 76.7 Investment as percent of GDPa 24.7 18.7 20.1 20.1 20.8 21.3 21.9 22.6 22.5 21.5 Exports as % of GDPa 28.1 23.2 19.4 17.2 17.7 18.3 18.2 22.6 22.8 18.3 Imports as % of GDPa 31.1 25.3 17.3 15.5 15.7 16.9 16.8 19.0 20.0 16.6 Trade balance as % of GDPa -2.9 -2.1 2.1 1.8 2.0 1.4 1.4 3.6 2.7 1.7 Current account as % of GDPa -5.8 -8.2 -4.0 -2.3 -L-7 -1.3 -0.9 -0.4 -1.0 -2.7 Revenues as % of GDPa 48.3 41.7 43.3 39.1 39.7 40.8 39.9 34.2 35.6 30.8 Expenditures as % of GDPa 49.5 53.2 46.3 41.2 41.9 41.9 41.0 37.7 37.0 37.0 Deficit as % of GDPa 1.2 11.4 2.9 2.1 2.2 1.2 1.1 3.5 1.4 -6.2 n.a. Not available. a. GDP. Gross domestic product. Source: World Bank and International Monetary Fund data. Soft Budget Constraints Within this approach there is money- and exchange rate- based stabilization, depending on which of these vari- A common problem in these two countries was the dif- ables is used as the nominal anchor. The heterodox ap- ficulty in imposing hard budget constraints on public sec- proach, which in effect is a version of the orthodox tor enterprises. A firm that has a soft budget constraint exchange rate-based stabilization, initially supports the has an incentive to run deficits, which in the end are fiscal adjustment with price and wage controls to deal financed by credit from the banking system. In effect, this with the inertial aspects of inflation. system amounts to granting firms the "privilege" of in- There is extensive literature analyzing orthodox and directly printing money (or getting seigniorage), which heterodox programs in Latin America and Israel (for ex- is a source of inflation. There could be an additional in- ample Bruno, et al. 1988, Kiguel and Liviatan 1988 and flationary bias if, as shown inAizenman (1989), each firm 1989, and Ramos 1986, among others). This paper sum- chooses its deficit optimally, taking the total level of marizes the main findings with the objective of drawing seigniorage in the economy as given (i.e., each firm lessons for Central and Eastern Europe. decides its "seigniorage" while keeping constant the amount of seigniorage that other firms receive). How- The Orthodox Approach ever, if all firms behave in this manner, two things happen: the rate of inflation increases; and each firm gets a smaller The orthodox approach has been effective in stop- amount of seigniorage than expected because the demand ping hyper-inflation episodes in low-inflation countries. for money falls as a result of the rise in inflation. Para- It has been less effective in the chronic high-inflation doxically, total seigniorage could decline if the economy ones, especially in the short run. is already on the wrong side of the Laffer curve.13 Soft budget constraints are not exclusive to socialist Money-based programs. There are few examples of economies. In Brazil, for example, it is well-known that money-based stabilization programs in chronic high- the central bank has had difficulty controlling the mon- inflation countries. The Chilean stabilization program of ey supply, as it has ended up monetizing the losses of 1974-75 is one of the few. It was a comprehensive effort public sector enterprises and government agencies and that combined a major fiscal adjustment with a mone- providing easy credit to the provincial banks. Likewise, tary crunch aimed at stopping inflation."5 The effect in Argentina much of the monetization in the eighties on inflation was disappointing. Despite a major anti- was caused by the losses of public sector enterprises and inflation effort, the rate fell only marginally between the financial sector and by the deficits of the provinces. 1973 and 1975. This disappointing outcome was accom- Those two Latin American countries have had the same panied by a dramatic rise in unemployment (from difficulty imposing hard budget constraints as the trans- 4.6 percent in 1973 to 16.8 percent in 1975), a large fall forming economies in Central and Eastern Europe are in GDP (exceeding 14 percent in 1975) and an improve- confronting now. ment in the current account. In early 1990 Argentina What the hard budget constraints entail is a central- and Brazil were pursuing monetarist programs to stop ization of control over overall macro-management, with their hyper-inflation (see the previous section). In these especial emphasis on monetary policy.14 In Central and two cases inflation fell initially-monthly inflation Eastem Europe the governments are imposing hard dropped at the beginning to one-digit levels-but the budget constraints on the enterprises at the same time success lasted for only several months: inflation later they are following policies aimed at decontrolling prices bounced back to rates above 20 percent a month in both and decentralizing production decisions. These two ob- countries. As in Chile, both programs were recessionary jectives are compatible, and in effect they represent the and led to an improvement in the current account. reversal of the previous system in which the decentrali- The sluggishness with which money-based programs zation of macro-decisions created an inflationary bias in reduce inflation and their high costs in terms of output the economies, while centralization of production deci- and unemployment explain why this approach is seldom sions resulted in widespread inefficiencies. used in the chronic high-inflation countries. In most cases where money-based programs have been adopted, Stabilization Strategies the monetarist phase has been short, with the authori- ties eventually shifting to a strategy that uses the ex- Broadly speaking, two alternative stabilization strat- change rate as the nominal anchor.'6 egies can be distinguished. The orthodox approach is based on elimination of the budget deficit and the use of Exchange rate-based stabilization. Exchange rate- money or the exchange rate to anchor the price level. based stabilization is a program designed to reduce infla- 94 tion that combines a package of fiscal adjustment with in both cases the exchange rate was the main nominal the exchange rate as the nominal anchor. The exchange anchor and was fixed at the beginning; and both main- rate rule can take the form of a fixed exchange rate (as in tained the fiscal adjustment throughout.'9 There were Chile in 1980) or a pre-announced rate of devaluation differences in the degree to which the two countries ap- (as in the tablitas in the late seventies in Argentina, plied controls. In Israel, the controls were economy- Chile and Uruguay).17 wide, while in Mexico the government opted to allow a Exchange rate-based stabilization is usually more large number of prices to be determined freely. However, effective than money-based stabilization in bringing these differences were ones of degree and not substance. down inflation. In Chile, for example, the combination of Basically, the philosophy behind the two programs was a tight fiscal stance and the announcement of a schedule the same.20 of future daily values of the exchange rate embodying de- One feature of heterodox programs is the ease with creasing rates of devaluation was more effective in re- which they bring inflation down during the early phase ducing inflation than was the monetarist phase. of the program. This initial fall in inflation, however, is Nevertheless, the inflation was very persistent, remain- not an indication of success because this outcome is ing above the pre-announced rate of devaluation for a common to both successful and unsuccessful programs prolonged period. As a result, the real exchange rate ap- (the latter include those that imposed price-wage con- preciated, and in the end the economy was stuck with an trols but did not persist on the fiscal side). A second fea- overvalued currency, difficulties in the external sector ture is that the initial costs of bringing inflation down and the threat of a balance-of-payments crisis. are not very large (as opposed to under money-based or- The experience with the pre-announced exchange thodox programs). In Israel, unemployment went up rate in Argentina and Uruguay, as well as in other coun- marginally for just two quarters, while in Mexico there tries that adopted exchange rate-based stabilization pro- were no indications of costs in terms of unemployment grams, indicates that overvaluation is unavoidable in and output growth as a result of the stabilization efforts. programs of this type and needs to be viewed as a neces- The first stage is, however, the easy part of a hetero- sary cost for stopping inflation. Because of this problem, dox stabilization program. Typically, the difficulties however, countries need to devise a strategy for solving appear later. By using the heterodox approach, policy- the overvaluation issue. Typically the solution is to makers tend to postpone many of the problems that ap- change the nominal anchor as devaluation of the ex- pear early on in money-based orthodox programs. The change rate is used to change the real exchange rate. At use of controls does not remove the credibility problem least transitorily, money should replace the exchange that orthodox programs face in chronic high-inflation rate as the nominal anchor (the strategy adopted in Chile countries, and this difficulty has to be confronted later in 1982-83 to overcome the real appreciation). when prices and wages are liberalized (the stage in which flexibility is created). This stage is the more diffi- The Heterodox Approach cult one in a heterodox program, since only then do policy-makers have to rely on traditional nominal an- Heterodox stabilization programs are those that sup- chors (either money or the exchange rate) to bring infla- plement orthodox measures-namely, tight fiscal and tion down. The typical problems in the flexibility stage audit policies and a fixed exchange rate-with income are a resurgence of inflation, an increase in real wages, policies. They are usually the first stage in a long-term high interest rates, an appreciation of the real exchange stabilization effort. The distinctive feature of these pro- rate and a deterioration in the current account. grams is the initial and temporary use of price and wage controls and a fixed exchange rate to achieve a rapid re- Outcomes under Exchange Rate- and Money-based duction in inflation. Once the controls are removed, the Stabilization program essentially becomes an orthodox exchange rate-based stabilization program. Most orthodox pro- The outcomes of money- and exchange rate-based grams start with a period of tight money before they stabilization not only differed in terms of inflation but switch to the exchange rate as the nominal anchor-one also the evolution of output, the current account, the difference between orthodox and heterodox programs. real exchange rate and other key macroeconomic vari- Few programs satisfy the definition of heterodox pre- ables.21 sented here. There were just two examples in the eight- ies: the Israeli program of 1985 and the Mexican Exchange rate-based stabilization. Most of the ex- program of 1987-88.'1 Both programs initially used change rate-based stabilization programs are character- income policies to achieve a rapid reduction in inflation; ized by a sequence of expansion and recession in GDP 95 growth and by a deterioration in the current account.22 ing) even when the expenditure boom subsides. This fac- This pattem is robust in the sense that it is observed in tor exacerbates the recession. programs that started as heterodox as well as those that started as orthodox and is common to those that did and Money-based stabilization. In view of the difficulties did not undertake a robust fiscal adjustment. For exam- found with exchange rate-based stabilization, is it not ple, a large number of programs in Argentina and preferable to use the money supply as the nominal an- Uruguay began with a reduction of the fiscal deficit, chor? A basic feature of money-based stabilization is that which, however, was maintained only for a limited time the recessionary costs appear up front, in contrast to ex- (from one to three years).2u These stabilizations were change rate-based programs, where the costs appear at a characterized by initial booms in output, increases in later stage as a result of the real appreciation and the de- real wages, real appreciations and subsequent reces- terioration of the current account. The immediate reces- sions. In some cases the collapse of the exchange rate re- sionary effect of money-based programs is mainly gime was associated with a severe crisis (such as in attributable to the short-term downward rigidity in 1980-81 inArgentina and the 1972 collapse in Uruguay). prices and wages and the lack of credibility, which in part Similar developments were also observed in exchange may be related to skepticism about whether the govern- rate-based stabilization programs that continued to be ment will stick with tight money. supported by a fiscal adjustment. This pattern prevailed Another difference between the two strategies is that, in the Chilean and Uruguayan tablita-type stabilizations with money-based programs, it is unusual that targets around the end of the 1970s and in the Israeli stabiliza- be set for expansion of the money supply, while under ex- tion of 1985. In the case of Chile, the fiscal accounts were change rate-based programs it is very common to an- in surplus throughout the last three years of the pro- nounce targets for the exchange rate (including a full gram, while Israel ran a surplus in the first two years. Yet peg or a pre-announced rate in the tablita fashion). A in all these cases output cycles were observed, ending in number of reasons can explain this difference. First, severe recessions in the Southern Cone countries and a there is a stronger incentive to set targets for the ex- prolonged recession in Israel (Kiguel and Liviatan change rate because its definition is clear and it is easy 1990a) . Real appreciation of the domestic currencies and to measure and control. With the money supply, on the the emergence of current account problems were also other hand, there are different monetary aggregates that thebeergen(ecept ofth curentf acco . pols werenals could be subject to targets, and information on these enon of temporary growth cannot be explained solely by aggregates is not readily available. Second, the difficulty enonck of teprar growth c of predicting the demand for money during disinflation Wat isK OTh expalanation for the difficulty of maintain- makes it very hard to establish targets for the money What is the explanation for the difficulty of maintain- supply (especially in the case of the narrow aggregates). ing a stable exchangertw nath onmns bu Increases in the money supply are not inflationary if ac- getmisnbalanced?lacks Onedpibley e ppltonise that anth companied by an increase in money demand. Under an nominal anchor lacks credibility. Suppose that agents exchange rate rule, on the other hand, this problem does expect that the government will not sustain the program not arise because money is endogenously determined over the longer run and that it may end up in a balance- through the balance of payments. Third, the costs asso- of-payments crisis, as has been true in the past. In this ciated with a lack of credibility (and the persistence of in- case agents will expect tight credit conditions in the fu- flation) are easier to absorb in the short run under an ture or the introduction of trade restrictions, expecta- exchange rate rule than under a monetary rule. In the tions that will induce them to shift expenditures to the former case, they usually lead to a real appreciation and present period. In an exchange rate-based stabilization a deterioration in the current account; however, because these expectations tend to become self-fulfilling, since real interest rates are relatively low, an exchange rate the central bank supplies the foreign exchange reserves rule does not usually have strong recessionary effects. to finance the expenditure boom and/or capital flight. Under a monetary rule, on the other hand, persistence in This situation may explain the actual consumption inflation leads to even tighter money, high interest rates booms often observed in these programs. and a more pronounced recession. This expansionary tendency may also explain part of In comparing the two strategies, it is observed that the rise in real wages and the real appreciation.24 In ad- exchange rate-based stabilization has the advantage of dition, a lack of credibility in the persistence of the offi- avoiding large initial recessionary costs.25 However, it cial exchange rate policy will give rise to expectations of should not be thought that the recessionary costs of devaluations in excess of the actual ones, a situation that bringing inflation down can be avoided altogether. There will tend to raise real wages (or prevent them from fall- is usually a period of real appreciation even with ex- 96 change rate-based programs, as inflation persists while There are at least two reasons why these exchange the government adheres to the exchange rate rule. A rate-based programs were more recessionary that simi- maxi-devaluation is-not usually the solution to this re- lar programs in Latin America. First, as argued by Calvo cessionary period because it amounts to abandonment of and Coricelli (1990), in part the recession was driven by the rule. In addition, the country needs to reduce domes- a tightening of domestic credit that worked through the tic expenditures in the future in order to service the ex- supply side. In addition, as happens in most monetarist ternal debt that accumulated during the period of programs, the current account improved. In this re- current account deficits. spect, the programs had a stronger monetarist flavor than did many exchange rate-based programs in Latin The Stabilization Programs in Yugoslavia America. and Poland A second, important factor for the recession cum im- provement in the current account is related to the delay An Impressionistic Look at the Outcomes in defining a policy for the privatization of public sector enterprises. It is now certain that most public sector en- The stabilization programs in Yugoslavia and Poland terprises will be privatized in the near future and that combined elements of both heterodox and orthodox this move will create great opportunities for those indi- strategies. The heterodox component was the use of wage viduals who are liquid. It will also create an incentive to controls (either directly or through tax-induced incen- increase private saving, so that domestic absorption will tives), which in Yugoslavia were supplemented by price decline and the current account improve. The improve- controls for a small number of goods (around 20 percent ment in the current account should not, however, be of the consumer basket). Prices were by and large more taken as evidence that the programs are credible, in part flexible than in most heterodox programs in Latin because it is not yet clear whether they will succeed and America.26 The orthodox part was the fiscal adjustment, what will happen to the exchange rate in the case of fail- of which a central part in both countries was the impo- ure. Instead, the improvement in the current account sition of hard budget constraints on public sector enter- indicates that the potential gains from purchases of pub- prises. This measure was crucial because to a large extent lic enterprises (the saving motive) outweigh the increase the deficits of these enterprises rather than that of the in demand resulting from the temporary nature of the central govemments were the source of the money cre- stabilization effort. More importantly, it also suggests ation (in this respect the nature of the deficits were sim- that resolving the privatization issue will add to the re- ilar to Argentina and Brazil's). Both programs fixed the covery of economic activity. exchange rate (which in effect became the main nominal Finally, it seems that both countries are still in the anchor) and tightened domestic credit to the enterprises. early stages of their stabilization process. While they While there are similarities regarding the behavior of succeeded in bringing inflation down before experienc- inflation and output, developments were not exactly the ing full-blown hyper-inflation, none has achieved sus- same. Inflation fell drastically in the two countries: in tainable price stability. Experience shows that the most Yugoslavia it was stopped in its tracks (as in past hyper- difficult task is sustaining a low rate of inflation. Many inflations), while in Poland inflation remained at a low programs in chronic high-inflation countries have level (around 4-5 percent a month). The disinflation led maintained low levels of inflation for prolonged periods, to large output costs in both countries, which were espe- but eventually their initial success was reversed because cially large in Poland. it could only be maintained at the cost of an overvalued How do these outcomes compare with the experience exchange rate, very high real interest rates, low real wag- of other countries? First, it seems that the costs of reduc- es, or unrealistic expectations about the sustainability of ing inflation were larger than was true under the ex- the cuts in government expenditures or the increases in change rate-based stabilization programs in Latin tax revenues. America and Israel, and certainly larger than those in- curred in stopping hyper-inflation. In fact, the reduction Policy Issues Based on the Latin American Experience in output and the improvement in the current account were more common under money-based stabilization There are three additional lessons from Latin America programs (e.g., Chile 1973-75). At the same time, the that could be relevant for Central and Eastern Europe. rapid initial fall in inflation is characteristic of programs First, none of the successful stabilization programs that rely (even if partially) on some type of wage controls (including Israel, Mexico and Bolivia's) has been able to (the heterodox component), an element that was present bring annual inflation down below 20 percent. This situ- in both countries. ation is a puzzle, because the same outcomes are observed 97 in Bolivia, which stopped a hyper-inflation, and in the and inventories, and less concern about the long-term chronic high-inflation countries (e.g., Chile, Mexico and viability of the firm. Israel). In addition, the phenomenon does not seem to be The above discussion indicates that imposing a hard related to the specific form of the exchange rate rule- budget constraint on firms while the issue of ownership Mexico adopted a crawling peg, while Israel used discrete is not settled could have detrimental effects on the capital devaluations-and cannot be explained by fiscal consid- stock and on the long-term viability of firms and hence erations, since deficits do not appear to have been a major growth. It argues in favor of settling the ownership issue factor in affecting inflation in any of these countries. before imposing the hard budget constraint on firms. Second, stabilization has proved to be a long-term en- deavor in the chronic high-inflation countries in Latin Notes America where markets already existed. It is likely to take even longer in Central and Eastern European coun- 1. As discussed in Lizano and Charpentier (1990) with respect to tries where a large number of structural reforms need to 2. For a discussion of this topic see Kiguel and Liviatan (1988). be undertaken simultaneously. 3. These were prepared as part of the World Bank research Third, and a related factor, the costs of failure should project, "Stopping High Inflation," RPO 674-24. not be underestimated. Each new stabilization attempt 4. While indexation usually lowers the costs of nominal shocks, at the same time, by reducing the variability in relative prices, it also that fails to bring inflation down makes stabilization makes the adjustment to real shocks more difficult and costly. more difficult the next time around. The repeated failure 5. The costs of inflation appear much later, usually as inflation of stabilization attempts in Argentina and Brazil since the reaches hyper-inflationary levels. Experience indicates that there is no Austral and Cruzado plans are good examples of what such thing as a high stable rate of inflation. Once inflation is high, if it is not brought down through a drastic and persistent stabilization pro- happens. Both countries experienced a period of infla- gram, it will eventually run out of control. tion-stabilization cycles in which the inflationary out- 6. The Bolivian hyper-inflation and subsequent stabilization are bursts became more pronounced after each failed analyzed in Sachs (1986) and Morales (1988), among others. attempt at stabilization (see figure 8-2), eventually reach- 7. For a thorough discussion of this topic see Kiguel and Liviatan ing hyper-inflationary levels, while the difficulty in keep- (1988). 8. See Kiguel and Liviatan (1990b) for a comparison of the recent ing inflation low, even for short periods, became larger inflation-stabilization experience in Argentina and Brazil. over time. These arguments can be used to make a case 9. Boliviadid not fully eliminate the inflation: it has held at an an- for maintaining and deepening the stabilization effort. nual rate of around 20 percent since stabilization. Finally, it is important to examine briefly in what 10. Rocha (1990) provides an extensive analysis of inflation in Yugoslavia in the eighties. ways stabilization in centrally planned economies could 11. There are many examples of inflationary episodes caused by a be different from that in market economies as a result of policy of aggressive devaluations aimed at achieving a real depreciation the differences in institutional arrangements and the (e.g., Mexico in 1987 and Israel in 1984). The perception that the real system of incentives. Specifically, it is argued here that exchange rate (a relative price) can be affected by manipulating the ex- change rate is misleading, because it ignores that domestic prices ad- the imposition of hard budget constraints on firms when just to bring the real exchange rate to a level that is consistent with the the issue of ownership is not settled, as is now the case underlying policies and overall macroeconomic conditions. in both Poland and Yugoslavia, changes the nature of the 12. There are also cases in Latin America of increases in inflation relation between firms and workers and could lead to an that were driven by wage push pressures of the type just described. In Argentina, for example, in 1983-84, first the military government and undesirable response by managers. Prior to the imposi- then the Alfonsfn administration attempted to increase real wages by tion of a hard budget constraint, there is an implicit raising nominal wages. The policy backfired in the end, as is now hap- long-term relationship between a firm and its workers pening in Poland, because inflation increased the rise in nominal wag- because the firm is likely to continue functioning in the es and thereby reduced the real wage. future and hence workers have implicit tenure. 13. Some of these issues are analyzed in Aizenman (1989). 14. In Argentina, for example, the finance minister recently re- Although the state is the formal owner of the firm, work- quested personal control over the budgets of public sector enterprises, ers still identify with the interests of the firm. a measure that increased centralization of decision-making. This ar- Once the hard budget constraint is imposed, the na- rangement was devised as a way to impose hard budget constraints on ture of this relationship changes. Workers in effect lose the enterprises. 15. For a thorough analysis of this program see, for example, their tenure, since they are not sure whether the firm Ramos (1986), Corbo (1986), Edwards and Edwards (1987), and Corbo will survive under the new rules and whether they will and Solimano (1990). retain their jobs. This situation causes a shift in the ob- 16. This sequence was observed in the Chilean program just dis- jective function of the workers (and hence managers), cussed, as well as in the 1959 and 1976 stabilization programs in Argentina. who place more emphasis on maximizing short-term 17. These programs are described and analyzed in Corbo and de revenue, even at the cost of depleting the capital stock Melo (1985) and Ramos (1986), among others. 98 18. The Mexican program cannot be fully evaluated yet, since two Cardoso, Eliana, and Albert Fishlow. 1990. "The Macroeconomics of and a half years is not a long enough period according to the authors' Brazilian External Debt," pp. 269-391. In Jeffrey Sachs, ed., Devel- criteria. Progress has been good, however, and the prospects are prom- oping Country Debt and Economic Performance. Chicago, Ill.: ising. University of Chicago Press. 19. The Israeli program is analyzed in detail in Bruno and Piterman (1988), Bruno and Meridor (1990) and Liviatan (1988 and Corbo, Vittorio. 1986. "The Use of the Exchange Rate for Stabilization 1990), among others, and the Mexican program in Ortiz (1990). See Purposes: The Case of Chile." In M. Connally and C. Gonzalez, eds., also Kiguel and Liviatan (1989). Economic Reforms and Stabilization in Latin America. New York: 20. There also were a number of heterodox programs in the six- Praeger. ties in Latin America. The Brazilian stabilization program of 1964-67 (analyzed in Simonsen 1974 and Cardoso and Fishlow 1990, among Corbo, Vittorio, and Jaime de Melo. 1985. "Liberalization with Stabili- others) is an example of a successful one. The Austral plan in Argentina zation in the Southern Cone of Latin America." World Develop- in 1985 (analyzed in Heymann 1987 and Kiguel 1989, among others) ment, special issue (August). started as a heterodox program, but the government later relaxed the fiscal adjustment and abandoned the program. Corbo, Vittorio, and Andres Solimano. 1990. "Chile's Experience with 21. Heterodox programs are considered here as exchange rate- Stabilization Revisited." Mimeo. World Bank. Washington, D.C. based stabilization. Although initially the exchange rate is part of a broader set of anchors, it usually becomes the main nominal anchor in the second stage once the controls are removed. Dornbusch, Rudiger, and Stanley Fischer. 1986. "Stopping Hyperinfla- 22. This finding is documented in Kiguel and Liviatan (1990a), tion Past and Present." WeltwirtschaftlichesArchiv (April). where a large number of money- and exchange rate-based stabilization programs in Latin America and Israel are examined in detail. Edwards, Sebastian, and Alejandra Cox Edwards. 1987. Monetarism 23. Kiguel and Liviatan (1990a) discussed these episodes. and Liberalization: The Chilean Experience. Cambridge, Mass.: 24. Calvo (1986) and Calvo and Vegh (1990) explain how a cash- Ballinger Publishing Co. in-advance model can generate these transitory aspects. For alterna- tive explanations, see Kiguel and Liviatan (1990a). Heymann, Daniel. 1987. "The Austral Plan." American Economic 25. 'fypically, heterodox exchange rate-based stabilization pro- Review 77 (2):284-87. grams exhibit some initial recession. However, it was usually small rel- ative to the ones observed under money-based stabilization in the International Monetary Fund. Various issues. Intemational Financial sample of countries studied by the authors and in industrial countries. Statistics. Washington, D.C.: World Bank. 26. The Mexican program is undoubtedly the most similar be- cause most prices continued to be freely determined. In practice, how- Kiguel, Miguel A. 1989. "Inflation in Argentina: Stop and Go Since the ever, there were informal pressures on firms to limit their price Austral Plan." World Bank, Policy, Planning and Research (PPR) increases. Working Paper. Washington, D.C. References Kiguel, Miguel A., and Nissan Liviatan. 1988. "Inflationary Rigidities and Orthodox Stabilization Policies: Lessons from Latin America." Aizenman, Joshua. 1989. "Competitive Externalities and the Optimal The World Bank Economic Review 2 (3)(September):273-98. Seigniorage." Mimeo. University of Chicago. Chicago. -_____ 1989. "The Old and the New in Heterodox Stabilization Pro- Bruno, Michael, and L. Meridor. 1990. "The Costly Transition from Sta- grams: Lessons from the 1960s and the 1980s." World Bank, Policy, bilization to Sustainable Growth: Israel in Its Third Phase." Mimeo. Planning and Research (PPR) Working Paper No.323. Washington, D.C. Bruno, M., and S. Piterman. 1988. "Israel's Stabilization: A Two Year Review," pp. 1-47. In M. Bruno, et al., eds., Inflation Stabilization. . 1990a. "The Business Cycle Associated with Exchange Rate Cambridge, Mass.: MIT Press. Stabilizations." Mimeo. World Bank. Washington, D.C. Bruno, M., G. Di Tella, R. Dornbusch and S. Fischer, eds. 1988. Infla- - 1990b. "The Inflation Stabilization Cycle in Argentina and tion Stabilization. Cambridge, Mass.: MIT Press. Brazil." Mimeo. World Bank. Washington, D.C. Calvo, Guillermo A. 1986. "Temporary Stabilization: Predetermined Liviatan, Nissan. 1988. "Israel's Stabilization Program: A Three Year Exchange Rates." Joumal of Political Economy 94:1319-29. Perspective." World Bank, Policy, Planning and Research (PPR) Working Paper No. 91. Washington, D.C. September. Calvo, Guillermo, and Fabrizio Coricelli. 1990. "Stagflationary Effects of Stabilization Programs in Reforming Socialist Countries: Supply . 1990. "The Process of Restoring Macroeconomic Balance in Side Vs. Demand Side Factors." Mimeo. World Bank. Washington, Israel." Mimeo. World Bank. Washington, D.C. D.C. Liviatan, N., and S. Piterman. 1986. "Accelerating Inflation and Calvo, Guillermo, and Carlos Vegh. 1990. "Credibility and the Dynam- Balance of Payments Crises: Israel 1973-1984," pp. 320 46. In ics of Stabilization Programs: An Analytical Framework." Mimeo. Yoram Ben-Porah, ed., The Israeli Economy. Cambridge, Mass.: International Monetary Fund. Washington, D.C. Harvard University Press. 99 Lipton, David, and Jeffrey Sachs. 1990. "Creating a Market Economy in Rocha, Roberto. 1990. "Inflation and Stabilization in Socialist Coun- Eastern Europe: The Case of Poland." Brookings Papers on Eco- tries: Some Lessons from the Yugoslav Experience." Mimeo. World nomic Activity 1:75-147. Bank. Washington, D.C. Lizano, Eduardo, and Silvia Charpentier. 1990. "External Debt and Sachs, J. 1986. "The Bolivian Hyperinflation and Stabilization." NBER Economic Policy: The Case of Costa Rica." Mimeo. World Bank. Discussion Paper No. 2073. National Bureau of Economic Washington, D.C. Research. New York. Morales, Juan-Antonio. 1988. "Inflation Stabilization in Bolivia," Sargent, Thomas J. 1982. "The End of Four Big Inflations." In R.E. pp. 307-46. In Michael Bruno, G. Di Tella, R. Dornbusch and S. Hall, ed., Inflation: Causes and Effects. Chicago, Ill.: University of Fischer, eds., Inflation Stabilization. Cambridge, Mass.: MIT Press. Chicago Press. Ortiz, Guillermo. 1990. "Mexico Beyond the Debt Crisis, Towards Sta- Simonsen, Mario Henrique. 1974. "The Anti-Inflation Policy." In Mario bilization and Growth with Price Stability." Mimeo. January. Henrique Simonsen and Roberto de Oliveira Campos, eds., TheNew Brazilian Economy. Rio de Janeiro: Crown Editores Pazos, E 1972. Chronic Inflation in LatinAmerica. New York: Praeger. Internacionais. Ramos, Joseph. 1986. Neoconservative Economics in the Southem Yeager, Leland B. 1981. Experiences with Stopping Inflation. Cone of LatinAmerica, 1973-83. Baltimore, Md.: The John Hopkins Washington, D.C.: American Enterprise Institute. University Press. 100 9 Stabilization Programs in Eastern Europe: A Comparative Analysis of the Polish and Yugoslav Programs of 1990 Fabrizio Coricelli and Roberto de Rezende Rochal Dissatisfaction with their performance and concern The contraction of activity was particularly strong in the over the acceleration of inflation led the governments of industrial sector, as indicated by negative growth rates of Poland and Yugoslavia to implement stabilization cum 20 percent and 10 percent, respectively. Although the restructuring programs at the end of 1989. Their pro- contraction in output was concentrated in the first se- grams, which they launched at approximately the same mester of 1990, the prospects for a sustained resumption time (December 18, 1989 in Yugoslavia and January 1, of growth are still uncertain in both countries. 1990 in Poland), comprise measures designed not only The implementation of two apparently similar pro- to stabilize inflation at low levels but also to change the grams by two reforming socialist countries at almost the structure of their economic systems. same time has produced great interest in a comparative The stabilization components of both programs would analysis of their design, performance and sustainability. ordinarily be classified as heterodox, since they include in- Such is the objective of this paper. To this end, the paper comes policy measures in addition to a significant fiscal investigates the possible differences that underlie the adjustment, the imposition of credit controls and the re- similarities of the two programs and that may account laxation of exchange rate and trade controls. The restruc- for the better initial performance of Yugoslavia's pro- turing components of the two programs also share some gram (a sharper reduction of inflation with smaller loss- similarities. Both attempt to force inefficient enterprises es in output). The paper identifies significant differences into bankruptcy, albeit through different methods. How- in the initial conditions in the two countries, as well as ever, in neither case is there a clearly defined strategy to in the sequence and degree of some policy measures. deal with bankrupt enterprises, including what the pre- These differences may explain the disparities in the early cise role of privatization in the overall reform strategy will results. Finally, the paper identifies the most important be. In fact, both countries are still elaborating several as- issues that the two countries will have to address in the pects of their restructuring components. second stage of their programs. These include the un- Both programs brought about a substantial reduction freezing of nominal variables and resolution of the criti- in inflation in the first semester of 1990 without re- cal structural problems affecting both economies. course to widespread price controls. In the case of Po- The next section provides some background informa- land, inflation was reduced from a peak of 78 percent in tion on inflation in the two countries during the 1980s. January to 3.5 percent in June. In the case of Yugoslavia, The following section examines the conditions in the inflation went from a peak of 60 percent in December two economies before stabilization. A comparison of the 1989 to almost 0 percent in May and June 1990. The per- components of the two programs is presented in the sub- formance of the two programs in the second semester of sequent section, followed by a section that analyzes the 1990 was less satisfactory, with monthly inflation rates initial results of the programs and points to the possible of around 5 percent in Poland and ranging between 3 causes of the more severe recession in Poland. It also percent and 8 percent in Yugoslavia. identifies the likely causes of the revival of inflation in Both countries experienced a sharp contraction in Yugoslavia during the second semester of 1990. The final economic activity following the implementation of their section assesses the main issues facing the second stage programs. During 1990, real GDP fell by 12.5 percent of the reform in the two countries. Much more detailed and 7.5 percent in Poland and Yugoslavia, respectively. analyses of the two programs appear in the two appendi- 101 ces (appendix 1 deals with Poland, appendix 2 with Yugo- a system of self-management in the early 1960s. As a re- slavia). sult, there was much less interference in the price sys- tem in Yugoslavia. Unlike in Poland, price controls in Background Yugoslavia were rarely strong enough to repress infla- tion. In fact, they bore a greater resemblance to the con- The histories of inflation in Poland and Yugoslavia trols prevailing in other high inflation countries such as during the last decade are quite different. As shown in Brazil (which has always had controls of some sort). figure 9-1, inflation in Poland was more erratic than that The phenomenon of high inflation in Yugoslavia was in Yugoslavia. The dominant feature in the first half of closely associated with the turnaround in the current ac- the decade in Poland was a jump in 1981-82, while in the count during the 1980s-typically inflation accelerates second half inflation showed an upward trend until as the current account shifts from a deficit to an increas- 1988, followed by another jump in 1989 when it reached ing surplus. Two major factors underlay the close associ- 640 percent. In Yugoslavia inflation was more or less sta- ation between inflation and the current account in ble until 1983. After that year it accelerated almost con- Yugoslavia. First, the need to reduce the real wage to lev- tinuously, reaching 2,700 percent in 1989. In general, els consistent with a given depreciated real exchange the rates of inflation in Yugoslavia were much higher rate required an increase in the rate of inflation-the Pa- than in Poland (except in 1982) and than the OECD av- zos-Simonsen mechanism (see Dornbusch 1987, Pazos erage for the decade. 1978 and Simonsen 1989). This factor became more im- These differences in the path of inflation reflect in portant at the end of the decade, with the increase in real part differences in the structure of the two economies. wage rigidity occurring after the formal introduction of Until the early 1980s Poland had a centrally planned wage indexation in 1987. economy, when it initiated a gradual decentralization. Second, the conditions that prevailed after the rever- Yugoslavia had already replaced its central planning with sal of the flows in external financing considerably wors- Figure 9-1: Inflation in Poland and Yugoslavia, 1980-90 (percentage change per annum) Log Scale (%) 10,000 1,000 1L00 1 0 80 I 81 82 1 83 1 84 85 1 86 1 87 1 88 89 1 90 .................... Poland Yugoslavia 102 ened the financial situation of Yugoslav enterprises, spectively). In addition, the negative transfer of resourc- where the fundamental imbalances of the Yugoslav es from abroad, as measured by the non-interest current economy were located (the non-financial public sector account surplus, was much smaller in Poland (2-3 per- ran surpluses throughout the 1980s). Large losses in for- cent of GNP) than it was in Yugoslavia (5-7 percent of eign exchange as a result of interest and principal pay- GNP).2 ments, combined with stagnant output and increasing These numbers indicate that the underlying rate of overstaffing, resulted in greater enterprise losses in the inflation consistent with "fundamentals" was lower in 1980s. The enterprise imbalances spilled over to the cen- the case of Poland. At the same time, the much stricter tral bank, since the latter had to absorb large amounts of price controls in Poland make it more difficult to associ- foreign liabilities in an effort to protect the enterprises ate the movement of inflation with the fundamentals.3 and commercial banks from even greater financial diffi- In addition, there is some association between price culty. As a result, the central bank ran a quasi-fiscal def- controls and budgetary performance via the subsidies. icit that became an independent source of monetary For instance, the outburst of inflation in 1982 coincided expansion. The various attempts at stabilization failed to with a sharp reduction in the deficit from 1981 to 1982, address the hidden losses in the system and were aban- a phenomenon that occurred again from the first to the doned soon after implementation. second half of 1989. Unlike in Yugoslavia, the non-financial public sector in Poland ran deficits during the 1980s, while enterprise Conditions Preceding Stabilization accounts showed small surpluses. The fiscal deficits were, however, kept at moderate levels (1-2 percent of Conditions in the Polish and Yugoslav economies pre- gross national product [GNP]) during most of this peri- ceding the two stabilization programs are summarized od, the exceptions being 1981 and 1989, when the deficit in table 9-1. In Poland, the sharp increase in inflation was much larger (10 percent and 8 percent of GNP, re- during 1989 was the result of not only the increased fis- Table 9-1. Initial Conditions in Poland and Yugoslavia Variable Poland Yugoslavia Background Inflation 640% in 1989 2,700% in 1989 Output growth 1.4% in the 1980s, 0.7% in the 1980s, 0% in 1989 0.8% in 1989 Rate of unemployment 0% in the 1980s, 13% in the 1980s, 0% in 1989 14% in 1989 Other Conditions in 1989 GNP US$66.2 billion US$67.0 billion Current account US$1.8 billion deficit US$24.billion surplus Gross debt US$40.6 billion US$17.3 billion Reserves US$2.5 billion US$6.1 billion Black market premium 400% average in 1989 10% average in 1989, 40% before the plan 30% before the plan Fiscal/enterprise Fiscal deficit of 7.2% Fiscal surplus of balance of GNP 0.7% of GNP Enterprise losses of 15% of GNP Central Bank losses of 4-5% of GNP Large volume of non- performing loans Seigniorage 13.5% of GNP 12% of GNP Ratio of M3 to GNP 47.8% 33.1% Share of foreign exchange deposits in M3 69.3% 69.5% Source: Various national sources. 103 cal deficit (7.2 percent of GNP, the bulk of which was in- as pledged against obligations). In addition, in Yugosla- curred in the first semester) but also the successive price via the excess demand for foreign exchange at the official shocks related to the liberalization of food prices (Au- rate was mild, as suggested by a barely positive black gust), the first round of adjustment of energy prices (Oc- market premium (figure 9.2). In Poland, the black mar- tober) and repeated devaluations accompanied by the ket premium was still very large in 1989, despite the real introduction of formal wage indexation. devaluations after 1987. The size of the premium there In Yugoslavia, various factors contributed to the open reflected the magnitude of the exchange rate misalign- inflation of 2,700 percent in 1989. First, a large real de- ment and possibly some residual monetary overhang. valuation in mid-1988 in the context of formal wage in- It is important to bear such differences in mind when dexation caused wages and prices to escalate. Second, examining the rationale for some of the preparatory although the government abandoned its target for the measures taken in 1989, especially in view of the goal in real exchange rate in mid-1989, an explosion in real wag- the two programs of substantially reducing the exchange es in the second semester sustained the inflationary rate and trade controls. As shown in figure 9.3, both pressure. Finally, the hidden losses in the system, which countries devalued their currencies in real terms in No- had not been corrected, constituted a permanent source vember and December 1989. However, in the case of Po- of monetary expansion and inflation. Pressure on the land, the more adverse external conditions and the large money supply was further increased by the large build- exchange rate misalignment dictated the need for much up in foreign exchange reserves over the year. In 1989, sharper real devaluations. The same factors seem to ex- seignorage revenues on base money reached 13.5 per- plain why Poland restricted the introduction of convert- cent and 12 percent of GNP in Poland and Yugoslavia, re- spectively. ~~~~~~~~ibility to current account transactions, while in spectively. Yugoslavia convertibility was also extended to the capital The fiscal deterioration in Poland during 1989 was acut partly attributable to the fall in real tax revenues, which account. in turn resulted in part from lags in the collection of tax- Figre 9-2: Black Market Currency Premia in es-the Olivera-Tanzi effect (see Olivera 1967 and Tanzi Poland and Yugoslavia, 1980-90 1977). As shown in table 9.2, the revenue loss amounted to 7 percent of GNP. 1,600 In Yugoslavia, on the other hand, the erosion of fiscal 1,400 revenues in the 1980s tended to be offset by adjustments 1 in the tax rates and contributions, as well as by the par- 1,200 tial indexation of some taxes. In the years when these 1,000 measures proved insufficient to stabilize real tax reve- nues, the government adjusted expenditures so as to 8 ' prevent a deficit. These factors make estimation of the 600 1 ' . i Olivera-Tanzi effect more difficult in the case of Yugosla- AtJ " V A IX via. ' Yugoslavia enjoyed a more comfortable external posi- 200 tion before 1990 than did Poland. The large current ac- 0_ count surplus of 1989-US$2.4 billion-and the successful debt rescheduling of 1988 allowed foreign re- -00 1 1 1 1 1 1 serves to rise to US$6.1 billion at the end of 1989. Po- 1980 1981 19821983198419851986 19871988 1989 1990 1991 land, in contrast, ran a current account deficit of US$1.8 ......... Pbland - Yugolavia billion in 1989 and ended the year with much lower re- serves-US$2.5 billion (much of which was not useable Table 9-2. Total Fiscal Revenues in Poland and Yugoslavia, 1985-89a (percent of GDP) 1985 1986 1987 1988 1989 Poland n.a. 49.4 46.9 48.0 40.8 Yugoslavia 33.7 38.2 34.5 33.5 34.7 n.a. Not available. a. Using a broad definition of the public sector in both countries. Source: World Bank. 104 Figure 9-3. Real Effective Exchange Rate Indices, Poland of M3 to GNP in Poland and Yugoslavia had declined to and Yugoslavia 47 percent and 33 percent, respectively. (1988=100) 180 A Comparison of the Two Stabilization Programs 160 A* A summary comparative description of the Polish and 140 i Yugoslav stabilization programs is provided in table 9.3. . '~, On first examination, the two programs look strikingly 120 ,8 -<,, similar. The income policy components are basically the 100 */" - ...** / ! " ~same, including: (1) a temporary freeze on the exchange 10. ,,*t 's \ Frate that both countries extended to one year; (2) wage so A /\ controls (with a complete freeze in Yugoslavia and par- 80 . tial indexation with very low coefficients in Poland); and 60 X (3) a temporary freeze on public sector prices.4 60 However, note that Yugoslavia made all its price cor- 40 . . . . .. . . . . . . rections before the start of its program, whereas Poland Jan Jan Jan Dec made large corrections to its energy prices (400 percent) 1988 1989 1990 1990 on January 1, 1990. .,, Poland - Yugoslavia Both programs included a significant fiscal adjust- ment-7 percent and 5 percent of GNP in Poland and Both countries adjusted public sector prices at the end Yugoslavia, respectively. In the case of Poland, the ad- of 1989. However, in Poland some of the administered justment was designed to close the fiscal deficit, whereas prices were so low that even after the very sharp correc- in Yugoslavia it was essentially designed to cover new, tions made in that period (100 percent increase in energy non-traditional expenditures. These included the quasi- prices in October), the prices remained substantially be- fiscal operations of the central bank (servicing of foreign low international levels and had to be further corrected liabilities and subsidies to favored sectors), transfers to at the start of the program. the bank restructuring program and the social program, The two countries shared a severe problem of excess and coverage of some enterprise arrears. personnel in the enterprise sector. In the case of Poland, Both programs set targets for the growth of net do- the extent of the problem is indicated by a rate of unem- mestic assets. In the case of Poland, the net domestic as- ployment barely above 0 percent throughout the 1980s. sets of the banking system were to grow at 20 percent in In Yugoslavia the rate of unemployment was around 14 the first quarter and 8 percent in the second, whereas in percent in 1989, a slight increase over the 13 percent av- Yugoslavia these assets were to expand at slightly nega- erage for the whole decade. Although this higher rate of tive rates in the two first quarters. Both programs al- unemployment (concentrated on new entrants) in Yugo- lowed for the monetization of foreign exchange inflows. slavia seems to rule out the existence of overstaffing, a A first reaction to these figures might be that the major cause of enterprise losses was indeed excess per- monetary program in Yugoslavia was much tighter. sonnel, estimated to have reached 20 percent of the labor However, two critical factors necessitate modifying this force in 1988 (Mencinger 1989). conclusion. First, the Polish program allowed for an in- The two countries also shared a problem of flight from crease in domestic credit in anticipation of the price domestic assets and increasing velocity. While hardly a jump that would result from the large correction in en- novelty in high inflation countries, in both countries the ergy prices in January 1990, a jump that later proved to payment of negative real interest rates on domestic de- have been severely underestimated. Second, the Yugo- posits-a policy designed to provide subsidized credits to slav program was preceded by a substantial real mone- enterprises-accelerated the flight from domestic assets. tary and credit expansion in December 1989. As shown In addition, the foreign exchange deposits in the financial in the next section, these and other factors resulted in a systems of both countries are also likely to have facilitat- much tighter monetary stance in Poland. ed the shift out of domestic money. Between 1980 and Both programs called for a substantial relaxation of 1989 the share of foreign exchange deposits in M3 in- the exchange rate and trade controls, although Yugosla- creased from 20 percent to 69 percent in Poland and from via maintained some quota restrictions (amounting to 30 percent to 65 percent in Yugoslavia. In 1989 the ratio 12 percent of total imports). Yugoslavia also introduced 105 Table 9-3. Summary Descrzption of the Polish and Yugoslav Programs Components Poland Yugoslavia Incomes policy 3-month freeze on the 6-month freeze on the exchange rate (dollar) exchange rate (deutschemark) after large adjustment in after some adjustment in December 1989, later December, later extended to extended to 1 year 1 year Partial and lagged wage 6-month freeze on nominal indexation, with very wages small coefficients 6-month freeze on energy 6-month freeze on energy and prices, after large other public sector prices corrections in January (20% of the Consumer Price 1990 Index), after corrections in December 1989 Fiscal policy Expected fiscal adjustment Expected fiscal adjustment of 7% of GNP of 5% of GNP Monetary policy Controls over net domestic Controls over the net assets of the banking domestic assets of the system Central Bank Setting of the discount Setting of the discount rate at 36% per month in rate at 23% per annum; full January 1990, up from 7% liberalization of other in December 1989 interest rates Exchange and Partial convertibility "Full" convertibility (some trade policies (current account only) capital restrictions have remained) Elimination of quotas and Reduction of quotas and relaxation of licenses relaxation of licenses Enterprise/bank Enforcement of bankruptcies Enforcement of bankruptcies reform through the "dividend tax" through 60-day tolerance limit for arrears Evolving plans for Bank restructuring program; enterprise restructuring evolving plans for enter- and privatization prise restructuring and privatization Source: The authors. full currency convertibility, while Poland restricted con- resources. Although the agency was created in mid- vertibility to current account transactions. 1990, enterprise restructuring moved slowly, pending Interruption of the explicit or implicit subsidization the establishment of specialized regional institutions. In of inefficient enterprises and free operation of the Dar- addition, the strategy for linking restructuring and winian-Schumpeterian law of natural selection were im- privatization was still being debated during the course of plicit in both programs. Thus, the programs contained 1990, partly as a result of different positions among the measures to preclude enterprise recourse to inter-enter- various republics. prise credits and arrears,5 so that inefficient enterprises In Poland, the government made a commitment to a would be forced into bankruptcy. In the case of Yugosla- large-scale privatization plan at the start of the program, via, a limit of 60 days of arrears was established as the and the legal framework for the plan was spelled out in a trigger for bankruptcy procedures. The government was privatization law passed in July 1990. However, the leg- able to enforce this rule by means of a centralized system islation was "enabling" rather than "programmatic." of payments among enterprises and banks. In Poland, Thus, no significant cases of bankruptcy (40 to 50 minor enterprises were ordered to pay dividends to the govern- cases of bankruptcies occurred), restructuring or priva- ment based on the book value of their funding capital. tization were observed in the first nine months following The two governments announced plans to restruc- initiation of the program. The recognition that further ture and privatize enterprises and banks. Yugoslavia is- delays in privatization could ultimately jeopardize the sued a bank restructuring program, to be managed by a success of the program led to the introduction of a much bank rehabilitation agency and supported by budgetary more ambitious plan in December 1990. The objective of 106 the plan is to privatize 50 percent of industry in three Figure 94. Monthly Price Changes in Poland and years. To this end, the plan relies on a free distribution of Yugoslaviat vouchers to the public and the creation of financial in- (percent change over the previous month) termediaries to control the enterprises (see chapter 21, 80 "Markets and Institutions in Large-Scale Privatization: An Approach to Economic and Social Transformation in Eastern Europe" by Roman Frydman and Andrzej Rapac- 60 zynski in this volume). Before moving on to an examination of the initial re- sults, some comments about the implementation of the 40 heterodox programs in the two socialist countries are in order. Incomes policy support to stabilization is com- monly justified on the grounds that it reduces the un- necessary losses in output that are usually associated '/'0 ' with purely orthodox programs. The possible reasons for o these losses have been investigated extensively, includ- ing imperfect information (Lucas 1973), long-term or staggered wage contracts (Fischer 1977 and Taylor - an Jan Jan Jan 1979), backward wage indexation, coordination prob- 1988 1989 1990 1991 lems (Dornbusch and Simonsen 1988) or monopolistic .......... Poland - Yugoslavia price setting (e.g., Rotemberg 1987 and Blanchard and second semester. Therefore it is useful to examine first Fischer 1989). the results for the first semester and then to identify the Although the microfoundations of socialist and mar- main problems that emerged during the second semes- ket economies are clearly different, the case for includ- ter. ing an incomes policy in the stabilization programs of A sharp contraction in economic activity followed im- the former may be even stronger. In addition to the im- plementation of the two programs in both countries, as portance of this policy in breaking the inflationary iner- shown by the deseasonalized indices of industrial pro- tia in the Yugoslav and Polish economies (both countries duction in figure 9.5. The decline in industrial produc- formalized backward indexation rules in the late 1980s), tion during 1990 was much more pronounced in Poland it has a role to play given the absence of markets for both (-20 percent, with a decline in socialized sector produc- labor and capital, and it can counterbalance the influ- tion of about 28 percent and an increase in private sector ence of the workers' councils in wage determination. Under these conditions, the link between aggregate de- Figure 9-5. Industrial Production Indices in Poland mand policies and the short-run behavior of wages and an85osa0 prices is likely to be weaker than in market economies,6 (1985=100) making the case for nominal anchors stronger. 135 Examination of the Initial Results 125 Analysis of the initial results of the two programs re- : veals two well-defined periods. During the first semester 115 t iv*.: 5 both programs achieved a substantial reduction in the 105 rate of inflation, as shown in figure 9.4. In the case of Yu- . V V V goslavia, the monthly inflation rates were reduced to 9 A zero at the end of the semester, while in Poland they V were around 3-5 percent (see also table 9.4 and the two appendices).7 The performance of both programs was more mixed : - . after June. In the case of Poland, monthly inflation re- 75V mained at around 5 percent, proving more persistent than anticipated. In Yugoslavia, there was a revival of in- 6s l I flationary pressures, as indicated by monthly inflation 1986 1987 1988 1989 1990 rates ranging between 3 percent and 8 percent in the .......... P oln t- Yugoslavia 107 Table 9-4. Results of the Polish and Yugoslav Programs During 1990 Variable Poland Yugoslavia Inflation 1989: 640% 1989:2,700% (Dec.-Dec.) 1990:250% 1990: 120% Output growth Industry: -20.0% Industry: -10.5% GDP: -12.5% GDP: -7.5% Rate of 1989: 0% 1989: 14.2% unemployment 1990 (Dec.): 6.1% 1990 (mid-year) 14.9% Trade balance 1989: US$0.2 billion surplus 1989: US$1.5 billion deficit 1990: US$2.2 billion surplus 1990: US$4.7 billion deficit Current balance 1989: US$1.8 billion deficit 1989: US$2.0 billion surplus 1990: US$0.7 billion surplus 1990: US$2.7 billion deficit Reserve position Dec. 1989: US$2.5 billion Dec. 1989: US$6.1 billion June 1990: US$3.9 billion June 1990: US$8.6 billion Dec. 1990: US$6.8 billion Dec. 1990: US$6.7 billion Black market Reduced to zero Reduced to zero in December; premium increased to 20% during the year Monetary Cumulative real changes: Cumulative real changes: aggregates Ml M3 Ml M3 Dec.-June: -15% -42% Dec.-June: 49% -5% Dec.-Dec.: 58% -42% Dec.-Dec.: 40% -20% Share of foreign exchange Dec. 1989: 69% Dec. 1989: 65% deposits in M3 June 1990: 42% June 1990: 51% Dec. 1990: 31% Dec. 1990: 46% Source: Various national sources. production of about 24 percent) than in Yugoslavia (-10 The first reason for this difference in output perfor- percent). Although the post-stabilization recession was mance was the much stronger direct supply shock aris- severe in both countries, it should also be noted that the ing from the increase in input prices in Poland. A second decline in economic activity cannot be blamed entirely likely reason is the tighter monetary stance in the Polish on the stabilization, since it started in the second semes- program in the first semester, even though a first inspec- ter of 1989, after a "peak" of activity in the first half of tion of the two monetary programs suggests otherwise. that year.8 Figure 9.7 illustrates the differences in the quarterly ev- The more severe recession in Poland was also reflect- olution of the real stock of money (Ml and M3) in the ed in the relative behavior of imports and the trade bal- two countries. Between December 1989 and June 1990, ance in the two countries, although in this case a real Ml grew by 50 percent in Yugoslavia and 15 percent number of other factors were also at work, most notably in Poland, as measured by calendar price indices. In the the sharper pre-plan real devaluation in Poland (figure case of M3, the difference is also large. There was a de- 9.3). The immediate import contraction was more severe cline of only 5 percent in Yugoslavia versus 40 percent in in the Polish case, as shown in figure 9.6. In addition, it Poland.10 was not followed by a recovery, unlike the case of Yugo- In the second semester the situation is somewhat dif- slavia. Whereas the Polish trade balance shifted into a ferent. The process of remonetization continues in Po- surplus, in Yugoslavia the trade deficit increased.9 land and is reversed in Yugoslavia. As discussed in more 108 Figure 9-6. Convertible Currency Area Trade Balance, Figure 9-7. Monetary Indices, Poland and Yugoslamia Poland and Yugoslavia 1988-90 (fourth quarter, 1988=100) A. Yugoslavia A. Real Ml Indicesa (illions of US$) 120 6 100 4 0 0~ ~ ~ ~ ~ ~~~~~~~~~~~~8 2 ~~~~~~~~~~~~~~~~~~~~~60 ,, 40 H -2 20 4 1 2 3 4 1 2 3 4 1 2 3 4 1988 i 1989 1 1990 1 I I I I I I i 1Q 1Q 1Q Imports Exports TradeBalance 1988 1989 1990 Poland Yugoslavia B. Poland (billions US$) B. Real M3 Indices a 4 (fourth quarter, 1988=100) 3 2 0~~~~~~~~~~~~~~~~~~6 :b, 0 1. 40 1 2 3 4 1 2 3 4 1 2 3 4, ..... 1 1988 1 1989 1 1990 1 20 ''''''- Imports Exports Trade Balance I I I i 01 1Q 1Q 1Q detail below, in Yugoslavia that pattern essentially re- 1988 1989 1990 flects excessive credit expansion in Yugoslavia in the sec- ..... Poland - Yugoslava ond semester, resulting in a revival of inflation and a loss of foreign reserves. The comparative evaluation of the two monetary pro- a. The number following the year rees to the quarter for that year. grams in the first semester centers on four major points. First, in Yugoslavia, all exchange rate and price correc- sequent impact on prices was also smaller. In Poland, the tions were carried out before the plan and were smaller various prior price and exchange rate adjustments com- in magnitude than those in Poland. Therefore, the sub- bined with the large increase in energy prices in January 109 1990 (400 percent) resulted in a price increase in Janu- reshuffling of the portfolios of asset holders. However, in ary whose magnitude (100 percent from the beginning both countries the exogenously determined discount to the end of the month) was not anticipated by the mon- rates influenced the determination of interest rates con- etary program. siderably, making them poor indicators of liquidity con- Second, as mentioned, the Yugoslav monetary pro- ditions. In Yugoslavia, interest rates on short-term gram was preceded by a large real increase in the stock deposits were relatively stable during the year (10-15 of base money in December 1989 (25 percent), resulting percent per year), in spite of the large fluctuations in from increased domestic credits and some increased for- monetary conditions in this period. In Poland, short- eign exchange inflows in the last two weeks of the year. term deposit rates moved from 15 percent per month in This growth in the pre-plan stock of money was not ob- January and February to 5-6 percent per month in the served in Poland. following months, becoming slightly positive in real Third, the net domestic assets of the central bank of terms. Lending rates were much higher in both coun- Yugoslavia during the first semester behaved according tries, but that situation to a large extent reflects the non- to the program-basically flat in nominal terms. In the performing loans in the banks' portfolios, and not just li- case of Poland, the net domestic assets of the banking quidity factors. Finally, the real devaluation in Poland system did not rise to the targeted level, essentially be- may also account for the stronger initial contraction of cause of the better than expected budgetary perfor- activity relative to Yugoslavia.i mance. There was no attempt to divert credit within the Indeed, the first indications are that the unambigu- ceilings to the enterprises. ously adverse supply-side effects of a real devaluation Fourth, large inflows of foreign exchange resulted in were not offset by the expansionary effects on the de- a substantial increase in net foreign assets in both coun- mand side. In the traded goods sector, it is unclear tries. In the case of Poland, the inflows were generated in whether the increase in foreign demand offset the con- the trade account, while in the case of Yugoslavia they tractionary supply-side effects. In the case of non-traded were generated by increased workers' remittances and goods, it is most likely that the demand-side effects were the repatriation of foreign exchange assets held abroad also contractionary. The wealth effect was largely nega- by enterprises. Although the increase in net foreign as- tive, as indicated by the 40 percent real decrease in M3, sets was even greater in Poland, the other factors, most while some redistribution of real income from wages to notably the price jump in January, produced a much profits is also likely to have depressed demand.13 more severe monetary crunch. A complete comparative evaluation of the initial re- The conclusion that monetary policy was tighter in sults of the stabilization programs requires a more de- Poland based on a comparison of real money stocks may tailed analysis of the impact of policies on real incomes not be valid, however, since it implicitly assumes that the in the two economies. In both countries, there was a real demand for domestic money was equally strong in substitution of explicit taxes for the inflation tax. How- the two countries. The large price shocks at the start of ever, the increase in explicit taxation was smaller than the program in Poland may initially have created adverse the volume of inflationary taxation in 1989, especially in expectations that lowered the demand for money. In any Yugoslavia. When this factor is looked at in isolation, case, the above comparison reveals interesting differenc- there seems to have been an increase in the net real in- es in the design of monetary policy in the two programs. come of households and enterprises. However, real wag- In particular, it shows the desirability of carrying out all es also fell substantially in the two countries (figure 9.8). price corrections before establishing the monetary tar- In part the reason was the removal of the credit subsidies gets. Another interesting point is that in neither country for enterprises. Thus the decline in real wages indirectly did households convert their stocks of foreign exchange reflects the elimination of inflationary financing of en- deposits into domestic currency deposits, despite the terprise losses. frozen exchange rate and higher interest rate on domes- It is difficult to assess the net impact of these factors tic deposits. In the Israeli stabilization program such a on real incomes. In addition, the situation is likely to be conversion constituted an important source of non-in- highly differentiated across households and enterprises. flationary monetary expansion (Liviatan 1988). The ab- While the real wage adjustment may have allowed sever- sence of conversion in Poland and Yugoslavia could al enterprises to operate without a loss (without mean- signal an initial lack of credibility of the program, partic- ing that they became efficient), in other cases there ularly of the exchange rate policy.1' would still have been a loss even with a further substan- It is natural to look at nominal interest rates as an in- tial reduction in the real wage. In fact, during the first dicator of liquidity conditions that result from an imme- semester the wage ceilings became non-binding for sev- diate contraction in the supply of money and the eral enterprises in both countries. The tightening of 110 Figure 9-8. Real Wage Indices, Poland and Yugoslavia Figure 9-9. Dinar Bank Credits and MI (1985=100) (billions of dinars) 200 A. Nominal 250 175 A _ _ __W_ _ _ _ _ 200 150 51 75~~~~~~~ .4 ~50 125F , yyO * S SS. 50 ,,,,,., ...........,, . O. . ......... ......... . Jan Jan Jan Dec 1988 1989 1990 1990 1989 1990 1991 .....10. - -land 0Yugoslavia ......... Dinarcedits Ml credit during the first semester had a strong impact on the enterprise sector. In Yugoslavia, 7,000 enterprises had difficulty making payments to banks and other en- B. Real terprises, 3,000 had accumulated arrears for 30 days and (base=1985) 350 were declared bankrupt. A large number of enter- 250 prises was reported to have interrupted their wage pay- ments (especially in May) to postpone bankruptcy. At the 200 end of the first semester the situation was, however, '. .- S S~~~~~~~~~~~~~~~~~~~~~~~J1 I *~~~~~~~~~~~~~~~~~~~~~J highly differentiated, with a group of enterprises in- ;.._ .. creasing wages above the ceilings and another group un- 150 able to make wage payments within the ceilings. _ Pressure to relax the monetary policy mounted in the 3 100 first semester and led to a relaxation after June. The shift to a relaxed monetary stance was achieved mostly through the implementation of measures to increase the 50 multiplier. These measures produced a rapid increase in the stock of bank credits ai er June (figure 9.9). The eas- 0 Real ing of monetary policy led to further increases in wages 1989 1990 1991 and contributed to a revival of inflation in the second se- mester. It also aborted the expected shake-out of the in- 2......... Dinar5aedits Ml dustrial sector by keeping loss-making enterprises afloat and allowing them to resume thge payments. billion equivalent in Dinars to finance the payment of Concern over the revival of inflation led the central wages and pensions (TheEconomist 1991) in the region, bankurn me to a tight monetary stance in early Octo- without notifying the board of govenors. The monetary ber. However, pressed by enterprises to continue lend- impact of the domestic credit expansion was largely off- ing, commercial banks reduced their reserves below the set by a substantial loss in foreign reserves during the required levels, a move that weakened the central bank's same period. However, these events raised obvious control over monetary policy.o 4 doubts about the sustainability of the stabilization pro- Further evidence of the difficulties faced by the cen- gram. tral bank in conducting monetary policy was the unusu- In Poland, despite the sharp contraction in industrial al episode of December 1990, whereby one regional output, no bankruptcies occurred. Nor did the expected branch of the central bank expanded credits by US$1.8 shake-out of the industrial sector, with inefficient enter- ill prises sinking under the pressure of the elimination of duction of a new wage policy and the adequacy of the fis- subsidies, take place.15 cal support for the restructuring component of the two The Darwinian-Schumpeterian process of natural se- programs. This group of issues also includes the need to lection did not operate. As noted, the main trigger mech- reduce the servicing of external debt in order to avoid a anism for bankruptcy procedures is a failure to pay the drain of scarce resources that ideally should be chan- dividend tax. The appropriateness of this mechanism as nelled toward restructuring the economy. The second the main trigger for restructuring and/or bankruptcy is, group comprises the restructuring and privatization of however, questionable. First, the tax was not burden- enterprises and banks and the development of the labor some (around 6 percent of total profits). Second, many and financial markets. enterprises paid the dividend tax while also failing to Recognition that the real exchange rate in Yugoslavia meet obligations with other enterprises and banks. How- had severely appreciated led to a 30 percent corrective ever, neither creditor firms nor the banks initiated bank- devaluation in January 1, 1991. In the case of Poland, the ruptcy procedures, the implication being some sort of real appreciation is probably not so severe as in the case collusive behavior among firms and banks. The ineffi- of Yugoslavia, given the large real devaluation at the ciencies arising from these inter-linkages among enter- start' of the program. Obviously, if inflation does not prises is one of the main issues that has to be tackled in quickly subside, the extent of the real appreciation is the second stage of the program. likely to become unsustainable during 1991. Both fiscal and credit policies were relaxed after July, Although the exchange rate correction in Yugoslavia in an attempt to stimulate the economy. The relaxation seemed unavoidable, given the rapid loss of foreign re- of credit policies opened the way to a significant increase serves at the end of 1990, it did result in increasing rates in real wages and contributed to a certain revival of in- of inflation at the start of 1991 (5 percent and 10 percent flation. In an attempt to reduce these inflationary pres- in January and February, respectively). This increase in sures, credit policy was tightened in the fourth quarter, inflation happened despite the introduction of a new with the imposition on each commercial bank of strict package of measures designed to contain the impact of ceilings on credit to the socialized sector. However, the the devaluation, which included a freeze on nominal overall credit expansion largely exceeded the targets, wages paid by loss-making enterprises and new regula- mainly because of sharp growth in credit to the non-so- tions designed to prevent other episodes of monetary de- cialized sector, which was exempted from the ceilings. control. Success at preventing the re-emergence of an Despite the tightening of credit to the socialized sector, exchange rate-wage-price spiral will obviously depend wage pressures in this sector remained unabated, a point on more effective implementation of wage and monetary that revealed the difficulty of containing wages under policies, a task that will require a high degree of consen- the present ownership structure. sus among the various republics. Developments in 1990 raise a more general issue: un- Despite the present uncertainty regarding the con- der the present structure of these economies, macroeco- tinuing implementation of these measures, they do re- nomic policies alone cannot generate efficient natural flect a recognition that some form of wage policy is selection. Inter-enterprise credits link "good" and "bad" needed, especially while the labor markets are not suffi- enterprises, opening the way for chains of bankruptcies ciently developed and the problem of ownership rights is involving "good" enterprises. Moreover, if the banking not adequately solved. The design of a wage policy for the system is inefficient and, in the case of Yugoslavia, con- transition is indeed a difficult task. On the one hand it trolled by enterprises, resources are not necessarily has to counteract the excessive influence of the workers' channelled to the more efficient uses. Under these con- councils. On the other hand, it should minimize the in- ditions, the attempt to enforce hard budget constraints efficiencies arising from strict and generalized wage may result in a generalized credit crunch-as seems to controls, which can hinder the recovery of economic ac- have happened in the first semester of 1990 in Poland- tivity. with unclear effects on efficiency. This point is also relevant in the case of Poland, where a wage policy was maintained in the first half of 1991. Assessment of the Main Issues However, the government introduced several changes. First, it shifted the control of wages from the wage bill to The issues that policy-makers in Poland and Yugosla- the wage per worker to eliminate the constraints on ex- via will have to address include some that belong to the panding firms. Second, wage increases became condi- stabilization component per se and others that are of a tional on profitability performance. Finally, private finns structural nature. In the first group, the most important were excluded from the wage ceilings, and corporatized issues are the unfreezing of the exchange rate, the intro- firms were partially exempted from the tax penalties on 112 wage increases above the ceilings. However, the govern- market will require that policy-makers address the is- ment retained the monthly indexation of wages, with ad- sues of housing ownership and financing, two notorious verse effects on inflationary inertia. obstacles to labor mobility in socialist countries. Fiscal support for restructuring is another critical is- While Poland and Yugoslavia acknowledge the impor- sue in the two countries. While the fiscal adjustment has tance of these issues, they have been addresssing them proven sufficient to close the deficit in Poland and to ab- too slowly. Failure to provide prompt solutions to these sorb the central bank's deficit in Yugoslavia, it does not problems could lead to a protracted stagnation of output seem to be consistent with the intention of implement- and a return to high inflation. In this regard, the model ing a serious restructuring program in both countries. of sequencing traditionally applied to Latin American For instance, it has become clear that the fiscal commit- countries, in which structural issues are relegated to lat- ments to the social program in Yugoslavia would have er stages of the adjustment programs, does not seem to been inadequate had the loss-making enterprises really be applicable to reforming socialist countries, in which been forced into bankruptcy. The initial fiscal commit- stabilization and structural reforms are much more ments to the financial restructuring also proved largely closely intertwined. insufficient at the end of the year. In the area of restructuring, a wide range of issues Appendix 1. Stabilization in Poland needs to be considered. At the level of each enterprise, restructuring entails laying off excess personnel, writing Background off debts in justified cases, making selective investments capable of improving the efficiency of existing capital, The Polish stabilization program of January 1990 is changing management and so on. It also entails closing undoubtedly one of the most radical attempts to trans- enterprises that cannot become profitable even with the form a socialist economy into a market economy quick- above measures. At the level of each bank, restructuring ly. In addition to the severity of Poland's macroeconomic involves a detailed evaluation of its portfolio, the remov- situation, the government chose to pursue a "cold tur- al of bad loans from the portfolios, a severing of the links key" approach to stabilization (whence the label "Big between banks and enterprises, and the introduction and Bang" ), because of the disillusionment with the gradu- enforcement of prudential financial regulation and su- alist reform strategy followed in the 1980s. That strate- pervision. A crucial question related to the restructuring gy, which relied on increased enterprise decentralization of enterprises and banks is who will be in charge of this but within a context of continuing state interference in formidable task and under what system of incentives. An the allocation of inputs, licensing of imports and ex- approach to restructuring that is excessively centralized ports, and determination of prices, had not provided vis- and excludes privatization may not only prove too slow ible improvements in efficiency. In fact, the partial to implement, given the institutional deficiencies, but decentralization of enterprise decision-making-remi- may also produce an undesirable selection of enterprises niscent of the idea of "market socialism" -in a context and sectors for restructuring. A more rapid move toward of absent factor markets and persistent soft budget con- large-scale privatization seems to be required to mini- straints actually contributed to a widening of the macro- mize the risks of wasting resources during the restruc- economic imbalances in the 1980s. Several rescheduling turing and to enhance the prospects of a sustained agreements of the external debt partly masked the sever- supply response. (For a detailed discussion of this issue, ity of the external position and allowed Poland to grow at see Hinds 1990 and Lipton and Sachs (1990), as well as relatively high rates in the period 1983-88 and to main- chapter 21, "Markets and Institutions in Large-Scale tain relatively high rates of investment growth (figure 9- Privatization: An Approach to Economic and Social 1-1). Transformation in Eastern Europe" by Roman Frydman On the other hand, the rate of inflation averaged and Andrzej Rapaczynski in this volume). above 50 percent in the 1980s, a condition that prima fa- While forcing inefficient enterprises into bankruptcy cie put Poland in with the "chronic inflation" countries is an important step in the right direction, in the absence (figure 9_1_1).16 of well-functioning labor and capital markets that per- However, the persistent inflation during the 1980s re- mit an efficient reallocation of resources across firms duced significantly the stock monetary overhang, a point and sectors, a large pool of unemployed could be gener- on which Poland's experience differed from the typical ated for a protracted period. The impossibility of devel- experience of the centrally planned economies with rigid oping a true labor market without developing in parallel price controls, such as Bulgaria and the Soviet Union. a market for capital again raises the issue of privatization Accordingly, the underlying inflationary pressures in the (see Hinds 1990). In addition, the development of a labor 1980s came, overall, more from flow than from stock fac- 113 Figure 9-1-1: Selected Macroeconomic Variables, Poland, 1980-90 A. GDP Growth B. Budget Deficit" (percent changes) (percent of gross domestic product) M 10 (%) 4 2 5 0 Average 1983-9=39 -2 0 -4 -5 -10 -10 -12 -15 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990a -14 a. EstimatedL 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 a. General govemnment defidt C. Gross Domestic Product, Consumption and D. Inflation Rate Investment Indices (1980=100)) (yearly percentage change) 120 700 110 ~~~~~~~~~~~~~~~~~~600 100 Soo~~~~~~~~~~~~~~~~~~0 100 ,.,,._ . \ [t400 90 * , \300 200 80 70 ~~~~~~~~~~~~~~~~~~~0 60 -100 60 , . .. . . , . . . . -100 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1971 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 GDP - Consumption --- Investment - Average 71-79 = 3.6% Average 80-89 = 53% E. Exports and Imports (Goods) (bilin US$) (in convertible currendes) b E Current Account 12 4 -2....... . 2. * . 6 -4 -2 ~~~~~~~~~~~~~~~~~~~~~~~~-2 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 ---.-,--Expotts Imports m Trade balance - Current account ....... Noninterest current account 114 tors (Lipton and Sachs 1990). In this regard, cost-push presence of inertia elements in the inflation. These can factors, such as the cost of labor and the prices of im- be traced mainly to the de facto full indexation of wages ports, played a fundamental role in the context of a price- to prices throughout the period 1983-89 and, after 1985, setting system dominated by "cost-plus" behavior. How- to an exchange rate policy aimed at stimulating exports ever, the widespread price controls that prevailed until through continued "competitive" devaluations. the end of 1989 and that affected about 50 percent of to- Discrete shocks to the price level, usually through a tal retail sales complicate the analysis of inflation. Al- change in administered prices, tended to be translated though repressed inflation does not seem to have been a into a persistent higher level of inflation. Thus, the sys- "chronic" phenomenon in Poland in the 1980s, the tem, with its generally accommodating monetary policy, mixed system of controlled and free prices tended to ad- lacked a nominal "anchor." In the mixed system of free just slowly to excess demand and to generate large gaps and controlled prices, price controls proved to be a pure between the controlled and free prices, with adverse ef- substitute for the missing anchor. Indeed, price controls fects on the budget. Thus, an inflation that behaved er- had an adverse effect on the fiscal budget-as they im- ratically, "stop-and-go" phenomena and the inertia plied subsidies-and with passive money this channel elements in the catch-up between controlled and free fueled inflation.18 prices were not only important features of Polish infla- The price controls gave rise to various types of subsi- tion in the 1980s but were particularly important in the dies to cover the difference between the price and pro- acceleration of inflation in 1989.17 duction costs; the subsidies were channelled to both Large disequilibria did indeed precede the "correc- households and enterprises. Moreover, when the input tive" jumps in the price level, mainly in 1982 and 1988- subsidies were reduced through increases in the admin- 89, with relatively stable rates of inflation of 20 percent istered prices, especially those of energy-as, for in- per annum inside the peaks, a rate that was high by Eu- stance, in 1988-sizable subsidies on interest rates ropean standards. cushioned the enterprises from these adverse supply Preliminary econometric results confirm this pattern shocks. Only in 1990 did this compensation disappear, (table 9-1-1). The regression, which tries to separate Table 9-1-1. Poland: Inflation in the 1980s (quarterly regression 1983-89)a DLCPI = 0.72*DLCPI(-1) + 0.42*DLWAGES + 0.1*DLOER + 0.23*DLM3/RS - 0.63*ECM 2 (7.1) (7.1) (7.1) (2.8) (-3.5) R = 0.97; DW = 1.33; e = 0.0042 Note: DLCPI = the rate of inflation; DLWAGES = the rate of change in wages; DLOER = the rate of change in the official exchange rate; DLM3/RS = the change in the ratio of household broad money to retail sales; and ECM = the error correction term, defined as the difference between prices and costs (wages plus the exchange rate). The t-statistics in parentheses a. The equation in the main text is part of a system estimation, in which prices and wages are estimated simultaneously. Source: Commander and Coricelli (1990b). cost-push factors from monetary disequilibria-as mea- with the simultaneous elimination of both types of sub- sured by the change in the ratio of household money to sidies. retail sales-indicates that, overall, cost-push factors In general, the system allowed the enterprise sector were the main determinants of price inflation in the pe- to register net profits in the aggregate without, as before riod 1983-89. 1989, large deficits for the consolidated government sec- However, if the coefficients in the above equation are tor (including transfers for the service of the foreign used, in the period of accelerating inflation in 1988 up to debt). Therefore, a "pure" fiscal explanation of underly- the third quarter of 1989, monetary imbalances exerted ing inflation does not seem to be relevant to the Polish important pressure on inflation. Indeed, 20 percent of experience in the 1980s. Nevertheless, as noted, sizable the increase in prices in that period is explained by the hidden subsidies, mainly in the form of implicit subsi- monetary disequilibrium, while cost-push factors ac- dies for interest rates, were channelled to the enterpris- counted for the remaining 80 percent, with 65 percent es, accounting for a significant portion of the attributable to wages and 15 percent to the exchange seigniorage from money creation. According to some es- rate. timates, the interest rate subsidies amounted to an aver- The fact that inflation remained relatively high fol- age of 6 percent of GDP in the period 1984-88.19 lowing the jump in the price level in 1982 suggests the 115 The subsidies on interest rates thus allowed firms to reduction in the fiscal and monetary disequilibria, pav- operate with positive profits, compensating them for the ing the way for the stabilization program of January 1, high costs of the inefficiency in production, such as 1990. overstaffing and the generally high intensity of inputs in production, common symptoms of soft budget con- The first phase. The wage push initiated in the second straints. The largely negative real interest rates on lend- half of 1988 gained strength at the beginning of 1989, ing were accompanied by even more negative real rates the result being sizable increases in real wages and a real on deposits. As a result, households increasingly shifted appreciation. In addition, the budget deficit deteriorated out of zloty-denominated assets and into foreign curren- sharply during the first half of 1989 because of a substan- cy deposits (table 9-1-2). The base for the inflation tax Table 9-1-2. Poland: Monetary Indicators 1985 1986 1987 1988 1989 )a (percent of GDP) Ml 20.1 19.5 18.5 14.6 10.1 Broad money 44.3 46.2 51.5 56.5 47.5 (percent of broad money) Ml 44.3 40.3 32.0 20.8 21.9 Currency 19.6 16.8 12.0 9.9 11.5 Quasi money 55.7 59.7 68.0 79.2 78.1 Zloty 75.1 68.8 53.7 35.1 30.7 Foreign currency 24.9 31.2 46.3 64.9 69.3 Socialized sector 26.9 24.9 26.6 19.4 25.1 Non-soc. sector 73.1 75.1 73.4 80.6 74.9 a. Average stock, geometric mean. Source: National Bank of Poland; Intemational Monetary Fund, Intemational Financial Statistics. was progressively reduced in the 1980s as the currency tial decline in revenues and, to a lesser extent, increased substitution advanced, a pattern that increased velocity expenditures.20 in terms of Ml and reinforced the inflationary pres- The consolidated budget of the general government sures-open or repressed-arising from the need to fi- moved from an initial balance in 1988 to a deficit of 7.2 nance both the budget deficit and the hidden subsidies percent of GDP in 1989. It should be noted that 80 per- to the socialized sector. cent of the state's budget deficit (excluding extra-bud- During 1987-88, and then again in 1989, the author- getary funds) in 1989 was accumulated in the first half of ities attempted to impose more stringent financial disci- the year. The deficit was financed mainly through central pline on the enterprises by tightening bank credit. bank credits and money creation. However, given the predominantly soft budget con- Interestingly, the worsening of the "fundamentals" in straints, the attempts to enforce financial discipline the first half of 1989 did not immediately result in hyper- through credit ceilings were largely ineffective, with en- inflation. In the first half of 1989 the monthly rate of in- terprises bypassing the ceilings by extending credit to flation averaged 7 percent and did not accelerate during each other. The ratio of inter-firm credit to bank credit the period. The cumulative increase in prices was about to the socialized sector went from about 50 percent at 60 percent with respect to December 1989. This lack of the end of 1987 to 70 percent at the end of 1988 and 155 synchronization between the acceleration of inflation percent at the end of 1989. and worsening fundamentals is consistent with the stop- and-go nature of the system, with its widespread price Preparation for the Stabilization Program and Other cntro nawhichta widening of monetary alance Developments in 1989 controls, in which a widening of monetary imbalances generally precedes corrective price increases. That grow- Two clearly distinct phases, separated by the change ing monetary imbalances were created in the first half of in government, characterized 1989. In the first, which 1989 can be inferred from the rapidly depreciating ex- corresponds to the first three quarters, there was a sharp change rate in the parallel market, a signal of a loss in deterioration in the macroeconomic situation. In the confidence in the domestic currency. The share of for- second, corresponding to the last quarter of 1989, when eign currency deposits increased from 65 percent at the the new government was in power, there was a marked end of 1988 to 71 percent at the end of June 1989. 116 The outburst in the rate of inflation occurred in Au- firm credit market, which partly cushioned the enter- gust, when retail prices jumped to 39.5 percent from 9.5 prise sector from the squeeze in official credit. Inter-firm percent in July. The liberalization of food prices in Au- credit declined by only 9 percent in real terms during gust at a time of significant excess demand caused a 1989, compared with a decline of 65 percent in real jump in those prices. Wage earners tried to catch up to terms in bank credit to the non-government sector. The the increased food prices by demanding higher wages, in result was a dramatic increase in the size of inter-firm effect indexing wages de facto to food prices, which were credit relative to official credit, from 63 percent at the rising more than the average price level. This phenome- end of 1988 to 145 percent at the end of 1989. non may have contributed to further inflationary pres- To sum up, the deterioration in the macroeconomic sures in the non-food sector as well. Formal institution situation was most severe in the first half of 1989. The rap- of a wage indexation rule in April and the policy of ad- id increase in the rate of inflation to almost hyper-infla- justing the official exchange rate to catch up with the tionary levels in the second half of 1989 was the result of black market exchange rate further increased the infla- a combination of factors, the foremost being: the liberal- tionary inertia. The official exchange rate, which had de- ization of food prices; large adjustments in administered preciated roughly in line with inflation in the first seven prices, particularly energy; continuous devaluations of months (about 65 percent), was depreciated by 680 per- the exchange rate; and the formalization of a backward cent in the remaining months of the year, when the cu- wage indexation scheme with progressively shorter peri- mulative increase in inflation was 320 percent. odicity for adjustments. Accordingly, wage indexation and the exchange rate policy tumed the "corrective" increases The second phase. The new government, which took in the price level associated with the reduction in the sub- office in September, tried promptly to halt the deteriora- sidies for food and energy prices (August and October) tion in the macroeconomic situation, instituting several into sustained higher inflation. Moreover, anticipation of measures to tighten credit policy, reduce the budget def- the January 1990 program is likely to have affected devel- icit, curb the wage increases and correct the prices of en- opments at the end of 1989 and to have exerted further ergy and other highly subsidized goods. The adjustments pressure on inflation. Significant purchases and hoarding in the prices of energy produced an initial acceleration by households were detected, along with the accumula- in inflation in October. A reduction of the periodicity of tion of inventories by enterprises. Finally, the wage spike wage adjustments from quarterly to monthly and the in December 1989 reflects the anticipation of the price frequent exchange rate devaluations also fueled infla- corrections carried out in January 1990, since, through a tion.21 complicated indexation rule, the December increase was In December real wages were 27 percent above their linked to both price changes in November and expected level in October, while the official exchange rate depreci- changes in January. ated by 260 percent in the period October-December. After October, however, the improvement in the fun- The Stabilization Program of January 1990 damentals did exert downward pressure on inflation, which decelerated sharply from 54.8 percent to 17.7 per- The stabilization program, launched on January 1, cent in December. At the same time, the premium of the can be defined as heterodox, with two nominal anchors, parallel market over the official exchange rate was sub- the nominal wage and the exchange rate, and fiscal and stantially reduced by the end of 1989, an indication of a monetary tightening. The wage policy consisted of a significant improvement in the monetary disequilibria. lagged indexation of the wage bill with low coeffi- Three main factors explain this reduction of the premi- cients.22 um: the devaluation of the official exchange rate; the The wage ceilings were to be enforced through a slowdown in the rate of growth of the money supply, steeply progressive tax penalty. The exchange rate was linked to the improvement in the budget deficit; and the frozen at 9,500 zlotys per US dollar, after the unification reduction in the monetary overhang as a result of the of the parallel and official markets, and the decreed "in- large cumulative increase in prices during 1989. ternal" convertibility of the zloty (for current account As to currency substitution, the shift from domestic operations but not for capital account operations). The to foreign currency deposits was halted in the second freeze of the exchange rate was preceded by a sharp de- half of the year, and the share of foreign currency depos- preciation of 46 percent, and the measure resulted in a its in broad money declined from 71 percent in June to significant overshooting of the parallel market rate pre- 69 percent at the end of December. vailing on average in December (about 30 percent).23 The tightening of bank credit to enterprises was par- It was expected that the exchange rate would be de- tially compensated for by the flourishing of a thick inter- fended by a special fund (of US$1 billion) to be made 117 available by foreign banks, as well as by the interest rate value of the funding capital of a firm) was established as policy. the main instrument for triggering these procedures. The fiscal components relied on a balanced budget, to To summarize, the design of the program was based be achieved in 1990. An increase in revenues of about 4 on the assumption that the nominal anchors would help percent of GDP and a small reduction in expenditures of reduce inflation very fast, and, accordingly, monetary around 1 percent of GDP were to produce an adjustment balances and credit, after an initial tight condition in the budget. The government undertook discretionary meant to defend the nominal anchors, were expected to measures-which were to have supplemented the auto- grow considerably in real terms starting in the second matic rise in real revenues from the expected reversed quarter of 1990. The planned decline in real terms in the Olivera-Tanzi effect-to increase revenues: it raised the first quarter was based on the assumption that the basic rate of the turnover tax from 15 percent to 20 per- growth of nominal credit and monetary aggregates cent; it revalued tenfold the fixed assets of the socialized would lag behind the rate of inflation, pushed up tempo- enterprises, the basis for the dividend tax; and it drasti- rarily by the increase in energy prices. However, it was cally curtailed tax exemptions and relief. On the expen- thought that the decline in real wages would give room diture side, the main gains were expected to come from to enterprises to absorb the price shock (see also Lipton a reduction in the subsidies on the order of 8 percent of and Sachs 1990). That the program was not intended to GDP. squeeze aggregate demand can be inferred from the ex- The monetary components relied on tight credit con- pectation of a current account deficit of more than 7 per- ditions in the first quarter of 1990 that were partially cent of GDP in 1990 (compared with a deficit of 2.6 loosened in the following three quarters of the year. Net percent of GDP in 1989). domestic assets were expected to grow by about 20 per- cent in nominal terms in the first quarter, a level that Initial Results implies a real decline of 30 percent, and to grow at an av- The initial results of the stabilization program can be erage quarterly rate of 7-8 percent in the rest of the year, characterized as stagflationary, at least in the first two a level that is above the expected rate of inflation. This months following the start of the stabilization measures. policy would have guaranteed a small real increase in net A sharp drop in output was accompanied by an accelera- domestic assets by the end of the year. The credit ceilings tion in the rate of inflation, which remained persistent were complemented by an interest rate policy geared to and relatively high given the depressed economy. Using maintain positive real rates throughout the year. Given a terminology suggested by Polish economists (Kolodko expected inflation of about 30-35 percent, the govern- and McMahon 1987), the economy experienced a shift ment set the refinancing rate of the National Bank of Po- from a condition of shortage-flation-caused by the land, which was to serve as a sort of leading rate, at 36 combination of high inflation and large shortages of percent in January. Amrcendments touay the bankinglawstrengthenedthein- goods-to one of stagflation. Amendments to the banking law strengthened the in- The rate of inflation, after jumping to 78.6 percent in dependence of the central bank. This measure likely sig- January 1990, declined to 23.9 percent in February and nalled to actors in the economy that the central bank to an average monthly rate of 5 percent in March-June. was abandoning its role as lender of last resort. After falling to 1.8 percent in August, it bounced back to Fundamental liberalization measures accompanied 4.7 percent in September and 5.7 percent in October, 4.9 the above macroeconomic policies. The price system was in November and 5.9 in December, pushed by the large liberalized almost entirely-only 5 percent of goods sold increases in the prices of oil products and administered at the retail level remain subject to price controls-and prices. Notwithstanding the external negative shock and simultaneously the administered prices of energy prod- successive increases in administered prices, the persis- ucts were raised more than 400 percent so as to reduce tently high rate of inflation, which oscillated around 4 the subsidies. The trade system was liberalized by abol- percent per month for more than 10 months, was unex- ishing the quantity controls on imports and replacing pected. Given the larger than anticipated fiscal adjust- them with tariffs and by reducing the quotas for exports ment and drop in output, this persistence was puzzling, of basic commodities. although it is similar in some ways to Brazil and Argen- Finally, the government announced a program of tina's experience with stabilization programs. When fis- privatization and restructuring of the industrial sector, cal and monetary explanations for this phenomenon are including new rules establishing triggers for the initia- ruled out, the main factors can be identified as: (1) the tion of bankruptcy and restructuring of enterprises. In slow movement toward a higher equilibrium price level this regard, failure to pay the dividend tax (a tax on the for both tradable and non-tradable goods; and (2) the de 118 facto high degree of indexation of wages to prices since gross reserves fully reflected the improvement in the March 1990. current account. The profile of the decline in output shows that the On the monetary side, net domestic assets grew well drop was concentrated at the beginning of the program below the ceilings of the program in the first half of the and that thereafter output remained practically flat, with year, while the growth of the stock of money was in line some signs of a possible recovery surfacing in August- with nominal targets because of the monetization of for- September. Along with the decline in output sold, initial- eign reserves (table 9-1-3). ly enterprises drew down significant quantities of inven- This trend was reversed in the second half of the year, tories of both inputs and finished products. Employment however, and the net domestic assets of the banking sec- in the socialized sector responded with a lag to the drop tor (NDA) overshot the credit ceilings of the program. in output, and in the year as a whole employment was While the contemporaneous increase in real wages and about 10 percent below its level in the same period in real credit raises concerns about possible excess liquid- 1989. Unemployment, practically non-existent before ity in the system, the evidence for the third quarter, par- 1990, increased after February 1990 at an average of ticularly the continued increase in foreign reserves, 100,000 people per month, reaching about 1.15 million points to a "remonetization" of the economy.24 people in December (more than 6 percent of the labor It is worth noting that part of the increase in bank force). "Statistical" real wages declined sharply in the credit appears to have resulted from the substitution of first six months of 1990, while they increased signifi- inter-firm credit with bank credit, a positive phenome- fthirdf 1990, quarter(aout0pecent theyincreasedsignifi- non. Developments in the fourth quarter were more cantly in the and quarter . Dente the worrisome, however, as the financing needs of the gov- second quarter) and fourth quarter. Despite this in- ernment sector increased with the shift to a budget def- crease, for the year as a whole they declined by about 30 icit. percent. The real decline in the stocks of money and credit has Marked improvements took place in both the fiscal been substantial. In the first quarter, net domestic assets and the external accounts. The fiscal accounts moved and broad money declined by 54 percent and 44 percent into a sizable surplus in the first half of 1990 (about 3 in real terms, respectively. In the second quarter net do- percent of GDP was expected for the whole year), despite mestic assets continued to decline in real terms, while the acceleration in inflation in the first quarter and the broad money increased in real terms because of the large decline in output. The improvement continued in the inflows of international reserves. third quarter, although a deficit was experienced in the The National Bank refinancing rate was set according last quarter of 1990. The main factor behind this unex- to expected inflation, the aim being to produce slightly pected surplus was a surge in tax revenues, mainly on positive real rates. In January, the refinancing rate was enterprise profits, which at the end of 1989 (which is the 36 percent, in February 20 percent, in March 10 percent, tax base for 1990 payments) were much larger than an- in April 8 percent, in May 5.5 percent and in June 4 per- ticipated as a result of large capital gains on enterprise cent. After declining to 2.5 percent in July-September, it dollar deposits. The trade balance displayed a remarkable was raised to 3.5 percent in October and again to 4.5 per- improvement, with a surplus of US$2.7 billion (for con- cent in November, following the increase in the rate of vertible currency trade) for the year as a whole, with ex- inflation. Ex post, the rates were negative in real terms ports in convertible currencies increasing by more than in January-February and roughly in line with inflation in 40 percent and imports by 10 percent. The change in the following months. Table 9-1-3. Poland: Monetary Survey, December 1989-September 1990 1989.1V 1990.1 1990.11 1990.11I 1990.1V Gross reserves 24.1 26.8 36.7 50.4 46.8 Domestic credit 40.0 50.8 63.4 81.8 110.4 Credit to government (net) 6.5 2.7 -11.1 -18.1 -8.2 Credit to non-gov't. 33.5 48.1 74.5 99.9 118.6 Other 31.8 35.3 36.5 35.0 34.8 Broad money 96.0 112.9 136.6 169.2 192.0 Inter-firm credit 47.3 75.1 82.6 77.3 82.6a Note: The roman numerals following the years refer to the quarters of the year. a. November. Source: National Bank of Poland. 119 The magnitude of the initial results was largely unex- that in 1989, at least in the first three quarters, the econ- pected; the discrepancy is so large that the normal bias, omy experienced excess demand, with worsening short- which often seems to characterize predictions about sta- ages. Therefore, when real wages are measured using bilization programs, whereby inflation tends to be un- official price indices, they reflect-in the terminology of derestimated and growth overestimated, cannot explain Polish economists-"statistical" rather than actual real it.25 wages (see also Lipton and Sachs 1990). These "statisti- Before turning to possible interpretations of the ini- cal" real wages largely overestimate real purchasing tial results, it is important to emphasize that there were power in 1989 and thus overestimate the drop in real no slippages in the implementation of the program dur- purchasing power in 1990, as measured by the decline in ing the first semester. More controversial is the issue of the "statistical" real wages. slippages in the second half in the areas of both credit Second, measured retail prices likely overestimate and wage policies. However, given that in both areas de- the actual rates of inflation, as a significant proportion of velopments in the first semester were well below the tar- retail trade in 1990 took place outside the state distribu- gets of the program, it is not too surprising that some tion channels, at prices estimated to be 20-30 percent overshooting took place in the second semester. below officially recorded prices. Obviously, the real issue is whether these trends are Third, the downturn in production actually began in sustainable in future months and consistent with a de- 1989, specifically, in the second half, when "statistical" cline in inflation. In the first semester the two nominal real wages were still well above their 1988 levels. anchors of the program, the exchange rate and the nom- Fourth, although a decline in domestic demand was inal wage, were maintained without trouble and actually clearly envisioned in the program, an increase in foreign appeared to be non-binding as the increase in wages demand should have sustained output. stood below the ceilings and the exchange rate stabiliza- Finally, the behavior of the financial savings of house- tion fund made available by western countries was not holds appears at odds with the view of a household sector used, with a build-up of international reserves. As to the experiencing severely constrained liquidity forced to ad- monetary program, the ceilings (on both net credit to the government and net domestic assets) were respected Just Its demand for goods downward. In particular, the ithe g nmplem ins and ineteoestic rasetes were respected aavailability of a large stock of foreign currency deposits with ample margins, and interest rates were set as pro- made by the household sector-which is an abstraction grammed. Although some overshooting of the targets from distributional issues, which may be important- took place in the third quarter, even then no pressure m thatiolds whichnmaine imporrow- was exerted on the exchange rate, and the wage increases means that households were not constrained in borrow- were consistent with the original indexation scheme, ing at the outset of the program for the most part. Given given that the increases derived from use of the "reserve" the temporary character of the income policy scheme, it accumulated in the first quarter, when wages rose well remains to be explained why households did not deplete below the norm. their stock of foreign currency deposits to protect their Assessment of the most recent trends and the policy levels of consumption. Indeed, in US dollar terms, prescriptions for the next months depend crucially on household deposits in foreign currency increased in the the analysis of the results of the program so far, in par- first six months following the start of the program (table ticular of the unexpected outcomes, such as the sharp 9-1-4). recession, the fiscal improvement and the sizable cur- More generally, a main limitation of the above view is rent account surplus. One possible, and actually wide- that it takes as exogenous elements that are actually en- spread, reading of the initial results of the program dogenous, particularly the decline in real wages, which interprets the recession as a standard Keynesian phe- is assumed to be the ultimate source of the decline in nomenon of demand contraction. For Poland, the ap- output.26 pearance of demand barriers would certainly represent a An alternative explanation that is favored here em- fundamental break with the past, when the economy was phasizes that the focus of the stagflationary effects of the traditionally supply constrained. The large drop in real program should be on the enterprise sector and its reac- wages would thus be the main source of the contraction tion to the large supply shock arising from the elimina- in output, and the larger than expected decline in output tion of the subsidies for both input prices and interest would seem consistent with the larger than expected fall rates. According to this view, the credit crunch associat- in real wages. ed with both the policy-determined tightening of official Without denying the possibility that this drop in de- credit and the sudden increase in credit risk played the mand could have been the source of the drop in output, main role in transmitting the effects of the initial supply several doubts can be raised. First, it is widely recognized shock. 120 Table 9-1-4. Poland's Foreign Currency Deposits (billions of US$) 1988 1989 1990 IV I II 111 Iv I II I1 Total 5.4 5.8 6.8 7.2 7.2 6.0 5.9 6.2 Households 4.3 4.5 4.5 4.8 4.9 5.0 5.2 5.6 Enterprises 1.1 1.3 2.3 2.4 2.3 1.0 0.7 0.6 Source: National Bank of Poland. While it has generally been observed that the jump in to accord with several phenomena characterizing the the rate of inflation can reflect the effects of the price lib- first results of the stabilization program. However, it eralization in the context of the liquidity overhang, the would imply a simple shift from a distorted equilibrium larger than expected increase in the rate of inflation in to an undistorted and efficient equilibrium. At the micro January can be mainly ascribed to cost-push factors level, this explanation suggests that firms operating only linked to the devaluation of the exchange rate and the thanks to subsidies should have gone bankrupt after the lifting of the subsidies decreed at the outset of the pro- elimination of these subsidies, while efficient firms gram as a main element in the rationalization of the should have survived after the Big Bang. However, the price system. While the importance of these targets to first months following the stabilization display a differ- achieving greater efficiency in the economy and to re- ent picture that indicates a widespread recession with- ducing pressures on the budget is not questioned here, out significant bankruptcies. This phenomenon casts it is important to stress that the recessionary impact of a doubt on the working of a Schumpeterian, efficient nat- tightening of credit when there is a large increase in the ural selection. prices of inputs has been overlooked. In Calvo and Cori- The explanation postulated here, without denying the celli (1990), it is estimated, although by an admittedly relevance of the above static factors, relies on a different rough calculation, that the credit crunch, measured as mechanism. Specifically, it emphasizes the role of the the gap between actual and required liquidity, necessary credit markets in transmitting the effects of the supply to permit enterprises to maintain the same level of pro- shock. The sharp reduction in the stock of real credit duction of 1989 was on the order of 60 percent at the be- from both the banking system and within the enterprise ginning of 1990. To appreciate the magnitude of the sector itself accounts for the observed recession.27 crunch, note that in January 1990 working capital credit By increasing the price of inputs, the stock of liquidi- from the banking system to the socialized sector de- ty necessary at the beginning of the period to operate at clined in nominal terms-in a context of increased pric- the old level of production increased sharply. The lack of es for domestic inputs of 400 percent, large increases in a sufficient stock of liquidity likely induced firms to re- the prices of foreign inputs as a result of the devaluation duce their purchases of inputs, with a consequent fall in at the outset of the program, and of a jump in interest output. This phenomenon, which is related to the ade- rate costs. quacy of the initial stock of liquidity, occurs indepen- Although the squeeze somehow loosened in the fol- dently of the behavior of the price of outputs. The lowing months, by the end of June 1990 the stock of increase in prices, accompanied by the compression in credit to the non-government sector was 20 percent be- real wages, helped firms reconstitute their liquidity. Ac- low its level in real terms at the end of 1989. This partial cordingly, both the fall in real wages and the good profit loosening of the credit squeeze seems consistent with performance of Polish industry during the first months the fact that, after dropping sharply in January, industri- of the deep recession can be explained. The good profit al output remained flat in the following months. A simi- performance of enterprises in particular is puzzling for lar argument applies to the third quarter, which shows those who share the view that a recession is induced by some recovery in economic activity. At the end of the "demand barriers." third quarter, real credit to the non-government sector The actual working of the credit crunch and the rela- had essentially recovered its level of December 1989. In tive role of the quantity of credit versus its cost (interest the fourth quarter credit to the socialized sector de- rate) are still uncertain, given the presence of direct con- clined in real terms while production remained flat. trols over bank credit. In a static framework, the sharp reduction in subsi- In this connection, the behavior of the inter-firm dies for enterprises represents a supply shock that even credit market may provide important clues. As noted, in in a competitive industry would cause a reduction in 1989 the stock of inter-firm credit, despite a slight de- output and an increase in prices. This simple story seems cline in real terms, grew much faster than official bank 121 credit did. According to data up to March 1990, although might have been the shift in the distribution of income smaller than the decline in official credit, the stock of in- in favor of profits. ter-firm credit declined by about 33 percent in real terms in the first quarter of 1990 (compared with a decline in Appendix 2. Stabilization in Yugoslavia bank credit of 46 percent in real terms in the same peri- od). It continued to decline in real terms in the second Background quarter and also in relation to bank credit. Starting in July it even declined in nominal terms. In the remaining The reversal of external financing flows during the months of 1990, the inter-firm market did not provide a 1980s forced Yugoslavia to undertake drastic measures relevant cushion to the squeeze of official credit. More- to balance its external accounts. The external adjust- over, it is conjectured here that a large component of in- ment consisted primarily of large real exchange rate de- ter-firm credit in 1990 represents an accumulation of valuations and, initially, quantitative restrictions on arrears rather than a "voluntary" extension of new cred- imports. Although these measures were able to generate it. This phenomenon is consistent with the view of a sud- increasing current account surpluses, economic activity den increase in credit risk, likely associated with the stagnated, and inflation accelerated almost continuously signaled withdrawal of the central bank from its role as (figure 9-2-1). Moreover, the absence of fiscal imbalances lender of last resort.28 in the economy might suggest that the nature of infla- The extremely simplified mechanism just described tion in Yugoslavia is entirely non-fiscal. has to be amended to accommodate the impressive im- While it is true that many hyper-inflationary episodes provement in the trade balance observed after January are triggered by balance-of-payments difficulties and 1990. Indeed, in general recessions led by temporary large exchange rate devaluations, the complete absence supply shocks should result in a worsening of the trade of fiscal imbalances from the overall picture in the case balance (consumption tends to decline less than out- of Yugoslavia is intriguing. Indeed, even the "balance-of- put).29 payments view of inflation" recognizes the role of fiscal However, if the behavior of inventories held by enter- deficits in the determination of inflation, although that prises is considered as well, the "model" used here can role is assumed to be less central than under the "fiscal account for a temporary improvement in the trade bal- view of inflation." In the former view, the exchange rate ance. The liquidity crunch will induce enterprises to de- has the central role, while fiscal deficits contribute to in- plete their stock of inventories to reconstitute their flation mostly through endogenous interactions with liquid balances-in other words, there will be a shift in the exchange rate and the inflation rate itself (see for in- the "portfolio" of enterprises from inventories to money. stance, Dornbusch 1987, Dornbusch and Fischer 1986, As a result, total absorption in the economy by house- Liviatan and Piterman 1986, and Montiel 1989). holds and enterprises can decline more than output-al- This appendix will show that, while non-fiscal factors though the decline in consumption falls short of the played a role in the acceleration of inflation during the decline in output-and thereby yields an improvement 1980s, especially at the end of the decade, the existence in the trade balance. The data on inventories are, howev- of fundamental domestic imbalances also contributed er, ambiguous, as inventory accounting is unclear. Ac- decisively to the inflationary process. These fundamental cording to the authors' calculations, the stock of imbalances were originally located in large segments of inventories declined sharply at the beginning of 1990 the enterprise sector, but gradually spilled over to the fi- and was, at the end of September, still well below its level nancial system, resulting, among other things, in a qua- in 1989. si-fiscal deficit in the central bank.30 Demand-side factors certainly played a role. House- A major non-fiscal factor that contributed to higher hold savings increased during the year so that no con- inflation was the need to reduce real wages and ensure a sumption "smoothing" took place. This increase given real devaluation of the exchange rate-the Pazos- occurred as households likely tried to reconstitute their Simonsen mechanism (see Dornbusch 1987; Dornbusch real stock of financial wealth, eroded by high inflation and Simonsen 1988; Pazos 1978; and Simonsen 1989). (see chapter 11, "The Ownership-Control Structure and This mechanism became more important after the intro- the Behavior of Polish Enterprises During the 1990 Re- duction of wage indexation in 1987. In particular, the forms: Macroeconomic Measures and Microeconomic large real devaluation of mid-1988 (figure 9-2-2) in the Responses" by Roman Frydman and Stanislaw Wellisz in context of wage indexation and monetary accommoda- this volume). In addition, the increase in unemployment tion, triggered a significant increase in the rate of infla- could have induced an increase in precautionary savings. tion (figure 9-2-3). During the last three quarters of Finally, a further element (mentioned in the main text) 1989, the relation between real wages and inflation 122 Figure 9-2-1. Selected Macroeconomic Variables for Yugoslavia, 1977-89 A. Exports and Imports, 1977-89 D. Inflation Rate 18 (billions of US$) Scale (Consumer price index: yearly change in percent) 18 ~~~~~~~~~~~~~~~~~~~~10,000% 16 4 ffi '""' ...................... """' ...........1,000%J 14 12 ................... * ~~~~~~~~~~~~~~~100% 10 8 10% 4. ... ...... 1% 1977 78 79 80 81 82 83 84 85 86 87 88 89 77 78 1 79 80 1 81 82 83 84 85 86 87 88 89 †....... Imports - Exports E. Public Sector Surplus, 1977-89 B. Real Exchange Rate, 1977-89 (% of gross social product) (1980=100) 170 1.5 160 1 150 140 0.5 130 90~~~~~~~~~~~~~~~~~~~~~~ 120 110 - 0.5 100 90 80 -1.5 1977 78 79 80 81 82 83 84 85 86 87 88 89 1977 78 79 80 81 82 83 84 85 86 87 88 89 C Current Account, 1977-89 E Real Growth in Gross Social P oduct, 1977-89 (billions of US$) (yearly change in percent) 6 ~~~~~~~~~~~~~~~~~~~~10 8 2~~ ~, ..... 6\ 4~~~~~~~~~~~~~~~~~~~ 2 -2 0~~~~~~~~~~~~~~~~~~~~ 2~~ -2 ~ ~ ~ ~ ~ ~ ~ ~~2.... 1977 78 79 80 81 82 83 84 85 86 87 88 89 1977 78 79 80 81 82 83 84 85 86 87 88 89 .. Non-interest CA - Total CA 123 Figure 9-2-2. Real Exchange Rate and Real Wages (1980=100) 180 170 160 140 Real Wage Rate 130 120 100 70 -~~~~~~~~~~~ ;!~~~~~~ so *, *I,S ,, S,F, . 90 * 2 .* * s 8 * .19 r 1977ed 1978n 1979 period 1981 demand 198e 1984e 198e Helpma 1987 1988erman 1990 fo1amde 9o1nfa Figy expectationthl Inlaio a ndtrewg freee, Readin Woage 40 tinbae8o7ea agncese) 1240 70~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~. - 0.6 -0.2 n Inflation~ ~ ~ ~ ~ ~~~~~~~~~-. ~100 1 90 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0.0 60 - . 1987 1988 1989 changed. During this period, wage demands were driven (see Helpman and Leiderman 1990 for a model of infla- by expectations of a future wage freeze, leading to a 40 tion based on real wage increases). percent increase in real wages and a dramatic accelera- As to the fundamentals, the imbalances in the Yugo- tion in inflation, as shown in figures 9-2-2 and 9-2-3. slav economy originated, as noted, in large segments of 124 the enterprise sector. Already in the late 1970s a signifi- the system. This redistribution gained momentum in cant number of enterprises were recording losses. Dur- the 1980s as a result of the reversal of financing flows ing the 1980s the amount of the losses increased and the successive exchange rate shocks in the economy. significantly (table 9-2-1). Although these numbers are The phenomenon of accelerating inflation in Yugoslavia Table 9-2-1. Enterprise Losses (% of GSP) 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Enterprise losses 1.6 1.3 2.2 2.9 2.1 2.8 3.0 6.6 5.7 15.0 Seigniorage on base money 1.8 2.7 2.7 1.6 3.9 3.6 4.2 4.9 5.7 12.0 Source: SDK, National Bank of Yugoslavia, Knight (1984) and Rocha (1991). affected by many accounting problems, they indicate a can also be interpreted as resulting from an external-in- deteriorating trend that is highly probable. One likely temal transfer problem, as in other debtor countries cause for this increase in losses was the sharp real deval- (Cohen 1988). The ultimate beneficiaries of the internal uations undertaken in the mid-1980s (enterprises ini- redistribution were the enterprises, which were the ma- tially held 80 percent of Yugoslavia's external debt). jor recipients of Dinar credits at subsidized terms, and Another likely cause was the growing numbers of redun- the holders of foreign exchange deposits. dant workers in the enterprises, estimated to have The precise magnitude of the inflationary financing of reached 20 percent of the labor force in 1988.31 hidden losses is difficult to assess. On the one hand, not An obvious question is how the enterprise losses were all the seignorage revenues on base money (changes in financed. During the 1980s there were two basic sources base money as shares of gross social product [GSP]) of financing. The first comprised various forms of inter- shown in table 9-2-1 were channeled toward the financ- enterprise financing, including voluntary transfers from ing of hidden losses. For instance, in 1988 and 1989 profit-makers to loss-makers inside the same industrial roughly half the seignorage revenues on base money holding, as well as inter-enterprise credits and arrears. were absorbed by the large build-up of foreign reserves The second consisted of bank credits on subsidized that preceded the stabilization program. On the other terms. The banks, which were controlled by the enter- hand, the base of inflationary taxation was broader than prises, were able to continue this policy of credit subsi- base money, since it included the whole Dinar deposit dies until almost the end of the decade only by paying base. The introduction of indexation on time deposits even more negative real interest rates on domestic de- above three months in mid-1988 reduced the base of in- posits. flationary taxation. However, the implicit taxation of During the 1980s enterprises' financial imbalances shorter term deposits remained substantial.33 spilled over into the central bank. Several enterprises re- The combination of accelerating inflation and nega- vealed themselves unable to service their foreign liabili- tive real interest rates on deposits resulted in a drastic ties. The central bank absorbed the foreign liabilities of portfolio shift out of domestic assets, as illustrated in fig- enterprises located in less developed regions, thus add- ure 9-2-4. The share of foreign exchange deposits in ing to its own foreign liabilities. The central bank also broad money increased from less than 30 percent at the absorbed the stock of foreign exchange deposits in the start of the decade to more than 65 percent in 1989. It is commercial banks, a measure designed to protect the quite possible that the existence of foreign exchange de- banks from foreign exchange losses.2 posits facilitated the shift out of domestic money and The reluctance of the central bank to charge positive thereby also contributed to inflation by accelerating the real interest rates on its credits, while facing mounting increase in velocity. payments on its foreign liabilities, resulted in a quasi-fis- The failure to correct the internal imbalances was a cal deficit that constituted an independent source of major cause of the failure of the several stabilization at- monetary expansion. tempts of the mid-1980s, which relied mostly on wage- Ultimately, inflation in Yugoslavia engineered a com- price controls. Particularly noteworthy was the stabiliza- plex redistribution of real resources from holders of do- tion program of mid-1988, which attempted to curb in- mestic assets toward the financing of hidden losses in flation essentially through the establishment of 125 Figure 9-2-4. Real M2, Foreign Exchange Deposits and M3 (base=1985) 8 7 M3 t. 2 ,,,_#-- -^,^8 Aw_"\.,- vi, _._,_,r s......j.~ Foreign exchange deposits .......... s-, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I I I I 1981 1982 1983 1984 1985 1986 1987 1988 1989 progressively declining targets on the growth of money Developments in 1989 and Preparations for and wages and through the indexation of time deposits. the 1990 Program This measure was introduced as an attempt to reduce the of crdt an to impose financial disci- The failure of the 1988 stabilization program showed subsidization ofcreditsandtoimposefinancialdis the futility of implementing another program without pline on enterprises. No fiscal policy support to stabiliza- addressing the fundamental domestic imbalances. Thus, tion was envisaged in the 1988 program. during 1989, consensus was reached about the need to The inconsistencies of the mid-1988 program were generate a surplus in the non-financial public sector in aggravated by a real devaluation whose rationale was un- order to cover losses elsewhere in the economy, even clear, since the country was already running a large cur- though there was less certainty about the required mag- rent account surplus and was also engaged in debt nitude of the fiscal adjustment, as well as the best strat- rescheduling negotiations with private foreign banks. In egy for dealing with the loss-makers. fact, the real devaluation defeated one of the main pur- In the case of the central bank's own deficit, the solu- poses of the external debt rescheduling, which was to re- tion was clear, consisting basically of the transfer of the lieve the pressure of extemal debt payments on the servicing of its foreign exchar-ge liabilities and agricul- domestic economy. ture subsidies to the federal budget. In the case of enter- The attempt to maintain the real exchange rate de- prises and commercial banks, the situation was less preciated in the context of formal wage indexation re- clear. Although enterprise losses were regularly calculat- sulted in a strong acceleration of wages and prices, as ed, frequent changes in accounting rules and a lack of noted above. The real devaluation also increased the bur- harmonization in accounting procedures decreased the den of foreign interest payments, thus offsetting in part reliance on available figures. The situation of the com- the benefits of the rescheduling of commercial debt. The mercial banks was not fully transparent either. While the pressure to finance hidden losses, particularly the cen- share of non-performing loans was known to be large, tral bank's own deficit, led policy-makers to abandon the the estimates were tentative. monetary targets soon after their implementation. More important, the design of a well-defined strategy Faced with higher real interest rates on bank credits, to deal with the loss-makers had not been completed. several enterprises simply stopped paying, a step that For instance, there were still doubts over whether to considerably worsened the already severe problem of provide a fiscal subsidy to loss-makers while submitting non-performing loans in the commercial banks. them to restructuring programs (involving lay-offs, debt 126 write-offs, selective improving investments, changes in of widespread price controls for goods and services as management, etc.), or whether to let the Darwinian- happened under other heterodox programs of the mid- Schumpeterian law of natural selection operate freely. In eighties. In addition to elements commonly found in this case, the number of bankruptcies was expected to other heterodox programs, the Yugoslav one included a increase much more rapidly, and the fiscal resources financial and industrial restructuring cum privatization would be directed toward social programs, as opposed to component that was supposed to affect most of the Yu- the loss-making enterprises themselves. Finally, al- goslav economic system, although this component had though privatization was already accepted as an essential not been fully worked out at the start of the program. component of the overall strategy, the mechanisms and The incomes policy consisted of a six-month freeze on timing of its introduction, as well as its interactions with the exchange rate, on nominal wages and on the prices the restructuring program, had not been fully worked of a set of goods (mostly in the areas of energy and trans- out. portation) that accounted for 20 percent of the consum- During 1989, progress in the establishment of the er price index. The prices of the other goods and services preconditions for a stabilization program was more ad- remained completely free. The exchange rate was frozen vanced in the macroeconomic area. First, the debt re- at 7 new Dinars per deutschemark right after a devalua- scheduling agreements of mid-1988 and the large tion of 20 percent on December 18. The devaluations current account surplus achieved in 1989 (US$2.4 bil- preceding the program were supposed to have corrected lion) resulted in an increase in international reserves the overvaluation that had occurred in the second half of from US$3.3 to US$6.1 billion. An adequate level of re- 1989, but they were rapidly eroded by the price increases serves was deemed essential to ensure the initial credi- that followed (figure 9-2-2).35 bility in the program, which was to include the Moreover, there was no attempt to use a larger real introduction of currency convertibility34 and a substan- devaluation to neutralize the impact of the introduction tial degree of import liberalization. In the fiscal area, all of currency convertibility and the reduction in import the changes in legislation required for the transfer of the restrictions. fiscal operations of the central bank to the federal budget The government froze nominal wages in December at and the introduction of a new sales tax were completed. the November levels adjusted for a 20 percent increase. The dramatic acceleration in inflation in 1989 (figure Although it was known that the freeze could cause a sub- 9-2-3) affected the elaboration of a stabilization program stantial erosion in real wages, they actually reached their that same year. As noted, several causes contributed to highest level of the decade in the second half of 1989 (fig- the acceleration. First, the attempt to maintain the real ure 9-2-2). Thus, it was felt that a larger initial correc- exchange rate depreciated in the presence of wage index- tion of nominal wages was not needed before the freeze. ation required an acceleration of inflation to achieve a The prices of energy, transportation and some other reduction of the real wage. Second, in the last three goods were frozen on December 18 after a series of ad- quarters of 1989 the real exchange rate target was partly justments in November and the first half of December. abandoned, but the very strong increase in real wages As with the exchange rate, the level of the price adjust- kept the pressures on inflation. Finally, the hidden losses ments seems to have been based on an expectation of in the economy had not been eliminated and still consti- rapidly declining rates of inflation after the program. tuted a permanent source of monetary expansion and in- The fiscal policy component of the program consisted flation. In addition, during 1989 the pressures on money basically of the assumption of several non-traditional ex- supply were compounded by the large build-up of for- penditures by the budgets of the federation and the re- eign reserves by the central bank. The perception that publics and the maintenance of a small budgetary inflation was running completely out of control motivat- surplus. The non-traditional expenditures included the ed the launching of the plan at mid-December, despite servicing of foreign liabilities of the National Bank of Yu- the awareness of some loose ends. goslavia, interest subsidies for agriculture, transfers to the bank restructuring program, transfers to the social The Stabilization Program of 1990 safety net, and coverage of some enterprise arrears. The overall expansion of fiscal expenditures and The stabilization program of 1990, as it was called, maintenance of the budgetary surplus required an in- contained a variety of measures in the areas of income, crease in tax revenues of approximately 5 percent of GSP. fiscal, monetary and trade policy. The program also in- The increase in revenues was expected to come partly cluded a currency reform that eliminated four zeros off from exogenous increases in tax revenues (3.5 percent of the old Yugoslav Dinar. The program can be classified as GSP) as a result of the introduction of a new federal sales heterodox, even though it did not include the imposition tax, higher customs duties and unspecified increases in 127 the republics' tax rates, and partly from endogenous in- Figure 9-2-5. Exchange Rate and Prices, Wages and creases in real tax revenues resulting from the stabiliza- Prices, and Base Money and Prices, 1988-91 tion itself (1.5 percent of GSP)-the inverse Olivera- Tanzi effect (Olivera 1967 and Tanzi 1977). Estimation of A. Exchange Rate and Prices the Olivera-Tanzi effect was complicated by the constant (monthly changes in %) changes in the tax rates in the 1980s, designed mostly to 70 offset the inflationary- erosion of real tax revenues. In ad- 60- dition, in 1989 a partial indexation mechanism was ap- 50- plied to the collection of the basic sales tax.36 40- / Reai i Monetary policy consisted essentially of a freeze on 30 * , * the nominal stock of the central bank's net domestic as- 20' sets, while also allowing the central bank to monetize 10 * Dinar/DM the foreign exchange inflows resulting from the pro- 0-o-..---- gram. However, it should also be noted that there was a -10 substantial real increase in the central bank's domestic 1989 1990 1991 credits and base money during December 1989 (figure 9- 2-5). Thus, monetary policy was less restrictive at the very start of the program than might have seemed at a B. Wages and Prices first examination. (monthly changes in %) Finally, the restructuring cum privatization compo- 60 nent included various measures that were still being de- 40 -N veloped or were running into problems with 50 , implementation. For instance, although the program 0 / 'Ret prices called for the financial restructuring of the commercial banks and the commitment of fiscal resources for this 20 Wages purpose, the agency that presumably was to lead the 10 Wages bank restructuring program was created only in mid- 0 V - 1990. -10. On the side of enterprise reform, it was finally decided 1989 1990 1991 that the program would not provide financing for loss- makers out of budgetary resources. It was expected that the program would therefore result in a large number of C. Base Money and Prices bankruptcies. To prevent enterprises from circumvent- (monthly changes in %) ing the lack of financing by lending to each other, on 110 100- January 1, 1990 the government introduced strict en- 190 forcement of the payments rules. The rules stated that 80 Base money enterprises which had not met their obligations within 70 enterprises ~~~~~~~~60-a 60 days would be declared bankrupt and closed. Enforce- 50o ment of this rule was possible because of a clearing sys- 40 prices 30 pie tem that centralized the payments among enterprises 20 * \ ' a and banks. 10 ,, , ,* ,,, ,, 0 Although the decision to force inefficient enterprises -10 into bankruptcy was in principle worthy of praise, the 1989 199 1991 strategy for dealing with bankrupt enterprises and the accompanying institutional and legal framework were Frst Results of the 1990 Program still incomplete. Regional institutions that would pre- sumably be placed in charge of restructuring and priva- Although it is too early to make a final judgment on tization had not yet been created. In addition, although the 1990 program, it is possible to assess the initial re- the program included the commitment of fiscal support sults and to identify the problems. As shown in table 9-2- to a social safety net, the amount of resources required 2 and figure 9-2-5, the first point to be made is the pro- seems to have been underestimated. gram's success in halting inflation during the first semes- ter without recourse to widespread price controls. The index measuring variations in the retail prices by calen- 128 Table 9-2-2. Monthly Variations of Retail Prices, December 1989-February 1991 (percent) 1989 1990 1991 Dec Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb (A) 59 42 13 5 3 0 0 2 3 7 8 3 3 5 10 (B) 64 17 8 3 0 0 0 5 2 1 10 n.a. n.a. n.a. n.a. n.a. Not available. Notes: (A) Classical Index, computed from mid-month to mid-month. (B) Supplementary Index, computed from beginning to end of month. Source: Federal Institute of Statistics. dar month (computed since December 1989) shows that Figure 9-2-6. Output, Real Tax Revenues and the Trade the decline in inflation at the start of the program was Balance A. Indusri P actually more pronounced than the classical index sug- (Deseasonalized, 1985=100) gests. However, table 9-2-2 also indicates a revival of in- 110 flationary pressures after July. 106 Industrial production fell by 10 percent in 1990, while 102 GDP fell by 7.5 percent. However, these figures might overstate the recessionary impact of the program. As 98 shown in figure 9-2-6, economic activity in the social- 94 ized sector, which was unusually strong in the first half 90 of 1989, had already weakened considerably in the sec- ond half of that year. In addition, the stability of electric- 86 ity production and the creation of large numbers of 82 small private sector enterprises during 1990 may indi- 1985 1986 1987 1988 1989 1990 cate that production in the private sector (not fully cap- tured in the statistics) was not affected to the same B. Real Tax Revenues degree. 150 (1988.1=1O0)a In spite of the decline in output, the increase in the 140 real tax revenues of the federation was impressive (figure 1 , 9-2-6). As mentioned, the increase reflects both endoge- 120 nous and exogenous factors. Total revenues rose less in 120 real terms, however, a reflection in part of the difficulty 110 of collecting taxes from an economy in recession and the 100 Total absence of significant exogenous tax adjustments at the 9.* local level. Although the public sector overall generated 80 * .. /Federation a small surplus in the first nine months of 1990, a devel- 70 opment that caused concern was the rapid growth in tra- 60 ditional expenditures by the federation, including the 1988.1 1988.2 1988.3 1988.4 1989.1 1989.2 1989.3 1989.4 1990.1 1990.2 1990.3 wages of federal employees, as they absorbed resources a. The figure following the year (.1-4) refers to the quarter of the year. that should ideally have been channelled to the restruc- turing programs. C. Imports and the Tfrade Balance, 1988-90 The accumulation of reserves was indeed substan- (in US$ millions) tial-US$4 billion between December 1989 and Septem- 6 ber 1990-and the stock of reserves rose to more than 5 TotaImpor US$10 billion, or 60 percent of the gross external debt. 4- On the side of the current account, the freezing of the 3 exchange rate and the introduction of convertibility led 2 to a rapid recovery in the level of workers' remittances Consumer Goods ,.._,,, .. starting in the last days of December. On the side of the . .° 0-:-.-. capital account, the same factors, combined with the im- -1 - rade Balance position of tight controls on domestic credit, led enter- -2 , prises to repatriate foreign exchange assets previously -3 held abroad. 1988 1989 1990 129 These foreign exchange inflows were the fundamental There are no clear indications that monetary policy source of the expansion in base money during the first was unduly restrictive at the start of the program. Inter- months of the program (table 9-2-3). After the December est rates on time deposits were around 12-15 percent per Table 9-2-3. Balance Sheet of National Bank of Yugoslavia, December 1989-December 1990 (billions of dinars) Dec. Mar. Jun. Sept. Dec. Net foreign assets 26 44 58 66 39 Net domestic assets 135 131 131 141 165 Base money 26 40 53 71 69 Foreign exchange liabilities to banks 135 135 136 136 135 Source: National Bank of Yugoslavia. "blip" (partly the result of increased inflows) and the Jan- year, an indication that liquidity conditions were not uary "crunch," base money expanded in keeping with the very tight.37 increase in National Bank of Yugoslavia's net foreign as- The level of the lending rates was much higher- sets. The result was a real increase of 70 percent in the around 30-35 percent per annum. These high intermedi- stock of base money between December 1989 and Sep- ation spreads reflect to a large extent the severe problem tember 1990 (measured by calendar inflation). of non-performing loans in banks' portfolios, rather than The reshuffling of the portfolios of asset-holders at the tight liquidity. In any case, the high lending rates tended start of the program is shown clearly in table 9-2-4. Dur- to aggravate the financial conditions of the enterprises ing the first three quarters, the increase in net foreign as- and increase their financial distress, creating obvious sets was accompanied by a rapid increase in Ml (driven complications for the program. especially by currency). The growth of Dinar time depos- Even though monetary policy did not seem unduly its was much slower, actually remaining below the rate of restrictive, the implementation of the program, com- inflation. Finally, the stock of foreign exchange deposits bined with strict enforcement of the new payment rules, was essentially stable. The fact that households did not had a strong impact on enterprises. During the first se- convert their foreign exchange deposits into Dinar depos- mester of 1990, 7,000 of them fell into some kind of ar- its could reflect an initial lack of confidence in the pro- rears, 3,000 were unable to make their payments within gram. In the Israeli stabilization program, the conversion 30 days, and 350 were declared bankrupt. A large num- of dollar-linked deposits into domestic deposits consti- ber of enterprises attempted to postpone bankruptcy by tuted a basic source of non-inflationary monetary cre- not paying wages. At the end of the first semester the sit- ation (Liviatan 1988). uation was, however, very differentiated, with several en- Of course, the evaluation of portfolio shifts resulting terprises increasing wages above the ceiling and several from the program also has to consider the repatriation of others unable to make wage payments within the ceil- foreign exchange assets previously held abroad by enter- ing. It was reported that a large number of enterprises prises and the conversion of deutschemarks and dollars did not make any wage payments in May to avoid the 60- held "under the mattress." The first was shown to be sub- day trigger mechanism for bankruptcy. stantial (around US$2 billion), while there is less infor- Pressure to relax the monetary policy mounted dur- mation about the second. ing the first semester of 1990 and produced a relaxation Table 9-2-4. Monetary Survey, December 1989-December 1990 (billions of dinars) Dec. Mar. Jun. Sept. Dec. Net foreign assets -61 -39 -29 -16 -46 Net domestic assets 306 314 333 355 388 Ml 51 75 100 125 127 Dinar time deposits 42 43 50 57 59 Foreign exchange deposits 152 157 154 157 156 M3 245 275 304 339 342 Source: National Bank of Yugoslavia. 130 starting in June. The relaxation was accomplished most- edly increased credits by US$1.8 billion equivalent of Di- ly by implementing measures to increase the multiplier, nars in order to finance pensions and enterprise wages although central bank credits also increased somewhat. ("No Such Thing as a Free Dinar" 1990), a step it took The result was a rapid increase in the stock of commer- without prior notification to the board of governors. The cial bank credits after that month (figure 9-2-7). This in- increase in domestic credits did not translate into mon- crease led to a further rise in wages and a revival of etary expansion in the fourth quarter because of the inflationary pressure in the second semester. The relax- US$3 billion loss of foreign reserves that happened dur- ation of monetary policy also aborted the expected ing the same period. As shown in tables 9-2-3 and 9-2-4 shake-out of the industrial sector by keeping loss-mak- and figure 9-2-7, the increase in domestic credits was al- ers afloat while allowing them to resume wage pay- most fully matched by a decrease in net foreign assets. ments. The result was roughly constant stocks of base money, Figure 9-2-7. Dinar Credits and Ml, 1989-90 Mi and M3. Thus, although there was no further escala- tion in inflation late in the year, the rapid loss in reserves A. Nominal created obvious doubts about the sustainability of the 2.2 stabilization program. The recognition that the real ex- 2.0 - change rate had become severely overvalued led to a cor- 1.8 _ rective 30 percent devaluation in January 1, 1991. In 1.6 - addition, the convertibility of the Dinar was abandoned 1.4 - in December, as the government interrupted sales of for- , 1.2 _ ,<. _._*jZg .. eign exchange to households to prevent a further loss of 1 / reserves. 0.8 - Dinar credits The devaluation of January 1, 1991 resulted in in- 0.6 _ / ,,.creasing rates of inflation in January and February (5 0.4 * Ml percent and 10 percent, respectively), despite the reim- 0.2 position of controls on the wages of loss-making enter- 0 ," , ,,, I ,,,,,,,,,,, prises and the efforts to reestablish control over 1989 1990 1991 monetary policy. Furthermore, the exchange rate again became severely overvalued as a result of the accumulat- ed increase in prices. B. Real (base=1989) An Assessment of the Main Issues 2.2 2.1 The Yugoslav program of 1990 is an additional exam- 2 ple of an exchange rate- based program lacking adequate 1.8 Dinar credits correction of domestic imbalances. The program was 1.7 - very successful in the first semester but started to falter 1.6 - in the second half of the year. Enterprises continued to t 1.5 - run losses, wages increased beyond the ceilings, fiscal re- 14 - sources were diverted toward wage payments of public 1.2 ,, sector employees, and, finally, monetary policy became 1.1 M1, largely accommodating. Although the maintenance of 1 , , , > , the exchange rate freeze and the open trade account pre- 0.8 .., .. vented a further escalation of inflation late in the year, 1989 1990 1991 the drastic overvaluation of the exchange rate and the rapid loss of reserves ultimately forced the government Concern over the revival of inflation led the central to make a corrective devaluation and to suspend the con- bank to shift back to a restrictive monetary policy in early vertibility of the Dinar. October. However, there are indications that the banks The challenge faced by policy-makers in 1991 is to avoided a contraction in credit by not complying with the avoid the reemergence of an exchange rate-wages-prices reserve requirements. A further indication of the difficul- spiral. The only solution to this problem is a return to ty the central bank was having conducting monetary pol- the discipline of the first semester of 1990, including the icy was the unusal episode of December 1990, whereby reinstitution of some kind of enforceable wage policy. Al- one of the regional branches of the central bank report- though the measures taken in early 1991 are a step in the 131 right direction, the prospects for their sustained imple- tually declined faster than indicated by the regular indices. mentation are still unclear, as it depends on a high de- 8. It appears that private sector activity is more reflected in the gree of consensus among the various republics. Polish figure than in the Yugoslav figure. If this is true, the difference gree in output performance may be larger than indicated. Fiscal policy also needs to be strengthened to support 9. The large current account surplus in Yugoslavia is generated the restructuring program, since the initial fiscal corn- from workers' remittances and tourism. mitments to the reforms were clearly insufficient. For 10. The real decrease in the Yugoslav M3 is a somewhat stronger instance, the resources allocated to the social program 12 percent when the foreign exchange assets held abroad by Yugoslav would clearly have been inadequate had loss-makers re- enterprises are included. However, it is still a much smaller decrease than in Poland. ally been forced into bankruptcy. Final estimates of the 11. In both countries, the decline in the share of foreign exchange quality of the portfolios of the commercial banks also in- deposits in broad money was almost entirely attributable to the real in- dicated the need for more fiscal support for the financial crease in the stock of domestic money. restructuring program. The fact that the public sector 12. In the case of Yugoslavia, the real exchange rate in the first ran a small surplus is almost completely irrelevant, as quarter of 1990 had already appreciated relative to the level in Decem- the imbalances in the Yugoslav economy were never lo- 13. The evaluation of the net impact of a real devaluation on out- cated in the non-financial public sector but rather in put is complicated by the notorious multiplicity of channels (see Liz- other sectors of the economy. ondo and Montiel 1989 for a recent survey). When the devaluation is Events in the second semester of 1990 also indicate carried out in the context of a stabilization program, it is further com- that, for the stabilization to be successful, there must be plicated by the simultaneous implementation of other policies. 14. This problem arose as a result of the control exercised on the much faster progress in the area of enterprise and bank banks by the enterprises. restructuring and privatization. To some extent, the slow 15. An interesting simulation by de la Calle (1990), based on data progress reflects institutional and human resource con- on domestic resource costs computed by Konovalov (1989a), estimated straints on implementation of the restructuring effort. It that the stabilization program would have led to a contraction of about also reflects the absence of adequate fiscal commitment to 20 percent in industrial output, a contraction led by the bankruptcy of the reforms. Finally, it reflects insufficient political re- enterprisesa particu arly in inefficie sion much more uneven than the solve to deal effectively with the loss-makers and the asso- one that actually took place. ciated increase in unemployment. Until these problems 16. The Polish rate of inflation in the 1980s was extremely high by are effectively tackled, however, the prospects for stabili- European standards, even excluding 1989. The average for the period zation and growth will be highly uncertain. 1981-88 was 35 percent, compared with 5 percent for industrial coun- tries, 38 percent for developing countries, 8 percent for Hungary and 1.3 percent for other Central and Eastern European countries, exclud- Notes ing Yugoslavia. 17. A macroeconomic analysis of the dynamics of inflation in a 1. The authors are grateful to Guillermo Calvo, Vittorio Corbo, mixed system of free and controlled prices can be found in Commander Alex Cuckierman, Miguel Kiguel, Nissan Liviatan, Branko Milanovic and Coricelli (1990a). and Jeffrey Sachs for their helpful comments on an earlier version of 18. Commander and Coricelli (1990a) show several cases in which the paper. The authors are also grateful to Rodney Chun for his very this mixed system of free and controlled prices, with "passive" money, competent research assistance. The views expressed in the paper are can generate explosive rates of inflation. the authors' and do not necessarily reflect the views of the World Bank 19. Saldanha (1989). The interest rate subsidies were measured by or its affiliated organizations. the difference between the nominal rates for credit and the rate of in- 2. In fact, Poland managed to run current account deficits flation. However, when the inflation tax on the monetary holdings throughout the 1980s, while Yugoslavia ran increasing current ac- (currency plus demand deposits) of enterprises is netted out, the net count surpluses after 1983. subsidies in the period 1984-88 fall to about 2-3 percent of GDP. 3. Even in countries with open inflation, it may be difficult to es- 20. Note that wage push and budget deficits are directly linked tablish a continuously close relationship between budgetary develop- through the wages of the public sector, which are tied institutionally ments, money creation and inflation. See Sargent and Wallace (1981), to wages set in the socialized sector (see Gomulka 1990). Bruno and Fischer (1987) and Drazen and Helpman (1888) for theoret- 21. The governmentdid notanticipate this effect, and ittried to re- ical demonstrations of the possibility of weak links in the short run and duce the wage-price spiral by lowering the coefficient of indexation and Kiguel and Liviatan (1990) for a recent examination of the Argentine forbidding wage increases in excess of the indexation rules. With accel- and Brazilian cases. erating inflation, however, the shortened interval of adjustment had a 4. In July both countries made some adjustments to the adminis- negative impact that dominated the moderating effects of the reduced tered prices. coefficients of indexation. 5. A common phenomenon in socialist countries that is known to 22. The wage bill was indexed to the changes in prices in the pre- weaken the power of monetary policy considerably. ceding month according to the following coefficients: 0.3 in January, 6, This point is acknowledged by Kornai (1990), who also stresses 0.2 in February-April (a coefficient that was ultimately maintained the importance of wage controls in the initial phase of reform pro- through June), 1 in July and 0.6 thereafter. This decision was taken to grams. See also Rocha (1991) for evidence on the absence of the Okun compensate for the effects on real wages of the administered increase relationship in Yugoslavia. of energy prices in July. The rationale is not fully clear, however, be- 7. Newly computed calendar indices for both countries, which are cause by July enterprises had accumulated significant room to increase provided in table 9.4 and in the two appendices, show that inflation ac- their wage ceilings, as in the first six months they had stayed well be- 132 low the ceilings imposed. References 23. The parallel market was very volatile in December: while the reduction in the overhang and the previous real exchange rate depre- Blanchard, O., and S. Fischer. 1989. Lectures on Macroeconomics. ciations pushed the parallel rate down to the level of the official rate in Cambridge, Mass: MIT Press. mid-December, in the last days of the month the rate approached 9,500 zlotys in anticipation of the decision actually made on January 1, 1990. 24. oreoer,almot hlf te icreae increit t no-govrn- Bole, V., and M. Gaspari. 1990. "The Yugoslav 'Way' to H-yperinflation." 24. Moreover, almost half the increase in credit to non-govern- Prcssd Begae ment was channelled to the non-socialized sector: households, farmers Processed. Belgrade. and private firms. 25. van Wijnbergen (1982) has emphasized the presence of a con- Bruno, M., and S. Fischer. 1987. "Seignorage, Operating Rules and the sistent bias of this type in IMF-supported stabilization programs. High Inflation Trap." NBER Working Paper No. 2413. National Bu- 26. The policy implications of this demand-led view of the reces- reau of Economic Research, New York. October. sion are straightforward: they imply some form of demand stimulus. However, the stimulus can be achieved with different policies and dif- Calvo, C., and F. Coricelli. 1990. "Stagflationary Effects of Stabilization ferent degrees of policy activism. A "non-activist" possibility would be Programs in Reforming Socialist Countries: Supply Side vs. De- not to change the policies and to let the decline in inflation, induced mand Side Factors." International Monetary Fund and World Bank. by the contraction in demand, reconstitute real monetary balances, Washington, D.C. August. and hence demand and output. A more activist approach would favor either increasing demaqd through a looser wage policy or more expan- Cohen, D. 1988. "The Management of Developing Countries' Debt: sionary fiscal policy or turning monetary policy around with a much Guidelines and Applications to Brazil." The World Bank Economic more accommodative credit policy (similarly to the recent experience Review 2 (1) (January): 1-48. in China in 1990). 27. Calvo and Coricelli (1990) provide a simple analytical model Commander, S., and F. Coricelli. 1991a. "The Macroeconomics of Price that illustrates this phenomenon. Reform in Socialist Countries: A Dynamic Framework." World 28. In addition, the central bank strengthened its supervision of Bank, Policy, Research and Affairs (PRE) Working Paper no. 555. the criteria that commercial banks are to follow in extending credit. It Washington, D.C. appears that an assessment of creditworthiness is a condition for the extension of credit. extension ohefall credit.supplyreducesthepermanentincomeofworke . 199 lb. Price Wage Dynamics and the Transmission of In- 29. The fall in supply reduces the permanent income of workers! flation in Socialist Economies: Empirical Models for Hungary and households, which accordingly reduces their consumption. Therefore, Poland. Policy, Research and External Affairs (PRE) Working Paper the fall in real wages and in consumption is a response to the credit Series. Washington, D.C.: aorld Bank, March. crunch and the attendant fall in supplies, and is not an exogenous source of the recession. 30. More detailed analysis of inflation in Yugoslavia is provided in de la Calle, L. 1990. "Macro and Microeconomic Linkages of the Polish Rocha (1991), Bole and Gaspari (1990), Gaspari (1988), Mates (1987) Reforms." World Bank, Washington, D.C. and Mencinger (1987). 31. See Mencinger (1989). That should be compared with an open Dornbusch, R. 1987. "Lessons from the German Inflation Experience unemployment figure of 14 percent.[ of the 1920s." In R. Dornbusch and S. Fischer, eds., Essays in Hon- 32. The stock of foreign exchange deposits (mostly held by resi- or of Franco Modigliani. Cambridge, Mass.: MIT Press. dents) is not part of Yugoslavia's external debt. In the early 1980s, the stocks of external debt and foreign exchange deposits amounted to Dornbusch, R. and Fischer, S. 1986. "Stopping Hyperinflation: Past US$20 billion and $11 billion, respectively. and Present." WeltwirtschaftlichesArchiv 122 (1)(April):1-47. 33. Therefore, the impressive correlation between enterprise loss- es and seignorage in table 9-2-1 is somewhat misleading, since the cen- Dornbusch, R., and M. Simonsen. 1988. "Inflation Stabilization: The tral bank did not finance enterprises directly. The inflationary Role of Incomes Policy and of Monetization." In R. Dornbusch, etc., financing of all hidden losses is more complex, involving the financing Exchange Rates and Inflation. Cambridge, Mass.: MIT Press. of central bank losses from base money creation and the financing of enterprise losses through subsidized bank credits. See Rocha (1991). Drazen, A., and E. Helpman. 1988. "Inflationary Consequences of An- 34. The dollar value of the stock of domestic currency and sight de- ticipated Macroeconomic Policies." Quarterly Journal ofnEconom- posits was $3.7 billion in December 1989. Thus, the level of reserves tci" would be sufficient to cover a large portfolio shift out of domestic as- sets. 35. However, the real effective exchange rate relative to December Fischer, S. 1977. "Long-Term Contracts, Rational Expectations and the 1989 in figure 3 understates the real devaluation that occurred right Optimal Money Supply Rule." Journal of Political Economy 85 before the freeze, since it is measured by the average exchange rates in (February). December adjusted by the Consumer Price Indexes in Yugoslavia and abroad. Gaspari, M. 1988. "Financial Crisis inYugoslavia Since the Early 1980s: 36. These measures succeeded in maintaining the ratio of fiscal Causes and Consequences." A paper presented at the Conference on revenues to GSP at around 35 percent during the second half of the Financial Reform in Socialist Countries, Florence, Italy, October 1980s. 1987. 37. However, the constancy of nominal rates raises doubts about its value as an indicator of liquidity. It is possible that the constancy of Gomulka, S. 1990. "Reform and Budgetary Policies in Poland, 1989- the free deposit rates reflects the maintenance of the discount rate at 90." In "Economic Transformation in Hungary and Poland." Euro- 23 percent per year during the first semester. pean Economy (43). 133 Helpman, E., and L. Leiderman, 1990. "Real Wages, Monetary Accom- Mencinger, J. 1987. "Acceleration of Inflation into Hyperinflation: The modation and Inflation." European Economic Review 34:897-911. Yugoslav Experience in the 1980s." Economic Analysis and Work- ers'Management, 21 (4):399-418. Hinds, M. 1990. "Issues in the Introduction of Market Forces in Social- ist Economies." World Bank. Washington D.C. January. . 1989. "Privredna Reforma i Nezaposlenost" [Economic Reform and Unemployment]. Privredna Kretanya Jugoslaviye Kiguel, M., and N. Liviatan. 1990. "The Inflation Stabilization Cycles in (March). Argentina and Brazil." World Bank, Policy, Research and External Affairs (PRE) Working Paper Series No. 443. Washington, D.C. Au- Montiel, P. 1989. "Empirical Analysis of High-Inflation Episodes in Ar- gust. gentina, Brazil and Israel." IMF Staff Papers 36 (September). Knight, P. 1984. "Financial Discipline and Structural Adjustment in "No Such Thing as a Free Dinar." The Economist. 1991. January 12, Yugoslavia." World Bank Staff Working Paper No. 705. Washing- p. 44. ton, D.C. November. Olivera, J. 1967. "Money, Prices and Fiscal Lags: A Note on the Dynam- Kolodko, G., and W. McMahon. 1987. "Stagflation and Shortageflation: ics of Inflation." Quarterly Review (Banca Nazionale del Lavoro) 20 A Comparative Approach," Kyklos 40:176-96. (September). Konovalov, V. 1989a. "Poland: Competitiveness of Industrial Activities: Pazos, F. 1978. Chronic Inflation in LatinAmerica. New York: Praeger. 196 1-86," World Bank. Washington, D.C. Rocha, R. 1991. "Inflation and Stabilization in Yugoslavia." Processed. World Bank. Washington D.C. 1989b. "Yugoslav Industry: Structure, Performance and Conduct." Mimeo. World Bank. Washington, D.C. November. Rocha, R., and F. Saldanha. 1991. "Fiscal and Quasi-Fiscal Deficits, Nominal and Real: Some Conceptual and Measurement Issues." Kornai, J. 1990. The Road to a Free Economy. Shifting from a Socialist Processed. Washington D.C. World Bank. System: The Case of Hungary. New York: Norton. Rotemberg, J. 1987. "The New Keynesian Microeconomic Founda- Lipton, D., and J. Sachs. 1990. "Creating a Market Economy in Eastern tions." NBER Macroeconomics Annual. Europe: The Case of Poland." Brookings Papers on Economic Ac- tivity 1:75-147. Saldanha, F. 1989. Processed. "Self-Management: Theory and Yugoslav Practice." World Bank. Washington D.C. Liviatan, N., 1988. "Israel's Stabilization Program." World Bank, Poli- cy, Planning and Research Department (PPR) Working Paper Se- Sargent, T., and N. Wallace. 1981. "Some Unpleasant Monetarist Arith- ries No. 91. Washington, D.C. September. metic." Quarterly Review (Federal Reserve Bank of Minneapolis) (Fall). Liviatan, N., and S. Piterman. 1986. "Accelerating Inflation and Bal- ance of Payments Crises, 1973-84." In Yoram Ben-Porath, ed., The Simonsen, M. 1989. "Inercia Inflacionaria e Inflacao Inercial" [Infla- Israeli Economy. Cambridge, Mass.: Harvard University Press. tionary Inertia and Inertial Inflation]. In F. Barbosa and M. Simon- sen eds., Plano Cruzado: Inrcia x Inpcia. Rio de Janeiro: Editora Lizondo, J., and P. Montiel. 1989. "Contractionary Devaluation in De- Globo. veloping Countries: An Analytical Overview." IMF Staff Papers 36 (1)(March). Tanzi, V. 1977. "Inflation, Lags in Tax Collection, and the Real Value of Tax Revenue." IMF Staff Papers 24 (March). Lucas, R. 1973. "Some International Evidence on Output-Inflation Trade-Offs." American Economic Review 63 (June). Taylor, J. 1979. "Staggered Price-Setting in a Macro Model." American Economic Review 69 (May). Mates, N. 1987. "Some Specific Features of Inflation in a Heavily-In- debted Socialist Country." Economic Analysis and Workers'Man- van Wijnbergen, S. 1982. "Stagflationary Effects of Monetary Stabiliza- agement 21 (4):419-32. tion Policies, a Quantitative Analysis of South Korea." Journal of Development Economics 10:133-69. 134 10 Comments on "Stopping Inflation: The Experience of Latin America and Israel and the Implications for Central and Eastern Europe" by Miguel A. Kiguel and Nissan Liviatan John Williamson It is a tribute to the authors of this paper that and important point that in both groups of countries sta- "Kiguel-Liviatan" is rapidly becoming a compound noun bilization seems at best to take inflation back down to akin to "Friedman-Schwartz." Their paper provides an where it was before the inflationary acceleration began. admirable framework within which to organize an Kiguel and Liviatan argue that inflationary pressures assessment of the policy problems facing the countries may persist for a time after stabilization if the public of Eastern and Central Europe, which are trying not to doubts the government's determination to stick with its succumb to chronic inflation. However, precisely be- proclaimed nominal anchor (the phenomenon of infla- cause their conclusions, which are based on a careful tionary inertia). If the government stands by the anchor, and broad-ranging analysis of the comparative experi- the result will be short-run deflation; to the extent that ence of different countries, are becoming so influential, wages are sticky downwards, the recession may persist it is the duty of a discussant to be particularly conscien- into the medium term. Because prolonged inflation tious in examining whether the policy conclusions may is politically painful and (at least superficially) economi- depend on questionable analysis. cally wasteful, the government will be tempted to seek The intellectual framework of Miguel Kiguel and some way of accommodating what it hopes will be a Nissan Liviatan is summarized in the 2x4 classification once-over price rise. Unfortunately, this accommodation displayed in table 10.1 of this paper.' They argue that fis- is difficult to carry out without undermining the credi- cal consolidation is a necessary but not sufficient condi- bility of its commitment to stand by its nominal anchor. tion for stabilization, which they also see as requiring a As a result, what the government initially intends as a nominal anchor that the authorities are prepared to once-and-for-all accommodation often becomes a re- make credible even at the cost of a recession. The au- sumption of inflation. thors observe that a recession typically comes sooner On examination, this account of inflationary inertia with a money-based nominal anchor than with an ex- embodies two distinct hypotheses. Hypothesis one is change rate-based one. They draw the important distinc- that agents (unions in particular, although not necessar- tion between chronic high-inflation countries and what ily exclusively) can and do choose to enforce a set of real by analogy can be called low-inflation countries. They income claims that are higher than what would be con- claim that the latter seem to have far less difficulty sta- sistent with a full employment equilibrium on the basis bilizing inflation after some shock pushes the rate far of announced macro policies. Hypothesis two is that above its previous level. They also make the interesting what causes agents to act in this way is their lack of Table 10.1 Kiguel-Liviatan Framework Fiscal Nominal anchor consolidation None Money supply Exchange rate Wage rate Yes No Source: The author. 135 confidence in the credibility of the government's an- Figure 10-1. Simonsen's 'saw-tooth diagram" nounced policy intentions. The principal criticism of Kiguel and Liviatan's paper w/p is that it goes along with practically all the literature of A i the last 15 years in assuming that hypothesis two is the only possible explanation for why agents choose to be- W have according to hypothesis one. However, another strand of the literature (it largely died out in the English language after Hirsch and Goldthorpe [1978] but has wi ... survived in the Portuguese language in Brazil) argues that agents make real income claims above the level con- sistent with equilibrium because they have become ac- customed to thinking that such a level is their just entitlement. For example, claims that are now collec- tively inconsistent may have been established at some time in the past, before the economy confronted an oil _ _ t price increase or a debt crisis and when its ability to pay t, t2 was greater. Alternatively, and of particular interest for present purposes, the inconsistent real income claims may have been nurtured by past inflation itself. Consider figure 10-1, "Simonsen's saw-tooth dia- Note also that inconsistent real income claims seem gram," which is familiar in the Brazilian literature. It to provide a readier explanation than the credibility ar- gram," w igument does for Kiguel and Liviatan's stylized fact that shows the movement of the (representative) real wage after a successful stabilization the rate of inflation tends over time under conditions of steady-state inflation. to return to its pre-disturbance level rather than to zero. After the wage settlement at time t, the real wage rises It is really rather difficult to see why a lack of credibility to its peak level w*. It then is gradually eroded by infla- shreally rarldiffcultt eew a lac ofered t tio unil he extwag inrese t tme 2. he ypoh- should so regularly prevent even a new government tion until the next wage increase at time t2. The hypoth- from being believed when it proclaims its determination esis is that labor tends to think of its just wage as w* to eliminate inflation, to the point where the govern- rather than the mean w'. If after stabilization-after the ment feels compelled to accommodate more or less the government has cut back its own demands on the econ- rate of inflation prevailing before destabilization started. omy-labor attempts to establish the real wage at w* In contrast, if agents are content to acquiesce in the re- rather than at w', the authorities will again have to establishment of their former real incomes but unwill- choose between recession and the resumption of infla- ing to abandon their former claims, then inflation will tion. A refusal to accommodate inflation will create a re- settle down again at its former rate under passive money cession of whatever severity is needed to batter the real (i.e., in the absence of a nominal anchor). income claims down until they are consistent with The credibility explanation of inflationary inertia reality (i.e., to w'). views government as engaged in a game with labor in Liviatan, in replying to these sorts of comments at the which its optional strategy is to pre-commit. The alter- conference in Pultusk, made the point that, if labor real- native view suggests that pre-commitment may not have ly believed it was entitled to w* rather than w', then much effect because people do not pay much attention to rational expectations would lead it to claim a still higher whether their micro claims are consistent with the gov- real wage than w* in order to raise the average real wage ernment's macro plans. They revise their claims only from w' to w*. The implication is that a steady-state in- when they perceive the government's plans to be in con- flation such as is shown in figure 10-1 is not a feasible flict with their micro realities. The government may be equilibrium. To the contrary, rather than implying that better advised to devote energy to educating the public inconsistent real income claims cannot explain infla- about the implications of its macro plans for the agents' tionary inertia, Liviatan's observation may explain micro alternatives. Presumably Helmut Schmidt was the stylized fact that it is more difficult to prevent a doing so during his chancellorship, when he is reputed 20 percent inflation from accelerating than to prevent to have spent some 20 percent of his time talking to a 2 percent inflation from doing so, and likewise that a trade union leaders. 200 percent inflation is far more prone to explode than The alternative view casts doubt on the wisdom of a a 20 percent one. government making a strong commitment to a fixed ex- 136 change rate as the nominal anchor. It suggests that the ness, a condition that assures prospective investors in reason this strategy has so often collapsed is that agents export industries that their returns will not be held hos- tend to pursue their inherited real income claims re- tage to the need to reduce inflation). It is the most neu- gardless of whether they would be serving their own best tral nominal anchor available. It can be adapted to interests if the government indeed stuck to its plans. combat the inevitable setbacks rather than confronting As the currency becomes progressively more overvalued the government with a stark choice of surrendering or because of the persistence of inflationary inertia, then fighting to the finish whenever events turn out less people may start to notice, but by then the betting has favorably than anticipated. started on how soon the government will renege on its Economists have embraced the doctrine that policy pledges and devalue, an expectation that is only rein- should always be consistent with the public's forming its forced by the government's increasingly more desperate expectations rationally. Few economists would abandon proclamations of its determination to defend the parity. that position because of evidence, of which there is now The process ends when a devaluation occurs and leaves plenty, that expectations are often not formed rationally. the government's credibility in tatters. The conclusion that surely should be drawn from the Must Poland repeat this sad story? The remarkable so- evidence that expectations are often not formed rational- cial solidarity witnessed in 1990 and people's willingness ly is that policy should be robust against the mechanism to accept real wage cuts give rise to more optimism that by which expectations are formed. A pegged exchange it can be avoided than Kiguel and Liviatan express. How- rate as nominal anchor fails that test: it performs satis- ever, this optimism is not based on Poland's continuing factorily only if a lack of credibility is the sole source of to defend a fixed nominal exchange rate. To the contrary, inflation inertia. it appears that the exchange rate is too brittle to serve as It is to be hoped that in future work Kiguel and a satisfactory nominal anchor for more than the first few Liviatan analyze the merits of a nominal income tar- months, or at most for the first year or two, after a stabi- get-and that this analysis be made available in time to lization program is implemented. (The European Mone- prevent Eastern and Central Europe from relying on the tary System [EMSI is moving toward the point where ability of fixed exchange rates to conquer inflation. fixed exchange rates may make sense, but to imagine that Eastern Europe can leapfrog the 25 years of history Notes that have gotten Western Europe to where it is today is fanciful.) Relying on the money supply is little better, 1. It would have been helpful for Kiguel and Liviatan to have used since it too produces strong distortions (high real inter- this taxonomy to organize their discussion and to have abandoned the est rates, an overvalued currency, recession and balance- use of emotionally charged words such as "orthodox" and "heterodox." of-payments deficits) that are prone to undermine confi- ing to find the latter defined as a versie anofhyms and it is disconcert- dence that the government will maintain its policies. definition of heterodoxy has the paradoxical effect of excluding the The wage rate is worse, for it generates a disequilibrium Austral and Cruzado Plans from the category, despite the fact that the in relative wages whose effects become progressively term was initially coined to describe them. more severe. 2. Williamson and Miller actually suggested a target for the The most promising candidate for nominal anchor in growth rate of nominal domestic demand rather than nominal income so as to build in pressures to correct balance-of-payments disequilib- the longer run is a target for the growth of nominal in- ia. The difference is not, however, central to the present discussion. come (Gordon 1990; and Williamson and Miller 19872). It is true that a nominal income target is effectively the References same thing as a price target under conditions of high in- flation, and as such it is not a candidate for the role of Gordon, Robert. 1990. "The Phillips Curve Then and Now." In P. nominal anchor in the initial months after the introduc- Diamond, ed., Growth, Productivity, Unemployment: Essays in tion of a stabilization program. However, once inflation Honor ofBob Solow's 65th Birthday. Cambridge, Mass.: MIT Press. has been brought down to single digits, it has all the ad- vantages of a money anchor (in particular, it is about as Hirsch, Fred, and John Goldthorpe. 1978. The PoliticalEconomy of In- easy to track and control), plus the enormously impor- flation. London: Martin Robertson. tant attraction of being consistent with a real exchange Williamson, John, and Marcus H. Miller. 1987. Targets and lndicators: rate peg (i.e., a crawling peg that moves to offset differ- A Blueprint for the International Coordination of Economic ential inflation so as to maintain constant competitive- Policy. Washington, D.C.: Institute for International Economics. 137 I Part V Structural Rigidities, Distortions and Inertia in Behaviors: The Supply Response to New Environments and Constraints in Central and Eastern Europe I 11 The Ownership-Control Structure and the Behavior of Polish Enterprises During the 1990 Reforms: Macroeconomic Measures and Microeconomic Responses Roman Frydman and Stanislaw Wellisz' This paper analyzes the behavior of the socialized sec- turn, those rewards and constraints are deeply rooted in tor in Poland during the first seven months of the reform the ownership and control structure of the Polish econ- program launched by the government in January 1990. omy.3 The principal purpose of the January 1 reforms was to fight hyper-inflation. Optimists also hoped that financial The Economic Reforms discipline and demand constraints would induce enter- prises to rationalize pricing, production and employ- Under the "classical" Soviet-type system that pre- ment. They further hoped that market pressures would vailed, with some modifications, from the late 1940s to drive the least efficient enterprises into bankruptcy. The the end of the 1970s, the economy was organized along pessimists, on the other hand, feared that managers, fol- hierarchic-bureaucratic lines. Enterprises were looked lowing the cost-plus principle, would raise prices in re- upon as administrative units, with no associated owner- sponse to costs and, if demand proved insufficient, ship rights, and their behavior was subject to the dictates simply slash production. There was also much skepti- of the plan. Industry was heavily concentrated for ease of cism about bankruptcy: given the tightly knit network of control. The reward system was based on fulfillment of exclusive suppliers and subcontractors and exclusive plan targets, and management was accountable to the customers, the financially strong firms would, it was ar- higher levels of the command structure. Market consid- gued, prop up the financially weak ones lest bankrupt- erations played a subordinate role. cies produce a chain reaction. During the 1980s the government progressively re- In retrospect it is clear that pressures on the product laxed the bureaucratic controls. The reforms embraced market induced by the reforms have forced firms to an ever-widening number of industries, and by the end adopt pricing that has largely eliminated shortages. of the decade only some of the key sectors (mining, However, the reform-related loosening of bureaucratic transport and communications, and basic industries) re- controls and political change have made managers more mained under centralized control. In other sectors, en- than ever beholden to labor. As a consequence, cuts in terprises were permitted to choose their sources of production have thus far not been matched by corre- supply, market their products, change the assortment of sponding cuts in the labor force. There has been some products, and decide, within limits, on the volume and cost rationalization, but here, too, managerial incentives direction of investments made out of retained profits. are weak.2 The government also gradually phased out the allocation It is shown here that in the absence of external ac- of foreign currency at official, artificially low rates. It countability, a characteristic of the post-command permitted enterprises to retain an increasing share of phase in the Central and Eastern European economies, their foreign exchange earnings and to use them to im- the only external instrument of control over enterprises port approved products, mainly producer goods, or to has been the government's credit policy. That credit pol- sell the retained foreign exchange at auction.4 icy, coupled with the internal structure of incentives of Despite the loosening of controls, much of the bu- the socialized enterprises, can explain the wage dynam- reaucratic-administrative structure held firm. For ex- ics during the first eight months of the reforms. The ev- ample, until the end of 1988 major enterprises idence seems to suggest that, given the rewards and continued to have only one marketing agent and a single constraints, managers strive to behave rationally. In source for principal inputs. For other firms the heavy 141 concentration of industry and trade barriers limited the The Socialized Sector range of choice. Most important of all, ownership and control rights remained unclear. Enterprise managers In 1989 the socialized sector in Poland, which gener- continued to face the possibility of arbitrary bureaucrat- ated an estimated 75 percent of the gross national prod- ic interference. An enterprise that set its prices at a level uct, consisted of over 21,000 enterprises. Of these, 9,000 that government bureaucrats deemed excessive could be were funded by central or local authorities, and 12,000 accused of gouging and be ordered to make a roll-back. were cooperatives. In December of that year the public On the other hand, if an enterprise made losses, it could sector employed 8.6 million workers, of whom just over receive a subsidy. The subsidies, which were subject to some 4 million were in industry, the sector that is the fo- administrative discretion, were granted easily to strate- cus here (table 11.1). There were a total of 5,486 indus- gically important enterprises, for example, coal mines or trial enterprises, including 2,463 cooperatives, most of steel producers. them small units. Large- and medium-scale units were, The profit bonus payable to management and employ- with few exceptions, centrally funded. ees was an insufficient incentive to pursue profit-maxi- Who owns Polish public enterprises? Is there a resid- mizing policies. For purposes of calculating the bonus, ual claimant who has the right to dispose of the proper- the subsidy was counted as part of the profit. Moreover, ty? How does that owner exercise control? There are no the profit incentive was blunted by taxes designed to dis- clear answers to these questions, although they are crit- courage "excessive" wage raises. In contrast, there were ical to understanding the pattern of behavior of the significant incentives to create shortages by setting pric- economy. The initial capital of the enterprises has come es below the market-clearing level. In a shortage situa- from a budgetary allocation administered through a tion the entire output could readily be sold regardless of "funding body," i.e., a ministry or local authority.6 Enter- quality. Letting preferred customers jump the queue and prises have no legal standing. Therefore a going concern reserving high quality products for them gave power to cannot be sold, although the funding body may dispose the individuals who could dispose of the product. They in of the assets of a bankrupt enterprise. Management has therindiids could dis posen of the rocT incom the right to sell selected assets, subject to the approval of turn could exact favors that enhanced their real Income. The environment within which thesociathe Workers' Council.7 The Council's approval is also The nvirnmet wihin hic thesocilizd setor needed to change the enterprise into a genuine joint- enterprises were operating changed abruptly with the stock c anyo to privatize it. reforms introduced on January 1, 1990. Most of the sub- stock company or to privatize it. sidies were eliminated: before the reforms, administered In large enterprises the funding body chooses the pri'esappiedtoan stiate 50pecen ofthenomna managing director, also subject to the approval of the prices applied to an estimated 50 percent of the nominal Workers' Councils. In smaller enterprises the Workers' value of legal transactions, a proportion that fell to 10 Cucl oiaeti esnadtefnigbd p percent. The subsidies for coal, energy and transport poves T orkers Con are empoere to rmv * proves. The Workers' Councils are empowered to remove were drastically reduced. To compensate for the subsidy the managing director or to suspend him or her. reductions, the price of coal was increased on average by The banks that grant operating credit to the enter- 500 percent, electricity 200 percent and transport 200 prises exercise financial control. The funding body does percent.5 The zloty was devalued by 58 percent and made not exercise control except over enterprises that go into convertible for transactions on the current account. bankruptcy. Net income is divided between the state The government also took measures to tighten de- budget and the enterprise. The former collects the "div- mand. Real interest rates that had been negative prior to idend," a lump sum based on the value of the funding the reforms were henceforth to be positive. The ration- capital, and a tax on an enterprise's profits over and ing of credit at preferential rates was discontinued. To above the dividend.8 The profit accruing to the enter- stop the price-wage spiral, tight limits were imposed on prise is allocated to the "development fund," i.e., to in- wage increases. From April until December 1989, wages vestment, and to the "personnel fund," i.e., to wage and were indexed at 80 percent of the cost of living index, and salary premia and social amenities. The allocation of the workers could obtain even higher wages through bar- funds is at the discretion of the Workers' Councils, with gaining. Beginning in January the permissible increases the credit-granting banks influencing the choice. The were tied not to the wage level but to the wage bill, and Workers' Councils also decide how to split the bonus be- the degree of indexation was reduced to 30 percent; in tween labor and management. Under the rules now in February and March it was further reduced to 20 per- force, bonuses in excess of the amount permitted by in- cent; thereafter it was subject to monthly revision. En- dexation are subject to heavy penalty taxes. terprises that exceeded the permissible increase were Labor organizations are also important players. The subject to heavy penalty taxes. original, submissive, government-sponsored unions col- 142 Table ll-1. Employment and Value Added in Socialized Industry Employment Value added on on December 31, 1989 June 30, 1990 in '000 in % (billions of zlotys) Industry of which: 4,016.3 100.0 109,641 Coal 482.5 12.0 1,576 Fuel 53.5 1.3 8,564 Energy 119.1 3.0 4,480 Steel production 142.1 3.5 10,388 Non-ferrous metals 60.2 1.5 6,017 Metal working 241.4 6.0 5,152 Machine-tools 407.4 10.1 9,895 Precision-tools 68.3 1.7 1,495 Transport equipment 311.3 7.8 7,870 Electronics equipment 247.7 6.2 6,029 Chemical 278.4 6.9 10,160 Construction materials 131.2 3.3 2,842 Glass 47.9 1.2 1,015 Ceramics 24.9 0.6 451 Wood products 157.9 3.9 2,508 Paper 45.5 1.1 1,560 Textiles 320.2 8.0 5,681 Clothing 175.6 4.4 1,756 Leather 132.4 3.3 1,789 Food 408.6 10.2 18,391 Animal feed 6.1 0.2 181 Printing 44.4 1.1 745 Other 109.7 2.7 1,096 Source: Central Statistical Office (Warsaw) data. lapsed with the formation of Solidarity. When Solidarity Effects of the Reforms on Enterprise Behavior was outlawed, the government created a new union or- ganization, the OPZZ. To gain legitimacy among the The analysis of enterprise behavior presented here is workers, the OPZZ assumed a belligerent stance and sur- based on aggregate data for the socialized sector. Also vived even Solidarity's regaining its official right to exist presented are conclusions drawn from a preliminary in 1989. It is something of a paradox that under the new analysis of a survey of 315 enterprises conducted by the regime Solidarity is closely associated with the govern- advisory group to the economic committee of the coun- ment, while OPZZ aggressively represents workers' par- cil of ministers. ticularist interests. As elsewhere in Europe, Polish legislation protects Price and Output Behavior workers from sudden mass dismissal. Management must In January 1990 the retail price index rose by almost notify the union prior to a planned group lay-off, that is, 80 percent (table 11.2). In the aggregate, this increase a lay-off of 10 or more percent of workers in enterprises reflects the effects of the cost-push from the supply with 1,000 or fewer workers or 100 or more workers in shock. In December 1989 there had been a sharp rise in enterprises above that size. The union has the right to producer prices: those realized by the metal-working in- obtain financial information concerning the enterprise dustry rose 61 percent, machine tools by 56 percent, the and to present a counter-proposal. This step is followed chemical industry by 60 percent and construction mate- by negotiations, although, in the case of a disagreement, rials by 52 percent.9 The December price increases af- the manager has the last word. It must be borne in mind, fected the January production costs. In addition, as however, that the Workers' Council may dismiss the mentioned, the prices of coal, fuel, energy and transport manager. Therefore, when managers need to reduce the were raised drastically as of January 1, 1990. labor force, they generally opt for gradual attrition. 143 Table 11-2. RetailPr;ices Januaryl989-Augustl990 mining sector declined by 1 percent (7 percent), while the output sold by manufacturing declined by 11 percent Year Month December 1988=100 Previous month=100 (22 percent). Most of the branches of heavy industry ex- 1989 January 111.0 111.0 perienced little decline in sales; in some, such as metal- February 119.8 107.9 lurgy, the January 1990 sales were higher than those in March 129.5 108.1 December 1989. However, sales of machine tools de- April 142.2 109.8 clined by 25 percent (12 percent), of precision tools 34 May 152.4 107.2 percent (13 percent) and of transport equipment 34 per- June 161.7 106.1 cent (25 percent). The decline in light industry, 11 per- August 247.1 139.5 cent (18 percent), was slightly more severe than in September 331.9 134.4 manufacturing as a whole, and the consumer durable October 513.6 154.8 branches were hard hit. Clothing sales dropped by 11 November 628.5 122.4 percent (77 percent), leather goods by 14 percent (23 December 739.8 117.7 percent). 1990 January 1,328.6 179.6 When a recession occurs in Western market econo- March 1,7156 123.8 mies, a decline in production in heavy industry often April 1,844.2 107.5 precedes and is more severe than the decline in the con- May 1,929.1 104.6 sumer goods' sector. In the Polish case, exactly the oppo- June 1,994.7 103.4 site has been true. The decline started and was very deep July 2,066.5 103.6 in the consumer goods sectors, and enterprises in those August 2,103.7 101.8 sectors lowered their price mark-ups significantly. Out- Source: Central Statistical Office (Warsaw) data. put sold by the producer goods sector declined initially by a lesser percentage, while the mark-ups increased. It is hypothesized that different explanations stem- An upper-bound measure of the supply shock effect ming from the structure of the Polish economy are can be obtained by substituting the exogenous price needed to understand the behavior of the two sectors. It changes in an input-output table.10 This calculation, is the view here that the fall in output and reduction in based on a 39x39 input-output table, yields a 77 percent mark-ups in the relatively competitive consumer goods price increase in the retail price index, a rate that closely sector were caused by the decline in consumer demand. matches the actual increase (table 11.3). The strongly monopolized producer goods industries, As the input-output table indicates, the supply shock insulated to a larger extent from market shocks, affected the capital goods sector more strongly than it switched, under the impact of the stabilization mea- did the more labor-intensive consumer goods sector sures, from a policy of below-market clearing to monop- (middle column of table 11.3).11 In the producer goods olistic pricing. industries, actual price increases (right-hand column of table 11.3) exceeded the price increases based on the in- Household Behavior and the Consumer Goods Sector put-output prediction. The machine tool, precision tool and transport equipment industries are the three nota- The stabilization measures caused an immediate and ble exceptions. In the case of the first two, exports play profound decline in constant-price household in- an important role; hence, a significant component of the comes.12 The index of the money income of the Polish aggregate realized price is determined in the world mar- population declined from 100 in July 1989 to 68.5 in Jan- ket. Much of the transport equipment was sold under uary 1990 at constant prices. During the same period the long-term contracts. More importantly, as will be dis- constant price index of household expenditures declined cussed below, domestic sales of the three sectors de- from 100 to 63.5. In February there was a further drop in clined more drastically than they did for the rest of incomes and an even deeper drop in expenditures. These industry. In the construction, leather, textile and cloth- figures suggest that the propensity to consume de- ing sectors, the price increases were substantially lower creased in a period of declining real incomes. This phe- than indicated by the input-output table. All three sec- nomenon can be explained by the "Pigou effect." The tors experienced a greater decline in output than did in- hyper-inflation drastically reduced the value of zloty dustry as a whole. holdings. Following stabilization, the real value of the In January 1990, the output sold by socialized indus- US dollar on the Polish market dropped steadily. The rise try was 10 percent lower than in December 1989 (21 per- in the propensity to save is the consequence of the effort cent lower than in January 1989). The output sold by the by households to rebuild their assets. 144 Table 11-3. Results According to the Input-Output Table, January 1990 (percent) Hypothetical Actual increase of prices increase based on in prices increases in costs in January (December 1989=100) (December 1989=100) Industry Coal 285.0 404.0 Fuel 205.2 294.0 Energy 218.0 300.3 Steel production 218.9 235.0 Non-ferrous metals 174.7 285.0 Metal-working 194.4 197.0 Machine-tools 184.6 175.4 Precision-tools 174.3 167.2 Transport equipment 188.9 162.5 Electric machinery 180.1 186.3 Chemical 186.9 207.2 Construction materials 202.9 224.0 Glass 179.3 203.6 Ceramics 181.3 187.7 Wood products 182.5 170.4 Paper 192.1 182.4 Textiles 173.3 159.3 Clothing 164.5 135.0 Leather 170.7 152.8 Meat 175.8 190.0 Other food products 178.1 150.0 Animal feed 182.0 290.0 Printing 175.6 180.0 Other 181.1 183.0 Construction General 186.1 149.0 Production & services 184.4 148.0 Specialized 186.4 146.0 Other 173.0 145.0 Agriculture Vegetable products 199.5 220.0 Animal products 188.5 162.2 Agriculture services 182.8 220.0 Forestry 188.8 170.0 Transport 241.0 280.0 Telecommunication 182.0 180.0 Trade 183.8 160.0 Other branches of production 166.0 185.0 Public services 204.7 240.0 Average price realized by producer 193.7 196.8 Retail price 177.0 179.6 a. The profit rates in each industry are assumed to be equal to the average of historical rates. Source: Andrew Berg, Dariusz Jasczynski and Jan Rajski. A comparison of the average propensity to consume comes from changes in household assets: between Janu- during the first six months of 1990 with the correspond- ary and June 1990 real household incomes dropped 8 ing figure for 1989 supports this conjecture (figure percent, but dollar-denominated household holdings 11.1). Within that period, real household incomes (the main store of non-agricultural wealth) rose from dropped by 36 percent, household expenditures by 41 US$4.9 billion to US$5.2 billion (table 11.4). percent. As a consequence, the average propensity to Other, less readily quantifiable considerations also consume declined from 0.84 to 0.78. Other evidence support the demand constraint hypothesis. Hoarding of 145 Figure 11-1. Average Propensity to Save Out of durables, which occurred during the hyper-inflationary Money Incomes and Real Household Income period, subsided, and households rebuilt their cash bal- (in real terms, July 1989 =100) ances (figure 11.2). Foreign consumer goods became 2s0 , somewhat more readily available. It is also possible that APS ...-.. n the private sector produced more goods.'3 In the pre-re- APS = Average RI-l = Real form period, the public sector was able to sell all its pro- propensity household duction, regardless of quality. Since the reforms, to save income 200 defective goods remain unsold. The Producer Goods Sector 150 The producer goods industry consists of large enter- .5 /~\ / \1 1 I \ /~\ prises, interlinked by a traditional administrative struc- ture. Under the old regime these enterprises produced 100 - ' *@ mostly for each other, and a complex system of subsidies - t$' . - and protection from outside competition shielded them V ., ,r-., ,,, from market pressures and from the necessity of paying so ... much attention to the future use of their products by the consumer goods industry. As discussed, prior to the reforms below-equilibrium pricing was prevalent throughout the economy. The 19 J M M J S N , , , , Jul hard line taken on new subsidies and the reduction of ex- Feb Apr Jun Aug Oct Dec Feb Apr Jun Aug isting ones induced enterprises to cover their costs. Un- Months der the new rules enterprises were free to raise prices. Clearly, the shortage-creating pricing strategy was no Source: Central Statistical Office (Warsaw) data. longer possible, and enterprises now sought to set prices at profit-maximizing levels. Table 11-4. HouseholdDollarDeposits, January1989-August1990 Household dollar Year Month deposits Index (millions US$) (January 1989 =100) 1989 January 4,139 100.0 February 4,402 106.4 March 4,464 107.9 April 4,516 109.1 May 4,561 110.2 June 4,458 107.7 July 4,550 109.9 August 4,618 111.6 September 4,762 115.1 October 4,864 117.5 November 4,863 117.5 December 4,925 119.0 1990 January 4,911 118.7 February 4,969 120.1 March 5,040 121.8 April 5,123 123.8 May 5,214 126.0 June 5,252 126.9 July 5,409 130.7 August 5,599 135.3 Source: Central Statistical Office (Warsaw) data. 146 Figure 11-2. Real Wages and Stock of Money in Although both the consumer and producer goods sec- Households tors attempted to raise prices to profit-maximizing lev- (in constant July 1989 prices) els,14 as has been stressed, the initial outcomes in terms of prices and output were quite different in the two sec- -w Wages tors. An explanation of this disparity lies not in the dif- AZotydeposit ferences in motivation of the enterprises but in the 140 / \ - - - - $ Deposits difference in constraints. The consumer goods sector ex- / . Cash perienced a severe shift in its demand curve, while in- 120 dustries in the producer goods sectors traded with each other and so at first were less affected by the demand or 100 *. , \ credit constraints. 80 Price and Output Behavior in Later Months 60 .\.. The initial price adjustments took place under highly \'\ "S """"' uncertain cost and demand conditions, and profits 40 o^_,_ , ,,,+" __.- plummeted (table 11.5). There is evidence that as the X. .. , economic environment became more stable, enterprises 20 ................. engaged in search behavior that led to improved prof- its.'5 The survey (referred to above) shows that enterpris- o ,_,,__, __,, _,, __, __,es that initially raised their prices by a wide margin 1989 Jul Sept Nov 1990 Mar May July tended to lower their prices later on, whereas those that Aug Oct Dec Feb Apr Jun Aug raised them relatively little, or in some cases not at all, Monts later made upward adjustments. In January 1990, 92 percent of the surveyed enterprises raised their prices; of :these, 60 percent lowered them in February, 54 percent in March and 49 percent in April. Of the firms that raised Table 11-5. Sales and NetProfits of the Socialized Sector, January 1989-July 1990 (billions of zlotys, constant December 1988 prices) Net profitl Year Month Sales Net profits sale ratio (/) 1989 January 4,498 504 11.1 February 4,108 551 13.4 March 5,379 550 10.2 April 3,764 430 11.4 May 4,328 632 14.6 June 6,000 896 14.9 July 2,841 320 11.3 August 4,028 439 10.9 September 5,493 746 13.6 October 3,517 711 20.2 November 4,221 794 18.8 December 5,013 1,904 38.0 1990 January 3,315 425 12.8 February 3,050 467 15.3 March 3,414 535 15.7 April 3,039 353 11.6 May 3,399 434 12.8 June 3,265 436 13.4 July 3,001 304 10.1 Source: Calculated on the basis of Central Statistical Office (Warsaw) data. 147 their prices in January, 21 percent increased them in es from domestic producers as well as from the ruble February, 10 percent in March and 15 percent in April. area countries went down-a clear indication that Pol- However, these price increases were much lower than ish industry was seeking to modernize. those in January. This picture is representative of the economy as a whole: even as hyper-inflation subsided, Employment prices continued to rise (table 11.2). After the initial fall, aggregate output continued to Employment in the socialized sector declined decline, albeit at a lower pace; by July it stabilized, and in throughout the 1980s. In 1989 alone it fell from 9.2 mil- August it increased, following a real rise in wages in July. lion to 8.6 million, or close to 7 percent. There was, how- As time has progressed, the demand constraint has ap- ever, no overt unemployment, for the expansion of the parently reached heavy industry: by June 1990 most private sector created a sufficient number of jobs to em- branches experienced a decline in output of 25-30 per- ploy the workers shed by the socialized sector as well as cent relative to June 1989. Interestingly, this drop is new entrants into the labor force. about the same percentage as was experienced by light In 1990 the rate of reduction of the labor force in the industry in January. socialized sector quickened, but, unlike in production, there was no dramatic break with past trends. In Janu- ary, in the face of a drop in output of close to 30 percent, Foreign Trade employment fell by just over 1 percent. In the months that followed, the rate of labor-shedding accelerated Changes in the patterns of exports and imports dem- somewhat, although it never surpassed 2 percent a onstrate that the Polish socialized sector is flexible. The month. In total, during the first eight months of 1990 domestic recession and the devaluation of the zloty in- employment in the socialized sector declined by creased the profitability of sales to hard currency coun- 870,000, or about 10 percent (table 11.6). tries. During the first six months of 1990, such exports, During the first three months of the year, output per reckoned in constant dollars, were 20 percent higher employed worker was about 26 percent lower than dur- than during the same period a year earlier. Imports from ing the first three months of the previous year (table the convertible currency and ruble areas declined. It is 11.7). Group lay-offs were insignificant (table 11.8), the interesting to note, however, that imports of machines reason being that according to the rules such lay-offs re- from the convertible currency areas rose, while purchas- quire a 60 to 90 day notice. Moreover, the unions as well Table 11-6. Employment in the Socialized Sector, January 1989-August 1990 Year Month Employment Change over the previous month ('000) ('000) (%) 1989 January 9,189 - - February 9,184 -5 -0.05 March 9,092 -92 -1.01 April 9,009 -83 -0.92 May 8,954 -55 -0.61 June 8,874 -80 -0.90 July 8,864 -10 -0.11 August 8,771 -93 -1.06 September 8,721 -50 -0.57 October 8,715 -6 -0.07 November 8,690 -25 -0.29 December 8,600 -90 -1.05 1990 January 8,508 -92 -1.08 February 8,414 -94 -1.12 March 8,305 -109 -1.31 April 8,199 -106 -1.29 May 8,040 -159 -1.98 June 7,950 -90 -1.13 July 7,844 -106 -1.35 August 7,734 -110 -1.42 - Not applicable. Source: Central Statistical Office (Warsaw) data. 148 Table 11-7. Production Sold per Worker in Socialized Industry, January 1989-August 1990 Corresponding month of previous January Year Month Zlotys year=100 1989=100 ('000) (%) (%) 1989 January 246.2 117.9 100.0 February 217.2 100.3 88.2 March 243.6 101.8 98.9 April 224.8 107.5 91.3 May 227.4 99.4 92.4 June 241.8 103.3 98.2 July 189.4 98.0 76.9 August 205.1 94.9 83.3 September 221.9 98.8 90.1 October 214.7 102.9 92.5 November 214.7 91.0 87.2 December 227.2 102.1 92.3 1990 January 185.3 75.3 75.3 February 160.6 73.9 65.2 March 179.3 73.6 72.8 April 165.7 73.7 67.3 May 175.6 77.2 71.3 June 176.5 73.0 71.7 July 169.9 89.7 69.0 August 183.3 89.4 74.5 Source: Central Statistical Office (Warsaw) data. Table 11-8. Unemployment, 1990 (last day of the month) Jan. Feb. Mar. Apr. May June July Aug. Rate of unemployment (%) 0.4 1.1 2.0 2.6 3.3 4.2 5.2 6.1 Unemployment ('000) 56 152 267 351 443 568 699 820 of which (%): Group lay-offs 4.3 4.5 5.7 7.8 9.5 10.2 11.3 12.2 Other separations n.a. n.a. 64.3 7.8 9.5 10.2 11.3 12.2 Unemployed for more than 3 months before registration n.a. n.a. 30.0 43.8 44.8 42.0 39.2 37.0 n.a. Not available. Source: Central Statistical Office (Warsaw) data. as the Workers' Councils pressured management not to Wages discharge workers. The hope that the anti-inflationary squeeze would lead to a rapid improvement in the allo- During the first half of 1990 money wages consistent- cation of labor proved unjustified. Nevertheless, over the ly rose less than permitted by indexation (table 11.9). next four months productivity improved, and by August Month by month there was an accumulation of arrears- it recovered to within 13 percent of that of the previous sums that could be paid to workers without incurring year. The index does not, of course, take into account the tax penalties. However, the rate of accumulation de- noticeable improvement in the quality of products. clined uniformly. In July and August the wage increases 149 Table 11-9. Wage Indicators and the Evolution of Average Wages in Five Main Sectors of the Socialized Economy, January-August 1990 Jan. Feb. Mar. Apr. May June July Aug. (percent) 1. Indexation coefficient 0.3 0.2 0.2 0.2 0.6 0.6 1.0 0.6 2. Increase in retail prices (%) Forecast 45.0 23.0 6.0 6.0 2.5 3.0 5.5 3.0 Actual 78.6 23.9 4.7 8.1 5.0 3.4 3.6 1.8 3. Actual wage increase (%) (without bonuses from profits) 5.2 5.4 10.5 2.5 8.7 5.6 14.4 5.2 4. Maximum allowable rate of growth of wage fund (increase relative to the previous month) According to the price forecast 13.5 4.6 1.2 1.2 1.5 1.8 5.5 1.8 According to actual prices 23.6 4.8 0.9 1.6 3.0 2.0 3.6 1.1 ('000 zlotys per employee)a 5. Actual wage without bonuses from profits 618.3 651.9 720.1 738.2 802.2 846.8 969.2 1,019.9 6. Maximum wage (the wage norm) 734.3 778.1 795.8 819.1 860.3 887.6 934.8 958.3 7. Monthly employment relative to the previous month 98.9 98.9 98.7 98.7 98.1 98.9 98.4 98.6 8. Forecast norm 674.4 776.8 797.9 815.8 847.8 885.8 952.0 965.2 9. Unused norm based on the forecast 56.1 124.9 77.7 77.5 45.6 39.0 -17.2 -54.8 10. Unused norm based on the forecast as % of the previous month's wage 9.6 20.2 11.9 10.8 6.2 4.9 -2.0 -5.7 11. Unused cumulative norm 116.0 242.3 318.0 398.9 457.1 498.1 463.8 402.2 a. Indexation base in '000 zlotys = 587.9. Source: Calculated on the basis of Central Statistical Office (Warsaw) data. exceeded the amount permitted by indexation, and the presents a computation of the wage norms for the aggre- cumulative arrears started to decline. gate economy. The question is why this pattern of wage behavior oc- The January 1990 wage norm is obtained by multiply- curred, and more particularly, the extent to which the ing the December 1989 indexation base of 587,900 zlotys indexation rule actually constrained the wage increas- by the maximum allowable rate of growth of wages. es.16In examining this question, it is necessary to start by Since the actual rate of inflation in January was 78.6 per- explaining the indexation system introduced in January. cent and the indexation coefficient was .3, the maximum It is easiest to do so using actual data, as displayed in ta- allowable rate of growth of the wage fund was 23.6 per- ble 11.9. The indexation specifies the maximum total cent (line 4). When this figure is adjusted for the change wage payments (the maximum wage fund) for every en- in aggregate employment (line 7), the result is the norm terprise and every month. The maximum wage or wage for January, 734,300 zlotys. An enterprise that exceeded norm (line 6) is obtained by dividing the maximum wage the norm was subject to a 200 percent tax on the excess fund by the employment. A fall in employment automat- payments below 3 percent of the norm and a 500 percent ically increases the wage norm per worker. Table 11.9 tax on payments higher than 3 percent.17 150 Wages are set monthly. Since the norm is set on the wage increases granted in August further reduced the basis of the ex-post inflation rate, enterprises have to aggregate unused reserve. predict the inflation rate to predict the norm for the To conclude, although it cannot be denied that the in- coming month. The only widely publicized forecast in dexation cum penalty tax system had a restraining effect Poland is published by the Ministry of Finance. Hence ta- on wages in the case of some enterprises, the view here ble 11.9 supposes that this forecast is used to predict the is that this mechanism was not a major determinant of norm. Since a 45 percent rate of inflation was forecast the evolution of the average wage since January 1990. It for January, the forecast increase in the wage norm was is necessary to look for additional explanations of the assumed to amount to 13.5 percent (line 4) and the cor- wage dynamics in response to the credit and interest rate responding forecast of the norm to equal 674,400 zlotys policy. (line 8). As it turned out, on average the enterprises ac- tually paid a wage of 618,300 zlotys (line 5). Thus, the ex- Interest Rate and Credit post unused norm was 116,000 zlotys (line 11), and the unused norm based on the forecast was 56,100 zlotys In December 1989, the last month prior to the intro- (line 9). The ex-post January norm in turn serves as the duction of the stabilization program, retail prices were basis for February's calculations. rising at 17 percent a month. The discount rate of the basis for FerNational Bank of Poland was set at 13 percent. Since the To return to the question of whether the indexation stabilization program was widely publicized, expecta- system constrained the wage increases actually paid by tions were that prices would rise sharply in January and the enterprises, it is also possible that the low wage in- that credit would be tightened. In anticipation of these crease granted in January reflects management's cau- tion iventhe ncertinty Whie wags roe in developments, enterprises attempted to build up mnven- Fbon grven the uncertasnty. Whale wages rose m tories of raw materials, a decision that led them into Fiebruary, on average the increase was only 33,600 zlotys hevdbtTeprotinfivnoisfnacdy .. . . ~~~~~~heavy debt. The proportion of inventories financed by (line 5). Although this increase is a 5.4 percent rise over enterprises from retained earnings fell from 48 percent the actual January wage (line 3), the gap between the at the end of June 1989 to 38 percent at the end of De- wage norm (whether computed ex post or ex ante) and cember, while accounts payable rose from 30 percent to wages more than doubled (lines 5 and 6). The unused re- 34 percent. serve grew in February from 116,000 zlotys to 242,300 December also brought windfall profits. The US$2.4 zlotys per worker. The unused reserve grew further over billion in the enterprise accounts were revalued from the next four months, although at a diminishing rate, 3,800 zlotys per dollar to 6,500 zlotys to reflect the offi- and by June it amounted to close to 500,000 zlotys, com- cial devaluation of the zloty; the revaluation yielded pared with a monthly wage of almost 850,000 zlotys. close to 6.5 billion zlotys. Thus, at the beginning of the It is therefore difficult to claim that, during this peri- program, the enterprises were in a relatively strong fi- od, the indexation rule significantly restrained wages. It nancial position. should not be concluded, however, that it had no effect. On January 1, 1990 the National Bank of Poland Although average wages were below the norm, some en- raised the discount rate to 36 percent a month. It expect- terprises paid the maximum permitted under the norm ed that this rate would exceed the average rate of infla- or even surpassed it. In January the treasury received tion for January-February. However, prices rose faster 1.75 billion zlotys in penalty taxes, and these payments than foreseen, and the ex-post January real rate of inter- increased every month, reaching 267.8 billion in June. estwas strongly negative (table 11.10). The new rates ap- For the first six months of the year the penalty taxes plied, however, to old as well as new debt. The rate amounted to 608 billion zlotys, or about 2 percent of the increase imposed a burden on the heavily indebted sec- wage fund. tors of the economy, especially agriculture and con- The indexation rule also cannot explain the rapid in- struction, which were granted special relief for interest crease in wages that occurred in July and August. In July payments.19 the government adopted full indexation to gain public Inflation subsided in February, and although the acceptance of the administered price increases (notably nominal discount rate was reduced to 20 percent, the ex- for coal),18 but the rate of inflation was not expected to post real rate was positive. In the following months, in surpass 4 percent. The actual wage increases exceeded anticipation of a further decline in the rate of inflation, 14 percent. The raises were not confined to enterprises the discount rate was gradually reduced. with unused reserves, witness the fact that the penalty The monetary authorities were subject to strong, tax payments amounted to 640 billion zlotys-that is, contradictory pressures. On the one hand, high rates of more than the total for the previous six months. The interest were considered important to holding the price 151 line. On the other hand, the continuing recession and of self-financing. On the whole, thanks to the amassed deepening unemployment were used as arguments to inventories of raw materials and their highly liquid posi- ease credit. Under the policy that was adopted, the ex- tions, the enterprises did not seem to have been severely ante real discount rate declined month by month. Credit constrained by inputs of raw materials. In February and was eased. For the first six months of the year, the real March the growth of nominal credit exceeded the rate of ex-post discount rate was positive. In July, however, it inflation. At the end of the first quarter, credit in real once again became negative. terms was 37 percent lower than at the end of December During the first months of the stabilization program, 1990. With the decline of nominal and real interest rates, credit to the socialized sector did not change in nominal credit was gradually rebuilt. terms, so that it registered a decline of about 50 percent It is striking that the time path of the enterprises' in real terms.20 wage policy parallels that of credit policy. The tight cred- The implication is that the enterprises perceived the it period of February through April occurred at a time increase in the nominal rate as an increase in real costs. when wage payments lagged behind the norm. As credit There may have been other reasons, as well, why enter- eased, the percentage gap between the wages paid by en- prises reduced their indebtedness. In view of the accu- terprises declined. When, in July, the ex-ante real inter- mulation that took place toward the end of 1989 and the est rate turned negative, nominal wages increased by decline in demand, the enterprises appropriately re- 14.4 percent, and for the first time in 1990 they exceeded duced their inventories and thereby decreased the need the norm21 (table 11.9). This also occurred in August, for inventory financing. Given the high nominal rates on the last month for which data were available. loans and the uncertainty concerning real rates, the en- It is hypothesized here that this parallelism of the terprises also found it preferable to increase the degree time paths of interest rates and wages indicates the di- Table 11-10. MonthlyInterestRates, January-July1990 (percent) Jan. Feb. Mar. Apr. May June July (nominal) National Bank of Poland discount rate 36.0 20.0 10.0 8.0 5.5 4.0 2.5 6-month time deposits 17.0 13.0 6.5 5.0 3.4 3.5 2.0 1-year credit: Minimum 36.0 20.0 9.0 7.5 5.0 4.0 2.5 Maximum 62.0 23.0 12.0 9.5 8.0 5.5 2.6 (real based on the actual point-to-point inflation rates) National Bank of Poland discount rate -33.9 14.0 3.7 1.6 0.7 0.4 -1.2 6-month time deposits -43.1 7.3 0.4 -1.2 -1.3 -0.0 -1.7 1-year credit: Minimum -33.9 14.0 2.7 1.1 0.2 0.4 -1.2 Maximum -21.3 16.8 5.6 3.0 3.1 1.9 -1.1 (based on the forecast inflation rates) National Bank of Poland discount rate -6.2 -2.4 3.8 1.9 2.9 1.0 -2.9 6-month time deposits -19.3 -8.1 0.5 -0.9 0.9 0.5 -3.3 1-year credit: Minimum -6.2 -2.4 2.3 1.4 2.4 1.0 -2.9 Maximum 11.7 0.0 5.7 3.3 5.4 2.4 -2.8 Memo items: Rate of retail price inflation (point to point) 106 5.3 6.1 6.3 4.8 3.6 3.8 Retail price inflation rate 45.0 23.0 6.0 6.0 2.5 3.0 5.5 (forecasua a. Forecasts of inflation rates are available only for average rates. Also note that since the point-to-point inflation rate in February was much lower than the average rate, the ex-ante real rate in February was actually higher than the rate reported in the table (and was probably positive). Source: Calculated on the basis of Central Statistical Office (Warsaw) and National Bank of Poland data. 152 rection of causation from the government's credit policy Credit was very tight in February, but easier credit to the wage policy of enterprises. The missing link in this over the ensuing months lowered the cost of inputs. chain of causation is provided by the demand constraint With product prices and output unchanged, enterprises coupled with the incentive structure in Polish enterpris- were able to raise wages. This highly simplified model is es. consistent with the parallel movement of wages and in- terest rates. Incentives and Responses The argument thus far is that the lowering of interest rates reduces the costs of production and, with a given Price and Output Response demand, raises value added, so that there is a possibility of raising wages as long as the norm is not binding. How- In a situation of declining demand for labor such as ever, for marginal enterprises, that is, for those threat- the one that has prevailed since January 1990, the ened with bankruptcy, employment is a control variable. unions and the Workers' Councils have strongly opposed A Workers' Council is likely to agree to a group lay-off if group lay-offs. A manager is unlikely to gain approval for the alternative is bankruptcy or a reduction of all wages a group lay-off unless this measure is required to save below an acceptable minimum. Unemployment com- the enterprise from bankruptcy.22 As a consequence, as pensation is indexed to the overall wage level; hence, ce- shown by the unemployment statistics, the decline in teris paribus, this minimum is an increasing function of employment in the socialized sector is almost entirely the wage norm. the result of individual dismissals for cause and of people As a first approximation, it can therefore be assumed quitting (table 11.8). that in enterprises on the verge of bankruptcy there is no In enterprises that are not threatened by bankruptcy, labor-sharing and that workers' pay, which is deter- the number of workers remains constant as long as out- mined by the minimum acceptable wage, is equal to put does not exceed "capacity," defined here as the out- their marginal product. Thus, in the case of marginal en- put that the normal labor contingent could produce terprises, wage increases constitute a cost push. working the usual weekly hours. When output is below Wages also influence prices via agriculture. An aver- capacity, there is labor-sharing. The drop in measured age Polish family spends 50 percent of its income on output per worker that occurred in 1990 strongly sug- food. The limits imposed on exports maintain the Polish gests that labor-sharing took place.23 prices of most major foods at a lower than world price The above observations suggest that as long as an en- level, and, in the short run, supply in most cases is very terprise is not threatened by bankruptcy, the manager inelastic. Agricultural products have a 50 percent weight takes the size of his labor contingent as fixed. It can be in the retail price index. In the short run, a rise in wages argued that, under these circumstances, the manager induces a rise in agricultural prices and in the retail will strive to maximize value added. As long as payments price index. are below the norm, workers can pressure for higher This model, as they all do, simplifies reality. Neverthe- wages. Moreover, both management and labor derive less, it throws light not only on overall price and wage benefits from enterprise profits in the form of bonuses movements, but also on specific, seemingly anomalous, and social amenities. Workers also benefit from invest- instances. Thus, although in July and August credit was ments that lighten the work burden or improve the eased and wages rose above the norm, inflation in Au- working environment. gust was the lowest since the stabilization began.26 How- Insofar as a part of all current non-labor inputs are ever, in August the prices of the products of the light bought on credit, a rise in interest rates boosts produc- consumer industries rose, although, as noted, this sector tion costs ceteribus paribus.24 Costs obviously also rise is the least profitable. Its wages also lag the most relative with increases in the prices of material inputs. In Janu- to the norm. Hence it may be inferred that bankruptcy ary 1990, costs increased sharply for both reasons. In the looms as a possibility in this sector.27 absence of knowledge of demand conditions, and given the widely publicized forecast of a drop in demand, man- The Rationalization of Production agerial prudence called for a low wage policy. In the ex- ante sense, low wages were also in the interest of labor. Consider now the problem of the rationalization of The reasons for, and the probable effects of, the anti-in- production. Rationalization measures that increase val- flationary policy received wide publicity, and workers ue added without causing labor redundancy are clearly were well aware that enterprises might go bankrupt. Un- beneficial to workers. The efficiency incentive is, of der these circumstances job security was more impor- course, blunted by the norm and by the limitations im- tant than wages.25 posed on bonus payments. It is clear, nevertheless, that 153 the imposition of a hard budget constraint and elimina- by machines, and at a time of growing unemployment tion of the subsidies spurred enterprises to seek more they are not likely to favor labor-saving improvements as profitable markets, cheaper sources of supply and im- long as they have a voice in management and as long as proved production methods. Two-thirds of the enterpris- there is unemployment. Thus, the incentives for achiev- es covered by the survey reported that in 1990 they ing efficient production in the long run are weak. In- introduced changes in the organization of production. deed, they may be even weaker now than they were About the same proportion reported changes in the as- under the planning system, for at that time workers had sortment of products. Fewer than 50 percent reported no fear of unemployment. changes in production technology. At one time the belief that the withdrawal of subsidies Is There a Role for an Expansionary Aggregate Demand would cause widespread bankruptcies that would weed Policy? out production and increase overall efficiency was fairly common. In free market economies an enterprise goes Poland has overcome its 1989 hyper-inflation, but the bankrupt if it fails to meet its fixed obligations. In the ab- economy is now in a state of stagflation. Orthodoxy calls sence of long-term debt, the payment of the dividend was for the continued application of anti-inflationary mea- adopted as the solvency test. The dividend, however, is a sures until such time as price stability is firmly restored. misnomer: it is in fact a capital tax that at present is set Yet political pressures are mounting to relax the controls at 32 percent of the funding capital, that is, at the (infla- and apply standard Keynesian measures to cure the re- tion- adjusted) depreciated value of the capital provided cession. by the funding authority. The impact of the dividend is On the demand management side the authors tend to unequal. New enterprises are subject to a higher one agree with the opponents of expansionary policy. As long than are old ones, which have depreciated their original as the control structure of enterprises is not modified, capital and made new investments out of retained prof- this policy is likely to result in higher wages and lead to its. Overall, it is estimated that the dividend is equal to 8 renewed inflationary pressures. The cost of the short- percent of the book value of fixed capital. Since the an- run gains in output may therefore be too high. nual rate of inflation surpasses 8 percent of a wide mar- In conclusion, the authors do not think that, in the gin, the overall impact of the dividend is very small. absence of deep changes in the structure of ownership So far, except for a few enterprises in the food pro- and control of enterprises, restrictive aggregate demand cessing sector-the hardest hit by current recession- policy is an effective way to achieve rationalization and there have been no bankruptcies. There are two main growth. The crux of the matter lies in the definition of reasons, in addition to the insignificant level of the divi- ownership and control: there is no well-defined owner dend. An enterprise may delay bankruptcy through de- who seeks to maximize the present value of the returns capitalization-it may either fail to keep up the stock of on capital and who is able to determine the use of re- its fixed capital, or it may sell some of its operations sources. Under the orthodox planning system, faulty and (usually the profitable ones) to the private sector. By highly irrational as it was, the government exercised such means it might be able to meet its current obliga- control. In the present transitional state, there is none. tions and keep on paying its workers. There is no reason Emergence of a new private structure of control requires to believe, however, that decapitalization is, in all cases, widespread privatization.28 the rational course from the point of view of the manage- ment of capital. Second, given the tightly knit network Notes of suppliers, subcontractors and customers, the bank- ruptcy of a weak unit may deprive a strong one of an im- 1. The authors benefitted from discussions with Guillermo Calvo portant source of supplies or an important outlet. and Fabrizio Coricelli. They also thank Dariusz Jaszczynski for excel- portant source ~~~~~~~~~~~~lent research assistance and informative discussions and Jan Rajski for Therefore, financially strong enterprises have an interest help with the data. Roman Frydman gratefully acknowledges a grant in supporting financially weak ones through the exten- for this project from the C.V. Starr Center for Applied Economics at sion of inter-firm credit. This credit puts off the day of New York University. reckoning but increases the danger of a chain reaction in 2. For an analysis of the relationship between the disciplinary ef- which sound creditors will go bankrupt because of the fect of the pressures on the product market and the problem of control over enterprise management in the 1990 reforms, see Chapter 21, insolvency of the debtors. "Markets and Institutions in Large-scale Privatization: An Approach to The structure of control also discourages moderniza- Economic and Social Transformation in Eastern Europe," by Roman tion of the labor-saving type. It has been argued here that Frydman and Andrezj Rapaczynski, in this volume. redundancy is not a proximate reason for group dismiss- 3. For earlier related analyses, see Calvo and Coricelli (1990), Fry- als. However, since the Luddite movement if not before, dman, Wellisz and Kolodko (forthcoming), Lipton and Sachs (forth- workers have been well aware that they can be replaced coming) and Phelps' (forthcoming). 154 4. For earlier related analyses, see Calvo and Coricelli (1990), Fry- norm, 200 percent on a 3-5 percent excess and 500 percent on an ex- dman, Wellisz and Kolodko (forthcoming), Lipton and Sachs (forth- cess over 5 percent. coming) and Phelps' (forthcoming). 18. Administrative fixing of the price of coal was abolished in July, 5. From July 1990 on, the price of coal was no longer adminis- but the administration retained the right to delay price increases by tered. Instead, it was "negotiated," i.e., set by mutual agreement be- three months. In July, the mining industry was permitted to raise pric- tween the sellers and the principal buyers. The remaining subsidy for es by 15 percent; prices will be raised again in October. coal was discontinued. 19. All debtors were permitted to capitalize 60 percent of the inter- 6. As the government's structure has evolved, the number and des- est payments on old debt, with interest paid only on the remaining 40 ignation of the funding bodies have changed. At present, industry is percent. However, agriculture was granted a 50 percent subsidy for its under the tutelage of a single Ministry of Industry. Other funding bod- debt payments; in the case of credit for construction, the subsidy was ies include the Ministry of Transport and the Ministry of Telecommu- to be 32 percent of the interest on the total debt, so that debtors had to nications. pay interest on only 8 percent of the debt. 7. The Workers' Councils originated as part of the effort by anti- 20. According to the IMF, total credit to non-financial public en- Communist forces to loosen the Party's grip on the economy. The law terprises amounted to 30.6 trillion zlotys at the end of December 1989 creating them was passed in September 1981, atatime when Solidarity and 31.8 trillion zlotys at the end of January 1990. The index of prices exercised considerable influence. However, most were set up during realized by industry, as defined earlier, rose 110 percent. the ensuing period of martial law, when Solidarity was suspended and 21. Wages above the norm could be paid without incurring penal- eventually outlawed. The Councils are elected by secret ballot for a ties because of the accumulated reserves resulting from the shortfalls three-year period. In some enterprises, notably in several steel mills, in wage payments during the previous months. they have been captured by Solidarity; in others, the elections have in Ag paymen in the pevousemonths. ' . ~~~~~~22. As discussed in the second section, the manager needs the ap- caused little excitement, and in practice the Councils have come close proval of the Workers' Council to effect a group lay-off. He or she may, to being tools of management. However, Council membership confers however, dismiss individual workers for cause. power, and the Councils now constitute a lobby of some importance. 8. The dividend is a 32 percent tax on the funding capital. It does 23. An alternate hypothesis that this measure of output per not apply to capital generated through the reinvestment of profits. On worker is faulty, since it does not take into account the quality im- average, the tax amounts to 8 percent of estimated total capital, al- provements that took place since the introduction of the anti-inflation- though the percentage varies greatly from firm to firm. ary program. Doubtless there is some truth to that contention, but it 9. Realized prices are the weighted average of the prices obtained fails to explain (1) the increased frequency with which paid and unpaid by enterprises for domestic sales and for exports to the ruble bloc and leaves have been granted to workers and (2) the acceleration in the re- to convertible exchange countries. duction of employment in the socialized sector. 10. This statement applies, sensu stricto, to a perfectly competitive 24. Calvo and Coricelli (1990) have proposed a connection be- economy. In the Polish case the calculation assumed that profit mar- tween the costs of inputs and credit and restrictions on output on the gins remained constant. supply side. They also provide some empirical evidence that the credit 11. Because of the high degree of aggregation, the changes in sec- crunch has caused the initial drop in the output of the socialized sec- toral prices differ, in some cases very markedly, from those of the sec- tor. More empirical research is needed to disentangle the demand and tors' major products. For instance, the sectoral price rise indicated for supply explanations of output dynamics during the 1990 reform. the coal mining sector is much lower than the exogenous price rise of 25. From the partial equilibrium point of view of the average en- coal because the sector embraces, in addition to coal production, vari- terprise, the low wage decision was correct. By raising wages to the lev- ous ancillary activities and services, such as canteens, miners' housing el permitted by the norm, such an enterprise would have incurred and rest homes. losses. 12. A distinction is drawn here between the statistical concept of 26. As discussed, the 15 percent increase in the price of coal ac- changes in real income and the welfare concept. The latter assumes counts, in part, for the price increase in July. that the quality and assortment of goods remain constant, as does 27. A stronger test of the model would require examining the rela- search time. The former, which for the sake of clarity is called here tion between prices, wages and group lay-offs. At present, the relevant changes in money income at constant prices, makes no such assump- data are not available. tions. Stabilization and liberalization improved the availability of 28. For an analysis of these issues see chapter 21, "Markets and In- goods and their quality; hence, the real drop in income in the welfare stitutions in Large-scale Privatization: An Approach to Economic and sense was less severe than the statistical decline. Social Transformation in Eastern Europe," by Roman Frydman and 13. During the first six months of 1990, employment in private Andrezj Rapaczynski, in this volume. non-agricultural activities increased 2 percent. There is no sectoral breakdown, and no data on output. 14. As will be discussed, the enterprises are modeled as striving to References maximize the value added at the firm level, subject to various con- straints. Calvo, Guillermo, and Fabrizio Coricelli. 1990. "Stagflationary Effects 15. By March 1990, profits in constant prices rose to 97 percent of of Stabilization Programs in Reforming Socialist Countries: Sup- the 1989 level, although output was 27 percent lower, ply-Side vs. Demand-Side Factors." Mimeo. World Bank and Inter- 16. The following analysis, based on new data, substantially ex- national Monetary Fund. Washington, D.C. September. tends and revises the authors' earlier discussion in Frydman, Wellisz and Kolodko (forthcoming), Frydman, Roman, Stanislaw Wellisz and Grzegorz Kolodko. Forth- 17. The tax rates were changed on July 28, retroactive to January coming. "Stabilization in Poland: A Progress Report." In Emil- 1. Under the new rules a 100 percent tax is payable on the excess of Maria Claassen, ed., Exchange Rate Policies of Less Developed wages over the norm if the excess does not surpass 3 percent of the Market and Socialist Economies. 155 Lipton, David, and Jeffrey Sachs. 1990. "Privatization in Eastern Eu- Phelps, Edmund S. Forthcoming. "Sub-normal Unemployment in So- rope: The Case of Poland." Brookings Papers on Economic Activi- cialist Economies." In Emil-Maria Claassen, ed., Exchange Rate ty, 2:293-341. Policies of Less Developed Market and Socialist Economies. 156 12 Life after the Polish "Big Bang": Representative Episodes of Enterprise Behavior Erika A. Jorgensen, Alan Gelb and Inderjit Singh Polish firms recently experienced a series of econom- The result was that for the first few months of 1990 ic shocks that occurred in three overlapping waves. First uncertainty and chaos reigned in the industrial sector. there was the severe macroeconomic instability (from Firms struggled to gain a sense of the nature of their August 1989 to February 1990). Then in January 1990 markets for inputs and outputs. Domestic demand con- the government implemented a package of economic tinued to fall, credit remained tight, and expectations policy measures that has come to be called the "Big about the important CMEA market became ever more Bang". It consisted of strong stabilization measures, lib- pessimistic. eralization of the trade regime and creation of a convert- What Poland now needs to pull itself out of this eco- ible currency. The government also took important steps nomic quagmire is a strong supply response to the re- toward reforming the industrial system by freeing most forms in the form of an expansion of industrial prices and eliminating most subsidies. Last, there was production, which will come about as the aggregate ef- the exogenous shock of the disintegration of the regime fect of the behavior of individual industrial firms. The of international trade of the Council of Mutual Econom- question is if and when that supply response will occur. ic Assistance (CMEA), a process that is ongoing. By coin- Given the string of negative shocks cited above, it would cidence, the five-year cycle of planned CMEA trade ended not have been surprising had firms become paralyzed by in 1990. Through the spring of 1990 the enormous polit- the uncertainty and unable to react. ical and economic upheavals occurring all over Central Contrary to that expectation, it seems, based on an and Eastern Europe and in the Soviet Union made it in- analysis of a sample of nine Polish industrial firms, that creasingly clear that standard negotiation of the next five they reacted strongly and often positively. This paper ex- years of CMEA trade was unlikely. Even the remaining plores the response of those firms to the January 1990 contracts for 1990 came into question as doubts about economic reforms in Poland and the contemporaneous delivery and payment grew and as governments backed shock of the breakdown in CMEA trade. away from the subsidies promised to exporters as part of the original bilateral agreements.! At the time, Poland The Methodology and the Sample was selling 45 percent of its exports under the CMEA sys- tem, equal to 9 percent of GNP. Of these non-convertible The Methodology currency exports, 54 percent consisted of machinery and transport equipment. Thus, when the CMEA arrange- This paper describes some of the microeconomic re- ments broke down, a key market for Polish exports dried sponses of a small sample of nine industrial firms to the up completely. Big Bang during the first six months of the economic re- The first wave of economic shocks gained strength form (January to June 1990). The information about the through 1989 with a sharp rise in inflation and tighten- firms in the sample was obtained during visits by a World ing of credit. The January 1990 reforms (preceded by Bank team to each firm during June and July 1990, at some preliminary policy changes in September 1989) which time managers were interviewed and basic data on then fundamentally transformed the environment in financial and real variables were collected. which firms were operating, with promises of still more This direct approach was used for a number of rea- change. Growing doubts about the future of the CMEA sons. First, aggregate data often obscure important be- trading system became a serious worry by March 1990. havior at the sectoral or subsectoral levels and suffer 157 from lags in availability. In the case of this study, timeli- have been overlooked. For example, firms may be so in- ness was very important. Second, it is more difficult to fluenced by the uncertainty over ownership and control assess the condition of firms on the basis of centrally col- by the policy-makers or so paralyzed by the general un- lected data in a reforming socialist economy than it is in certainty of the economy that they react to new econom- a capitalist economy, where changes in the recorded ic policies in seemingly perverse ways. Last, there has rates of bankruptcy, profitability and investment provide been much variation in the performance of firms, with important information for tracking the impact of policy some doing well and others badly even within the same changes. Third, the macroeconomic data in socialist subsectors. Interviews with firm managers can provide economies may be of especially poor quality because of some sense of whether particular characteristics render non-standard collection practices and inadequate con- firms flexible in the face of change and whether success- ceptual definitions. This situation is exacerbated in Po- ful adjustment is mostly a matter of luck and historical land by the distortions in the economic environment, accident. such as the high rate of inflation in 1989. Fourth, when change is both drastic and ongoing, recorded data be- The Sample come more difficult to interpret, and the lags before which patterns become apparent may be overly long. Fi- State-owned industrial firms were the focus of the nally, it is difficult to analyze the impact of macroeco- field trip. They have accounted for almost 90 percent of nomic policies in socialist economies. As such, micro the net material product (NMP) in industry in recent analysis becomes all the more important. years in Poland and almost 90 percent of industrial em- Given the above problems, a decision was made to talk ployment. The private firms are too small in the aggre- directly with the managers of industrial firms to gain in- gate to absorb much labor in the short run or to affect sights into their responses to the economic reforms and to total production. As a result, it is the aggregate response use the results as context when examining the statistical of the state-owned firms in industry that is crucial for a data from the firms. This approach yields first-hand evi- macroeconomic supply response led by the industrial dence of the perceptions of the managers as to their firms' sector. changing budget constraints, evolving objectives and Since timeliness was a primary goal, only nine firms shifting scope of possible actions. The combination of data were visited: seven state-owned enterprises, one cooper- from the statistical agency and interviews with manage- ative and one private joint venture (for purposes of com- ment at individual firms can provide clear and timely in- parison). The firms covered a wide range of industrial sights into recent economic behavior, as well as guidance subsectors, including engineering, transport equip- on key areas for government action. This direct examina- ment, chemicals, electronics, light industry, food indus- tion can also assist in forecasting the likely direction of try, and wood and paper. They also covered the range of the supply response. product markets, including domestic, CMEA export and There are a number of other advantages to using hard currency export. firm-level data and interviews to analyze the reaction to Without examining the entire universe of firms in Po- the Big Bang. First, the microeconomic impact of mac- land, it is impossible to know whether the few firms in roeconomic policies is immediately evident from firm- the sample are representative of Polish industry as a level data. Retrieving the data directly from the enter- whole in terms of initial conditions or responses to the prise avoids the inevitable delays in the collection of Big Bang, as similar information on the entire industrial macroeconomic data. Interviews with managers give in- sector is not readily available. A few preliminary compar- sight into how these economic agents perceive current isons can be made, however. Table 12-1 displays some of reform policies and can disclose divergences between the basic characteristics of the firms in the sample, of the the objectives of the managers and the government. Sec- average firm in the Lista 500 (the top 500 state-owned ond and more important, the behavioral responses to firms when ranked by sales), and of the average firm in macroeconomic policy changes may well be different in Polish state-owned industry overall. The average firm in a reforming socialist economy than in a market econo- the sample had sales close in volume to the Lista 500 av- my, especially in the case of state-owned industrial firms. erage in 1989 but greater employment. The sample firms Problems of ownership and control may dominate their also had significantly higher shares of exports in total reactions, leading to objectives that are quite different sales. from those of privately owned firms. Third, important This range of firms provided a broad picture of indus- obstacles to adjustment by firms may have caused them trial behavior but ruled out any formal comparative to react in ways different than expected to the changes in analysis, since sectoral differences could not be captured the macroeconomic environment, and those factors may within the small sample. 158 Although a sample of only nine firms offers sparse in- economy much of the way toward a free market orienta- formation, their behavior does provide insight into the tion. issue of when and if a supply response will emerge to re- The stabilization component of the program was het- vive the Polish economy from the deep recession in- erodox in nature, using both the wage and exchange rate duced by the profound economic reforms. This first cut as nominal anchors for price stability. The predicted at understanding firm-level responses needs to be fol- budget deficit for 1990 was reduced to 1 percent of CDP lowed by more detailed and quantitative analysis of a by cutting the subsidies for consumption and produc- larger sample of firms to test the generalizability of the tion. The government tightened credit by increasing in- findings. Nevertheless, the information presented here terest rates sharply and reducing the quantity of credit. offers much-needed anecdotal evidence of the microeco- To create a nominal anchor, it instituted an incomes pol- nomic responses and focuses further debate and analy- icy that involved levying a heavy tax on firms that ex- sis. ceeded the government-mandated wage increases, a measure designed to repress the nominal wage. Another The 1990 Policy Package nominal anchor was the pegging of the zloty to the dol- lar at a fixed rate, resulting in a large devaluation of al- Beginning in 1981, Poland undertook a partial eco- most 60 percent. This devaluation allowed the nomic reform that significantly diminished the role of simultaneous introduction of many policies directed at central planning in the economy. Enterprises gained the liberalization of trade. The government unified the some autonomy, although with much variation across foreign exchange market, eliminated the restrictions on sectors. For example, firms that produced consumer access to foreign exchange and replaced the quantitative goods were allowed to make decisions about most as- restrictions on imports with a unified customs tariff. pects of current operations. Another measure estab- The Big Bang package also included important steps lished some self-management of firms through the toward the creation of a market economy. The govern- introduction of the Workers' Councils, with limited ment lifted almost all the remaining controls over rights in enterprise decision-making. However, the gov- wholesale and retail prices except for energy. In addition, ernment maintained its control through informal bar- it instituted unemployment insurance and other social gaining between its financial authorities and enterprises assistance programs that formed the beginnings of a that determined the levels of prices, interest rates and safety net to ease and encourage adjustment by labor. tax rates for the enterprises. Unfortunately, this system The unemployment insurance legislation passed in Jan- of compromised central control discouraged financial uary 1990 also allowed employers to fire employees on discipline by the firms and eventually created serious one month's notice. The combination of openness to the macroeconomic imbalances in the Polish economy. world markets and liberalization of domestic prices was The further reforms introduced from 1987 to 1989 intended to shift relative prices to reflect scarcity, an im- failed to improve the macroeconomic conditions, al- portant step in encouraging the reallocation of resourc- though they did help set the stage for the drastic reforms es toward more productive uses. of the new Solidarity government. For example, the sys- Firms had to cope not only with these macroeconom- tem of centralized allocation of inputs was pared back to ic policy shocks but also with other policy changes that just five commodity groups, and firms were given great- had a primarily microeconomic focus. The monopoly po- er freedom to set the prices of their outputs. In August sition of the central distribution system was eliminated 1989, the government freed the prices of agricultural in- by law, and anti-monopoly legislation was prepared to puts and products. These goods accounted for approxi- counteract the high concentration of Polish industry. mately 20 percent of GDP, so that the share of goods with The government instituted a number of policies to hard- decontrolled prices reached 50 percent of gross domestic en the budget constraint2 of state-owned firms. For ex- product (GDP). ample, the levy on the fixed assets of state-owned Despite these measures, by mid-1989 it became clear enterprises was modified as of January 1990.3 The assets the macroeconomic situation was unsustainable. Infla- of the enterprise that formed the base for the tax were tion was already 100 percent at the start of 1989, and in adjusted for inflation, and the link between the tax and the last semester of that year it hit 2,000 percent. current profits was eliminated. Non-payment of the tax When the new Polish government took power in Sep- now triggers rehabilitation (similar to bankruptcy pro- tember 1989, it responded immediately with a policy of ceedings except that the government as owner initiates steady devaluation of the zloty. In January 1990, it imple- the rehabilitation). The slashing of the subsidies, con- mented extensive stabilization and liberalization mea- traction of bank credit and strengthening of the threat of sures to reverse Poland's soaring inflation and push the bankruptcy proceedings together hardened the budget 159 constraint for state-owned enterprises and forced an im- mand and uncertain CMEA trade; and they established a provement in resource reallocation. system of inter-firm notes in response to the severe cred- The impact of the Big Bang at the aggregate level was it squeeze, as reflected in a sharp rise in receivables and generally in the direction intended by policy-makers, al- payables. though the disintegration of the CMEA trading system It also seems that firms were suffering from a vacuum intensified the negative shocks on demand. Inflation (as of ownership. However, the partial self-management did measured by the Consumer Price Index [CPI]) fell to a not appear to be a source of perverse behavior, perhaps monthly rate of 5 percent by March 1990 and stayed because the repression of the nominal wage as part of the there until September. Industrial production dropped government's stabilization package significantly con- more sharply than expected, contracting by 30 percent strained the power of the Workers' Councils, the repre- in the first five months of the year (compared with the sentative bodies elected by the workers in a state-owned same period in 1989). The industrial subsectors that ex- firm,4 and the trade unions. perienced the largest contraction were light industry, The numerous and dramatic changes in policy out- mining, food processing and minerals. At the same time, lined above drastically altered the environment in which registered unemployment rose sharply to 107,000 peo- Polish enterprises were operating. For one, the budget ple by the end of February and 500,000 by the end of May. constraints they faced were significantly hardened by the Real incomes fell by 35 percent from January through removal of most subsidies and the tightening of credit, May. On the other hand, the trade account improved as well as by the increased threat of enforcement of the greatly, despite the problems with CMEA partners. Total bankruptcy procedures. The harder budget constraints exports increased by 8 percent in real terms in the first tended to encourage a reallocation of resources from un- five months of 1990 as compared with the same period in profitable to profitable firms. In other words, as of 1990, 1989. The increase was driven by exports to hard curren- the profitability of enterprises mattered. For another, the cy countries, which rose by 15 percent, whereas exports liberalization of trade and devaluation of the exchange to the CMEA remained unchanged. Over the same peri- rate altered the relative profitability of the sectors and od, imports fell by 29 percent in volume, split between a together shifted relative prices in favor of tradables. The 32 percent contraction in imports from CMEA countries removal of the subsidies and the general freein of prices and 27 percent from convertible currency areas (World also changed relative profitability across firms and sec- tors. The generally worsening demand conditions had a The harshness of this contraction of the economy differing effects, since the CMEA market, formerly the succeeded in squeezing inflation from the system. The sere of since Almost foserly the important question is when a supply response will occur. most secure of selling places, almost closed and the do- This issue of the resumption of growth has crucial polit- mestic market, especially for investment goods, col- ical dimensions, since a lengthening and deepening re- lapsed. By process of elimination, the hard currency cession eventually wears down the patience of any export market becamethe strongestandsteadiestsource electorate. Over time the political costs of worsening of demand for Polish goods. economic conditions are high, especially in a country This shift in relative profitability (driven by changes such as Poland, where employment has always been se- in the structure of prices and in demand) should have cure and where many social benefits remain tied to em- encouraged the reallocation of resources toward newly ployment. profitable sectors. More specifically, it would have been expected that firms would lay off labor to cut costs and The Effect of the Big Bang Policy Package at therefore that labor and other resources would be freed the Firm Level to be reallocated across sectors to match the new relative prices. Some firms would go bankrupt, a move that Despite the great uncertainty, the firms actively ad- would free capital assets for reallocation, while others justed to the triple shock of the liberalization of the would sell off some plants and equipment, a step that product market, depressed domestic demand and col- would also encourage the reallocation of capital. Since lapse of CMEA demand for Polish exports. Since they en- the real devaluation of the zloty was large, in general tered 1990 with relatively large financial cushions and there should be a shift of labor and capital toward trad- since wages were not the dominant portion of their pro- able activities. Profit margins would fall since the import duction costs, they had made only modest cuts in their competition and perhaps new domestic competition (es- work forces. Moreover, they took assertive action in two pecially at the level of distributors and retailers) would areas: they energetically and creatively pursued new ex- severely curtail the ability to pass prices on to consum- port markets in response to the decline in domestic de- ers. As described below, an examination of the behavior 160 of industrial firms in the small sample corroborates With this obvious emphasis on market conditions, the these expectations. firms that were performing best were, as expected, ori- ented toward the hard currency export markets. The Responses of Industrial Firms to the Big The sample of nine firms entered 1990 in quite a Bang: Some Findings from a Field Trip strong position, as many Polish firms did, having earned high profits relative to sales in 1989-the average was 42 The average firm in the sample is significantly larger percent for the nine firms in the sample (figure 12-1). By on both counts than the average firm in all of industry. comparison, the Lista 500 firms displayed profits of 34 The sample firms on average also had significantly high- percent of sales and industry as a whole, 31 percent (Ta- er overall exports as a share of sales and somewhat high- ble 12-1). Oligopolistic market power may explain part of er profits in 1988 and 1989. Thus, the sample does not that pattern, but far more important was the near-hyper- consist of extraordinary firms (by these sample mea- inflation of the last quarter of 1989, which drove their sures). profits up. Profit as a share of sales doubled between 1988 and 1989 for all three groups of firms. The firms An Overview of Shocks andResponses at the Level of the had little net debt (because of gifts of capital, the erosion Firm of their bank debt by inflation, and transferring to the government foreign exchange losses on foreign debt). The behavior of firms and their responses to the com- (Table 12-2 shows an average ratio of total liabilities to bination of changed government policies and other eco- equity of well under two for 1989.) The amount of the nomic shocks are discussed in general and with respect government-mandated dividend tax (a levy on the origi- to the following areas: marketing and distribution; ex- nal capital of the firm) was modest relative to profits, av- porting; cost-cutting measures; and credit and financ- eraging 2 percent of sales in 1989. The reasons for the ing. low level of payments were past retention of earnings and the fact that the Basic Assets Funds (the original The Response in General capital of the firm) had not been revalued in line with in- flation. Further, the Enterprise Funds (investments The responses of firms in the sample reflected what is made from retained earnings) were large relative to the known of the industrial response overall. Output and Basic Assets Funds. The share of wages in costs was employment decreased, with the drop in output greater low-13 percent of sales. As such, while firms reported than that in employment, as would be expected in firms experiencing a serious liquidity squeeze starting near long used to hoarding labor in a socialist setting and re- the end of 1989, most had a deep cushion against bank- luctant to or unable to fire workers. More interesting are ruptcy as a result of retained profits in the past. the differences in the condition of the firms, in their as- Dupnn the interviews, the firms indicated that their sessments of the shocks that were most important to Duig teitriw,tefrsidctdtahi sessm,eantsd their choieksf respon, wespecy msto i rtat con output for the first six months of 1990 was running 20- them, and their choice of response, especially to the con- 30 percent below that of 1989 on average, although there tinuing uncertainty. was considerable variation: some firms were working at Managers were asked which were the most important no more than 50 percent of capacity while others, nota- external shocks to industrial firms since the "Big Bang" bly in garments and food processing, were running near reforms from a list of issues that included price liberal- ization, greater import competition, the breakdown in CMEA trading arrangements, the lack of a formal distri- Figure 12-1. Net Profit Before Tax bution system, the depressed domestic demand, the (average for sample firms) tight bank credit, the irregular inputs, the uncertain property and ownership rights, and the general uncer- 5 tainty about economic conditions in the near future. In the case of firms that were ailing, the managers identi- fied demand conditions as the overriding problem with 3 3 which they had to cope. Managers of firms oriented to- 25 ward the domestic market saw the depressed domestic 20 demand as crucial. In contrast, the electronics firm, be- 15 cause of its heavy orientation to the CMEA, said it was 10 suffering mainly from the evaporation of CMEA orders. 5 i The other issues were often seen as irritants and difficul- o - ties and not as fundamental to the survival of the firms. 1988 1989 1st Qtr 1990 Jan-May 1990 161 Table 12.1: Characteristics of the Sample of Polish Industrial Finns Net profit Sales Employment Tbtal exports Exports to Zone 11b before taxC Finn Main product Sector (millions US$) (thousands) (% of sales) (% of sales) (% of sales) 1989 1989 1988 1989 1988 1989 1988 1989 State-owned firms 1 Machine tools Engineering 4 0.5 n.a. 69 n.a. 64 38 132 2 Computer printers Electronics 27 3.1 67 80 0.1 0.2 31 41 3 Electric generators Engineering 13 3.3 10 15 0.5 3 26 33 4 Basic chemicals Chemicals 77 6.2 27 26 22 23 17 36 5 Textiles Light industry 18 3.8 7 11 4 7 10 32 6 Garments Light industry 8 2.4 n.a. 19 n.a. 12 23 36 7 Trucks Transport equipment 76 13.8 4 4 0.8 0.4 12 16 State-owned cooperatives 8 Food processing Food industry 63 8.7 81 83 62 73 14 19 Private joint ventures 9 Wood furniture Wood and paper 44 5.5 31 21 31 21 16 32 Averages for firms in sample 37 5.2 32 36 17 23 21 42 e Average for Lista 500 firmsd 31 3.2 21 21 n.a. n.a. 16 34 Average for all industrial firmse 4 0.6 18 18 11 12 15 31 n.a. Not available. a. Polish zloty have been translated to US$ at average exchange rate for 1989 of 3502 ZL/US$. b. Exports to Zone 11 are exports to hard currency areas. For ALL INDUSTRIAL FIRMS, reported govemment number may be too high due to conversion problems. c. Net profit before tax is before company profits tax, excess wages tax and dividend tax. It includes non-sales income and extraordinary profits and losses. d. Lista 500 firms are Poland's largest 500 state-owned firms ranked by sales, excluding mining. e. All industrial firms are state-owned firms and include fuel and power as well as industry. Sources: Interviews with firms; and data from Central Statistical Office of Poland, Central Planning Office of Poland and others. Table 12.2. Overallperformance and the structure of costs for the sanple of Polish industrial finrs Sales growtha Net profit before tax" Change in employmentC Total costsd Firm Main product (in %) (% of sales) (in %6) (%6 of sales) 1989 1990 1990 1989 1989 1990 1990 1989 1990 1990 1989 1990 1990 1st qtr Jan-May 1st qtr Jan-May Ist qtr Jan-May 1st qtr Jan-May State-owned firms 1 Machine tools 41 9 25 38 132 42 37 -8.0 -4.0 -4.0 41.7 60.4 73.5 2 Computer printers 18 -1 16 31 41 36 35 -21.8 -3.7 -6.3 55.4 62.9 74.1 3 Electric generators 15 19 68 26 33 16 21 -6.3 -6.2 -9.6 80.3 71.3 65.6 4 Basic chemicals 22 18 55 17 36 27 23 -2.7 -1.5 -5.3 77.5 83.0 79.4 5 Textiles 21 -6 11 10 32 15 4 -10.8 -9.2 -17.8 76.2 112.5 128.8 6 Garments 30 n.a. 8 23 36 n.a. 22 -1.6 n.a. -0.4 53.4 n.a. 99.5 7 Trucks 20 -17 -11 12 16 111 7 -4.5 -4.0 -18.3 63.6 84.6 102.4 State-owned cooperatives 8 Food processing 33 3 34 14 19 7 7 -1.1 -1.2 -1.2 139.8 87.1 77.9 Private joint ventures 9 Wood furniture 61 n.a. n.a. 16 32 n.a. n.a. n.a. 18.2 18.2 67.8 n.a. n.a. Averages for firms in sample 29 4 26 21 42 22 20 -7.1 -1.5 -5.0 72.8 80.3 87.6 Average for Lista 500 firmse 26 n.a. n.a. 16 34 n.a. n.a. -8.4 n.a. n.a. 64.8 n.a. n.a. Average for all industrial firmsf 26 7 n.a. 15 31 n.a. n.a. -3.0 -5.9 n.a. 68.2 n.a. n.a. (continued) Table 12.2. Overallperfonnance and the structure of costs for the sample of Polish industrial firms (continued) Wages Material costs g TaXesh Dividend taxi Firm Main product (% of sales) (% of sales) (% of sales) (% of sales) 1989 1990 1990 1989 1990 1990 1989 1990 1990 1989 1990 1990 Ist qtr Jan-May Ist qtr Jan-May Ist qtr Jan-May 1st qtr Jan-May State-owned firms 1 Machine tools 12.1 18.1 8.4 23.8 38.0 46.2 7.3 2.8 6.2 2.7 4.1 1.3 2 Computer printers 7.9 10.0 7.5 36.7 47.8 52.5 6.6 15.0 14.7 1.3 0.8 0.6 3 Electric generators 19.4 12.7 9.2 38.1 38.3 38.2 13.8 8.5 9.8 5.0 0.7 3.9 4 Basic chemicals 7.2 4.8 3.6 59.2 70.1 66.4 12.0 12.5 11.1 2.3 3.6 2.0 5 Textiles 13.7 12.4 13.7 51.4 86.1 82.2 16.9 20.9 18.1 2.7 5.6 4.0 6 Garments 17.6 n.a. 19.8 23.5 n.a. 60.9 29.6 n.a. 25.4 0.9 n.a. 1.1 7 Trucks 11.4 15.4 19.8 40.8 45.7 56.0 22.9 16.9 23.0 1.0 3.4 2.9 State-owned cooperatives 8 Food processingi 10.3 7.3 7.6 102.0 19.6 20.8 4.2 2.2 2.3 Private joint ventures 9 Wood furnitures 14.2 n.a. n.a. 31.2 n.a. n.a. 10.6 n.a. n.a. -- -- -- Averages for firms in sample 12.6 11.6 11.2 58.1 43.2 52.9 13.8 11.3 13.8 2.3 3.0 2.3 Average for Lista 500 firmse 8.2f n.a. n.a. 57.7 n.a. n.a. 26.0 n.a. n.a. 1.9 n.a. n.a. Average for all industrial firmsf 12.0 8.7 n.a. 43.9 n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. Not available. a. Sales growth is % change in sales from beginning of year measured in US$, using an average exchange rate of 3502 ZIJUS$ for 1989 and 9500 ZL/US$ for 1990. b. Net profit before tax is before company profits tax, excess wages tax, and dividend tax. It includes non-sales income and extraordinary profits and losses. c. Change in employment is % change since beginning of year. d. Total costs are material costs (including cost of raw materials and energy, other material inputs, maintenance and servicing, subcontracting, and depreciation) and non-material costs (including wages, payroll taxes and other overhead on wages, interest paid, property taxes and import duties). e. Lista 500 firms are Poland's largest 500 state-owned firms ranked by sales, excluding mining. f. All industrial firms are state-owned firms and include fuel and power as well as industry. g. Material costs include cost of raw materials and energy, other material inputs, maintenance and servicing, subcontracting and depreciation. h. Taxes include tumover tax, company profit tax, excess wage tax, payroll taxes, property taxes and import duties. i. Dividend tax is a levy on the original capital of the firm. j. These firms pay no dividend tax. Sources: Interviews with firms; and data from Central Statistical Office of Poland, Central Planning Office of Poland and others. full capacity. This general contraction does not show up * the setting up of factory shops that sold at very in the simple measures of sales growth for the first quar- low margins; ter and for May of 1990 (table 12.1). One reason is that * the sponsoring of new distributors, for example, sales are deflated only by the exchange rate; another is of clothing, to bypass the large state distributors that that many firms have strong seasonal patterns for their were forced in 1990 to widen their margins to finance sales. high and slow-moving inventories; In addition, production in many industrial branches * special auctions and attendance at exhibitions and (e.g., for the maker of large electrical generators) is for- fairs; ward-looking, based on orders for future delivery; thus, * a searching for foreign firms that could provide production will fall more quickly than will current sales. marketing expertise through a joint-venture arrange- The pattern of average inventories as a share of sales pro- ment (as well as the provision of technology and capital); vides some insight as well. (See table 12-2). From 21 per- * an increase in the quality and appeal of products, cent of sales in 1989, average inventories of materials for example, by importing certain critical components and products soared to almost 60 percent during the from the West or acquiring licenses for high visibility first quarter of 1990, in the after-effects of the high infla- brand-name products. Some of the firms were planning tion of January and February. For the truck manufactur- to install new capacity to meet the perceived changes in er, inventories of materials soared upwards to 400 demand, although they indicated that it was hard to as- percent between 1989 and the first quarter of 1990. At sess new investments in the rapidly changing situation. the same time, inventories of finished goods jumped by over 2,500 percent (but constituted less than 5 percent In some cases, these efforts to develop a marketing of total inventory). Average inventories for all the sample strategy seem to have been quite effective. For example, firms fell after March, to average 46 percent for the first the truck manufacturer interviewed had begun to sell 80 five months of 1990, as firms learned to economize on percent of its output to individuals or small clients for stocks o cash, whereas such clients had previously accounted for s a cose no more than 20 percent of sales. The garment-maker As a consequence of the price liberalization and tight etmtdta twsitouigsm 0 ifrn fiscal and monetary policies since January, domestic de- estimated that it was introducing some 500 different mand fell sharply, especially for firmns producing invest- garment products per year, compared with a few dozen a ment good . The inteciews s t h fir.sts year ago, including high-value lycra bathing suits made ment goods. The interviews suggested that the first six under license. This behavior was a dramatic change from months of economic reforms were a period of turmoil. udrlcne hsbhvorwsadaai hnefo mons oa product line previously dominated by low-value stan- Except in the case of agricultural goods, which had been dardized products, for which demand had fallen sharply. freely traded since mid-1989, the markets were volatile However, some firms that made significant efforts in and highly segmented for the first quarter of the year. marketing and distribution found they were to no avail. Firms found it hard to estimate demand and to price in- For example, the electronics firm, most of whose market puts and products, in part because the border prices pro- was in the USSR, began contacting past customers to sell vided an imperfect guide. In addition, in some cases the directly and was planning a junket across the USSR to supplies of inputs became more uncertain, mainly be- demonstrate its product line to enterprises that already cause of the disruption of the state-controlled distribu- had some of its equipment. It was clear this bold move tion channels and the problem of finding a replacement was unlikely to pay off. Similarly, the textile producer for a troubled or uncooperative monopoly domestic sup- was suffering from the strong competition for imports in plier. New private distribution channels and markets the depressed domestic market. It tried to coordinate were in the nascent stage of development. closely with garment-makers to increase specialization and customization, but their costs remained far too Marketing and Distribution high. Again, success seemed linked to previous experi- ence with the hard currency market. The behavioral response of the firms in the sample to the creation of markets was extensive. The steps they Export Response took in the areas of purchasing, marketing and distribu- tion to adjust to the post-reform environment included: A number of factors affected CMEA demand for Polish products. The appreciation of the zloty against the trans- * the creation of sales departments, sometimes with ferable ruble relative to the zloty rate with the US dollar travelling salespeople, and the development of more di- lowered the profitability of the negotiated prices for ex- rect links with clients, especially abroad. ports to the CMEA. In some cases, the exclusion of prod- 165 ucts from the inter-government trade protocols and the pressed demand elsewhere, but also from an awareness elimination of export subsidies also diminished profit- of capitalist ways of doing business. ability. In other cases, such as the manufacturer of a computer printer, the economic reforms turned exports Cost-Cutting Measures: Employment, Wages and the at previously agreed prices into loss-making sales, while Costs of Material Inputs attempts to negotiate price increases under the con- tracts met with resistance from foreign purchasers. By .Because many firms in the sample had cut production mid-year firm s with resiotance producteion c surchase. t significantly and were pessimistic about the outlook for mid-year, firms with long production cycles, such as the sales, they attempted strong cost-cutting measures. Giv- producer of electric generators, faced the immediate ef- en the sharp increases in the cost of inputs, notably pow- fects of the failure to negotiate the next five-year cycle of er and water, and the progressively more competitive CMEA agreements to begin in 1991. In addition, the gen- product markets, which limited the ability of firms to eral uncertainty, especially in the German Democratic pass the higher costs on to customers, their margins Republic, led to the suspension of orders for production were squeezed in 1990. As a result, as shown in figure 12- by Polish firms. 2, their profits before taxes as a percent of sales, which With respect to sales, the firms continued to use the had doubled from 1988 to 1989, fell from an average of foreign trade enterprises (FTEs), state-owned firms that 42 percent in 1989 to 22 percent for the first quarter of had monopolies over exporting, despite the strong 1990 to 20 percent for the first five months of 1990. movement away from domestic state-owned distribu- These profits include extraordinary profits and non-sales tors. The FTEs are likely to have a continuing role in ex- income, such as sales of assets and of inventory. By using porting to both Zone I (CMEA) and Zone II (hard the total costs of production for comparison, it can be currency) markets: managers perceived some FTEs to be calculated that in 1989 the firms reaped profits not di- operating effectively and providing good quality service rectly related to production of 15 percent of sales. This for the sales margin they demanded. This opinion con- share fell to 8 percent in the first five months of 1990, trasted strongly with the generally poor perception of but the textile firm (with extra profits of 33 percent of the usefulness of the former monopoly domestic distrib- sales) and the garment manufacturer (with extra profits utors. It may be that the FTEs possessed scarce and valu- of 23 percent of sales) in this period remained dependent able information and skills that Polish firms typically did on income sources other than sales of goods for their fi- not possess. On the other hand, those managers inter- nancial health (table 12-3). Also of interest is that the viewed during the field trip believed that the percentage ranking of the firms in the sample by profit rate does not sales commissions the state intermediaries charged for change much from 1988 to 1989 to the two points in some foreign and almost all domestic sales were too 1990, partly because of sectoral differences between the high, even though the rate was generally around 5 per- firms but also partly because of the general quality of cent of the sales price. These commissions seem modest each firm. by maktecnmnsadrd,apinhtugstht Policy-makers anticipated that a response of firms to ,y market economy standards, a ponmt that suggests that the Big Bang would be to fire labor. Although the data on Polish firms were not accustomed to the substantial idsra mlyetaedfiutt nepe,age costs of promotion and sales. At present, Polish firms see idsra mlyetaedfiutt nepe,age -c ofpromotion a *ndusls.A prese polish frmss gate employment (i.e., mainly in state-owned enterpris- such expenditures as unproductive andaspossibleareas es) declined by around 3 percent in 1989 for the for cost-savings rather than as a priority for increased in- industrial sector as a whole, while Lista 500 firms re- vestment. duced their labor forces by 8.4 percent. In general, the firms interviewed believed that ex- During the same time, the enterprises in the sample ports to the West were more profitable than those to reduced employment 7 percent. With these significant CMEA partners, that Western markets had superior po- cut-backs in 1989, it is not so surprising that firings in tential for growth and that such exports would be a cen- the first semester of 1990 were modest. For the sample tral part of their strategy for survival. Virtually all of the firms, employment declined by another 2 percent in the firms had had some exposure to Western markets before first quarter of 1990 while for all of industry, employ- 1990 (although sometimes indirectly through the ment fell by 6 percent. By the end of May 1990, the firms FTEs). The interviews strongly indicated that those had laid off about 5 percent of the labor force they had at firms with more prior exposure to Western firms had a the beginning of the year (table 12-3). Clearly, firms are definite advantage in formulating a viable strategy for being forced to lay off workers. Initially, however, they survival. This advantage seemed to come partly from the are doing so by laying off casual workers and those close relative attractiveness of hard currency exports given de- to retirement. In light of the nearly 20-30 percent de- cline in output, the 7 percent decline in the labor force 166 Table 12.3. Financing by the sample of Polish firms Total liabilitieslequity Average receivables' Average payablesb Receivables/payables Average inventories' Firm Main product (ratio at end of period) (in weeks of sales) (in weeks of sales) (ratio at end of period) (% of sales) 1988 1989 1989 1990 1990 1989 1990 1990 1988 1989 1990 1990 1989 1990 1990 Ist qtr Jan-May Ist qtr Jan-May March May March May State-owned cooperatives 1 Machine tools 0.53 1.47 20.4 8.7 7.2 8.4 2.1 2.7 1.7 2.4 11.1 3.4 21 39 36 2 Computer printers 0.93 1.12 9.3 4.6 4.6 3.8 2.0 3.3 2.1 2.5 2.1 1.0 25 44 40 3 Electric generators 5.20 1.50 8.0 6.2 6.8 7.1 4.2 5.0 2.3 0.9 1.7 1.4 36 57 40 4 Basic chemicals 1.12 1.43 10.9 5.0 5.3 6.3 1.9 2.6 1.7 1.7 3.5 2.2 8 20 13 5 Textiles 0.93 2.07 11.0 9.9 10.9 7.8 8.2 10.5 1.2 1.4 1.1 0.9 23 70 54 6 Garments 0.27 2.70 7.0 n.a. 7.8 1.9 n.a. 5.9 3.9 3.7 n.a. 1.0 12 n.a. 40 7 Trucks 0.99 2.10 9.0 5.8 8.0 7.7 9.3 16.5 1.3 1.1 OA 0.3 20 110 100 State-owned cooperatives: 8 Food processingd -- -- 9.3 4.7 4.3 3.4 1.3 1.4 2.5 2.7 4.6 3.3 50 73 38 Private joint ventures: 9 Wood furniture 4.22 3.13 32.0 n.a. n.a. 26.4 n.a. n.a. 1.1 1.2 n.a. n.a. 15 n.a. n.a. Averages for state-owned firms:e 1.42 1.77 10.8 6.7 7.2 6.1 4.6 6.7 2.0 2.0 3.3 1.5 21 57 46 n.a. Not available. a. Average receivables are the average of beginning of year and current receivables, expressed in weeks of sales during the period. b. Average payables are the average of beginning of year and current payables, expressed in weeks of sales during the period. c. Average inventories are the average of beginning of year and current inventory stocks, as a % of sales during the period. d. No comparable equity measure was available for these firms. e. Averages for state-owned firms exclude firms 8 and 9. Sources: Interviews with firms; and data from Central Statistical Office of Poland. is a clear indication that reducing labor and labor costs strengthened in January, was a powerful constraint on is the last resort. raising wages and bonuses,6 partly because of its finan- Despite the lay-offs, for a number of reasons there was cial cost to the firm and partly because payment of the still overstaffing. In the eyes of managers, employees val- tax was one of the indicators prompting government re- ued job security very highly. Wages were a secondary, al- view of the firm's condition. The tax was designed to of- though still important, consideration. Desire for fer an incentive to reduce the labor force, meaning the ownership of the firm ran a distant third. Despite this Wages Fund would be shared among fewer employees. In worker concern with job security and the legal con- the past, however, this has not seemed to be a major in- straints on group firing, employment did adjust, with an centive to lay off employees, although there are indica- average decline of 7 percent since the beginning of the tions that companies are now considering lay-offs to year for the firms in the sample. Some firms reduced permit higher pay. their labor significantly: since January both the textile A parallel process was occurring on the input side, firm and the truck manufacturer eliminated 18 percent with firms searching for more diversified suppliers. This of their labor force. At some firms negotiations were un- search was not easy because of the market dominance of derway to reduce staffing even further. Workers ap- a few producers and the higher costs of many imports. In proaching pensionable age and casual workers appear to addition, firms eagerly revised the nature of their con- have been the first to be laid off. The private wood furni- tracts with suppliers, since they no longer desired the ture firm stands out as the only firm taking on new labor long-term and inflexible contracts common and attrac- in 1989 or 1990. tive in a shortage economy. Potential import competi- The cost of labor also fell as a result of lower wages - tion was a factor in holding down prices, but in many real wages declined quite substantially. This drop was cases domestic goods were still cheaper than close for- offset in part by the elimination of the shortages of goods eign equivalents. This continuing gap offered a breath- and related queuing. By way of comparison, wages at the ing space to some firms hard-hit by the abrupt transition firms in the sample remained constant as a share of sales from virtually full protection to almost no protection. through the first five months of 1990 (figure 12-2 and ta- For example, steel prices were said to have been lower ble 12-3). However, the ratio of wages to profits rose: the than international prices for comparable qualities, a sit- first five months showed an average ratio of wages to uation that provided some steel-using firms in the sam- profits of .56, up from .30 in 1989 for the sample. The ple, such as the machine tool maker, with a competitive most dramatic increases in this ratio occurred with the advantage. textile manufacturer and the truck maker, followed by the food processor and the garment maker. The excess Credit and Financing wages tax, in existence through the 1980s5 but greatly The credit squeeze that began in the latter part of Figure 12-2. Cost Structure of Production 1989 affected the balance sheets of even the most profit- (average for sample firms) able firms. Since the firms in the sample were not heavi- 1W ly in debt, however, in most cases the rise in interest 1 _ rates did not in itself seriously damage profits. An excep- 9Oo -0 77777 tion was food processing, which faced high costs for credit because of the seasonal production cycle. In general the firms responded to the tight credit in 60 - . ..i ....several ways. For example, the manufacturer of electri- cal generators, fearing that the high rates of interest in s50 40 -January and February 1990 would continue, secured 30 - prepayment from a large client and paid off all its debt. 20 - More commonly, however, firms experienced a sharp rise lo in payables, which was mirrored in rising receivables. 0 _ For the firms in the sample, this pattern became evident 1989 lstQtr 19 Jan-Mays199 between March and May of 1990 (figure 12-3 and table 12-2). Average payables stood at six weeks of sales for the Othercastsandtaxes state-owned firms in the sample in 1989, fell to five IMateria costs weeks during the first quarter, and rebounded to reach Wages seven weeks for the first five months. All of the sample firms experienced a rise in receivables between March 168 Figure 12-3. Average Receivables Plus Payables The Credit Crunch (average for samp[e state-owned firms) 17 Credit from the banking sector was tight and interest 16 rates high, a central feature of the Big Bang's anti-infla- 15 tionary focus. It was expected that the cut-off of cheap 13 4 credit would push inefficient firms toward bankruptcy 12 - and encourage credit to be directed toward the most i 10 creditworthy borrowers. However, firms evaded this con- O 9 _ straint through substitution and innovation. Anecdotal 7 8 evidence supports the claim that banks simply chan- ; 6 - neled the remaining credit to their biggest borrowers. 9 5 _ Other firms were highly credit-constrained, although 3 many did not face an immediate danger of bankruptcy 2 since their profits were so high in 1989. Thus, liquidity I rather than solvency was the problem. 0 1989 st Qf 1990 Jan-May 1990 The managers agreed in the interviews that the prior- ities of their firms in making payments were, first, the e , , dividend tax due the government, second, other taxes E | Payables | | Receivableg and wages, and, last, the payables due suppliers, since it was possible to stretch these liabilities out with no real and May of 1990. At the same time, average receivables penalty. Thus, suppliers' credits were the buffer in the amounted to 11 weeks of sales in 1989 and then stabi- system, allowing firms a margin of adjustment during lized at around seven weeks during the first five months. the credit crunch. The firms in the sample kept ahead of the inter-firm credit game by delaying their own pay- Did the Big Bang Create a New "Sink or Swim" ables while accelerating their own receivables. This Environment for Firms? strategy reduced their creditworthiness but generated working capital. The end-of-period ratio of receivables to The Old Regime for State-Owned Enterprises payables rose from 2.0 at the close of 1989 to 3.3 in March 1990. It then fell dramatically to 1.5 by the end of There has been a general withdrawal of government May (table 12-2). The declining excess of receivables over from the active ownership of firms. The decentralization payables illustrates the credit generated by this behavior. of control in the 1970s had still left the government with The use of payables and receivables as a mechanism financial control over firms. In the 1980s, however, the for financing inter-firm transactions was not new. So- government paid a large portion of the direct subsidies cialist firms were notorious for accumulating unpaid in- to the citizenry rather than to the enterprises. For exam- ter-firm commitments. Prior to the January reforms, the ple, in 1988 enterprises received only one-third of the to- payables plus receivables for the firms in the sample tal subsidies, divided into two types: compensation for were around 17 weeks equivalent of sales (figure 12-3). distortions in the pricing of output, a measure that con- This ratio dropped in the first quarter of 1990, and the centrated on the coal mining sector; and transfers to ratio of payables to receivables rose. The implication is loss-making firms, a measure that was focused on the that once the firms were forced to be responsible for food industry and the fuel and power sectors. More im- their own cash flows, they probably insisted on some portant for industrial firms in other sectors was the ef- clearing of inter-firm accounts and improvement of fect of the indirect subsidies created by a complicated their own financial position. By the end of May 1990, vol- system of subsidized credit and foreign exchange. Schaf- ume of suppliers' credits had risen, and the ratio of re- fer (1990) documented the softness of the budget con- ceivables to payables had fallen sharply, a change that straints on the Lista 500 firms as a result of government suggested a growing need for credit. Managers reported policies, including firm-specific subsidies and tax re- both that bank credit was generally unavailable and that bates. The removal of the subsidies in 1990 eliminated they expected few or no losses from their own debtors, the last shred of active government involvement except on the assumption that if the govemment pushed the for the ongoing attempt to harden the budget constraint debtors into bankruptcy, it would take on responsibility through bankruptcy proceedings. for the bankrupt firm's liabilities. 169 A more interesting analysis emerges if the sample run firms saw the future of the CMEA markets as highly firms are split into two rough categories: firms deemed uncertain. Firms had doubts about receiving payments, likely to fail; and those more likely to survive. The elec- especially from the USSR, that made them hesitant tronics firm, totally dependent on the CMEA market, and about fulfilling existing contracts and worried about col- specifically the Soviet market, the textile manufacturer lecting on receivables. According to the managers of the with outdated machinery, few exports and buildings pro- sample firms, for the CMEA to become a viable alterna- tected by historic landmark status, and the truck maker, tive market more quickly, trade would have to be con- also with little export experience and growing import ducted at international prices, the rate for the ruble competition, and with soaring payables and inventories, would have to be adjusted and the uncertainty of pay- were clearly having trouble. The other six firms had ment would have to be resolved. Thus, in the short term comparatively good chances of surviving. These two they concentrated on investing in developing Western groups had different patterns of the ratio of receivables markets. to payables. All the firms showed a rise in this ratio for the first five months of 1990 as compared with the first Exit Policy for State-owned Enterprises quarter, but only the three failing firms also showed an increase for the first quarter as compared with 1989. Progress was made in defining more clearly the exit That is, the failing firms were using suppliers' credits to policy for firms through the gradual implementation of generate working capital from the beginning of the year, the January 1990 bankruptcy law and the growing focus while the surviving firms delayed until April. A signifi- on privatization as a way to shift most of the decisions on cant development in 1990 was the transformation of exit to private agents. Bankruptcy has become more of suppliers' credits into negotiable inter-firm notes that an institution although the process is not yet completely cost far less than bank credit. The very tight credit credible. For example, the firms in the sample still be- squeeze notwithstanding, firms found an innovative way lieved there was an implicit government guarantee of in- to avoid the disruption of the squeeze. ter-firm credit. This inter-firm credit yielded some benefits. In the The dividend tax-the levy on fixed assets-involved face of the rigidities of the banking system, it allowed monthly mandated payments that each state-owned firm firms to maintain a margin of flexibility for production. had to make to the government as the provider of the This credit was also used to finance new distributors, original capital of the firm. Non-payment of the tax was which in turn made the markets more competitive. supposed to trigger recovery or bankruptcy procedures, However, continued growth without regulation poses an according to the January 1989 law on the finances of increasing risk of cascading bankruptcy. These worries state enterprises. However, not until the government re- were not new developments: toward the end of 1989 the affirmed this threat of rehabilitation of non-paying firms government was considering a scheme to net out such in January 1990 did the dividend payment become a top debt across all firms. priority of firms.7 Beginning in January 1990, the divi- dend became a fixed 32 percent of the capital base, with ThePressuretofmportCompetionandtheNeedto no link to profits or other performance measures. Be- cause of the high rate in 1989, the government also or- The new trade regime had a strong impact on the dered a revaluation of the Basic Assets Fund of the markets for both inputs and outputs and on relative pric- enterprises in January 1990 by a factor of 14. Despite the es. The fierceness of the import competition across sec- tax's name, the fixed nature of the payment made it less tors varied, but the low value and continuing like the capitalist dividend that is normally paid to share- depreciation of the zloty moderated competitive pres- holders and more similar to an interest payment on se- sure overall. However, the exchange rate alone cannot nior debt. Whatever the case, its importance is small for protect domestic producers. For example, the truck most firms: the dividend tax averaged only 3 percent of manufacturer was charging sales prices far below world sales in the first quarter of 1990 for the firms in the sam- prices for a truck of similar size, but since the quality of ple, and for no firm did it reach higher than 6 percent of its trucks was far lower, the firm continued to face strong sales. competition from imported used trucks from Western As mentioned, firms assumed the government would Europe. International competition came almost entirely provide a bail-out on inter-firm credit, although they from the Western countries. Although Western markets also believed they had to cut costs and increase sales to were hard to penetrate, they were what the firms focused survive. This sort of confusion will persist until an ac- on for survival. For Polish firms, the best market in the tive, functioning bankruptcy procedure is in place. medium term may well be the CMEA, but in the short Firms said they were expecting privatization to occur in 170 the near future (within a year or so) and that the game the pleasure of the Workers' Councils. All but two of the for them was to survive until then and to place them- sample firms had very recently changed their general selves in as favorable position as possible for privatiza- manager. tion, including making themselves attractive to foreign The actual role played by the Workers' Councils, partners. whether active or passive, depends, however, on the indi- viduals elected to them. All the firms visited described Ownership, Control and Privatization relations between management and the Workers' Coun- cils as good, with no serious disputes. They frequently A central and continuing problem for the long-run commented that the Workers' Council was helpful, per- adjustment of state-owned firms is clear definition of the haps because it provided a counterweight to the trade rights of ownership and control, a problem intertwined unions. It should be noted that by constraining pay, the with the move toward privatization. Ownership and con- excess wage tax has defused one of the main potential ar- trol of state-owned enterprises in Poland, as in most so- eas of contention among management, Workers' Coun- cialist economies, are subtle issues because of cils and trade unions.8 complications posed by the legal and institutional rules Based on the interviews with management, the Work- governing operations, the important implications of the ers' Councils did not seem to consider themselves to be current structure of ownership on enterprise behavior, owners, although they exercised some rights that were and the influence that de facto ownership and control usually reserved for owners. Typically, management might have in determining what is a fair settlement of doubted that workers wanted to be shareholders in the property rights during privatization. At the time of the firm where they worked. Ownership was seen as posing interviews, the new Privatization Law of July 1990, extra risks for Colon workers without corresponding which circumscribes the possibilities for transforming benefits, unless the firms were very secure and profit- state-owned property, was not in effect. able. Only in the machine tool firm, which had an excel- The firms were suffering from a vacuum of owner- lent capital stock and almost no liabilities, did workers ship, with an ill-defined system of control over the enter- express a desire to buy shares. Without exception, man- prise. The managers at a minimum clearly understood agement and workers assumed that the employee shares that the formal owner of the firm was the treasury. How- would have to be purchased (perhaps at some discount) ever, the control and supervision the Ministry of Indus- and would not be distributed free of charge. try once exercised had declined to zero, even in strategic The possibility of joint ventures with foreign firms industries. The role of local government had increased, has been one source of pressure for speedy resolution of but in very specific areas - zoning approvals, environ- the ownership issue: the managers of Polish firms view mental control and social benefits for employees. Those this option favorably and are eager to be able to commit areas did not include direct control over the financial to a joint venture. Foreign partners are seen as bringing condition of firms. The Workers' Councils exercised some particular advantages: market access and market- some control, but it was limited. In practice, the general ing expertise; technology and licenses; modern manage- managers of the firms had almost all the power over de- ment methods; and new capital. cision-making, except in a few enterprises with strong Some involved breaking up the firm into different en- unions that had created particularly activist Workers' tities to facilitate joint ventures. The search for foreign Councils. It should be noted that job security for em- partners, while not the sole element of adjustment strat- ployees was almost universally quoted as an objective of egy, was intense. Partners were perceived to offer advan- management, partly because of the influence of the tages of market access and marketing expertise, Workers' Councils and partly because some managers technology and licenses, modern management methods, were former members of Workers' Councils and/or trade and new capital. With ownership status unresolved, union activists. A further reason was that Solidarity was however, potential foreign partners were asking for gov- the ruling party. ernment guarantees on their investments. The terms of All state-owned firms have elected Workers' Councils the offers were not always favorable to the Polish firm, with formal powers covering all aspects of management. especially if the firm was doing badly and was in great The Workers' Council of a firm must approve the distri- need of foreign assistance. For example, the electronics bution of profits, the sale of fixed assets and the appoint- firm was negotiating a contract with a Western firm to ment of deputy directors. Most importantly, they must provide some new technology under license, some joint recommend the appointment of the general manager, production and some guarantees of distribution in West- based on nominations by the Ministry of Industry (with ern markets. However, the prospective Western partner final approval by the Ministry). Management serves at was not providing any capital and was demanding a very 171 high fee for its services. On the other hand, the electrical rewriting contracts with existing suppliers. They were generator maker was negotiating a very favorable ar- actively searching for new markets and new products. In rangement that resulted in the acquisition of 75 percent response to the tightening of credit in the first semester of the firm by a Western firm in the same business in Au- of 1990, the firms expanded suppliers' credits and creat- gust. Overall, the favorable view of joint ventures among ed inter-firm notes. managers of Polish firms, and managers' eagerness to be Some institutional features were influential in the able to commit to a joint venture, placed pressure on the firms' behavior. In those in the sample, the role of the government to resolve the ownership issue. Workers' Councils appeared to be positive; in some cas- With the ownership issue unresolved and privatiza- es, the Councils were playing a useful role for manage- tion not proceeding, state-owned firms nominally be- ment vis-a-vis the trade unions. This relationship could, longed to the Treasury while local governments however, have been a honeymoon caused by the excess exercised some regulatory powers, and management and wages tax, which was effective in curbing wage increases. the Workers' Councils co-managed most aspects of these The excess wage tax alleviated the wage pressure that firms' operations. Decision-making by the firms under would likely otherwise have been induced by the vacuum these circumstances was subject to complex consider- of ownership created by the decentralized control of ations and considerable uncertainty. firms, especially under continued inflation. However, de- spite the favorable effect of this tax, the ownership issue Conclusions was beginning to inhibit the implementation of longer term survival strategies, especially the pursuit of joint Management interviews and basic financial data from ventures with Western firms. Resolution of this issue nine Polish manufacturing firms in June and July of was growing in importance, because continued delay, 1990 provide insights into the behavior of state-owned despite the willingness and ability of firms to change firms in the first six months after Poland's "Big Bang." their behavior, could stall adjustment. The observations at the firm level are based on but a par- This view of firm behavior was derived at a time when tial picture of the response to the Big Bang. Some con- the transition to a market economy was far from com- clusions, however, do emerge. In particular, the plete. Uncertainty was extreme. A new tax structure, in- aggregate unemployment rate is not a good measure of cluding a personal income tax and a VAT, were not yet in the speed of adjustment of the industrial sector because place to replace the socialist tax structure that had been of firms' abilities and incentives to continue to absorb la- heavily dependent on revenues from state-owned firms. bor costs. In general, the extensive adjustment to the The social safety net required further development and changed conditions and long-term horizon exhibited in refinement. The process by which privatization was to the sample of firms reflect flexibility in their behavior, al- take place had not yet been fully specified. Most impor- beit not necessarily with assured successful outcomes in tant, the medium-term success of individual manufac- all cases. turing firms, which in aggregate can underlie a positive As noted, it is unclear if the firms in the sample are supply response necessary to revive the economy, evi- representative of the response in the industrial sector. dently required new investment, both foreign and do- They may well have been above average in their adjust- mestic. ment. However, in the aftermath of the demand and There was life after the Big Bang, at least for the sam- shock of the liberalization of the product market associ- ple of firms that were investigated, although the quality ated with the Big Bang, there is evidence of significant of life differed. However, a revival of investment awaits positive adjustment by state-owned firms in some indus- clarification of future uncertainties, in particular re- trial sectors under conditions that were unfamiliar and garding privatization, the system of financial intermedi- uncertain. Although some firms were failing, others ation, the role of the excess wage tax, and the future role were actually thriving in 1990. Overall, the best-placed of the Workers' Councils under a Solidarity government. firms appeared to be those with previous exposure to Notes selling in Western markets. Positive responses occurred in a number of areas. 1. See Schrenk (1990a) for a discussion of the workings of the Firms were trimming and rationalizing labor use, al- CMEA and the trade patterns that resulted. though, for the better firms, lay-offs were not necessarily 2. This phrase comes from Komai (1979) and is further developed a good indication of the degree of adjustment because in Kornai (1986). wage costs were relatively small and could be absorbed 3. In 1989 the government had introduced a statutory tax on the original assets of state-owned firms that it misleadingly called a "divi- by the generally high profit cushion. The firms were also dend," to be paid to the I'easury as the owner. cutting input costs by searching for new suppliers and 4. The Workers' Councils were introduced in 1981 as part of 172 Poland's first program of reform. They were given limited powers in References decisions of firms. 5. During the 1980s, as part of a continued effort to re-establish fi- Calvo, Guillermo, and Fabrizio Coricelli. 1990. "Stagflationary Effects nancial discipline over firms, the government imposed a series of taxes of Stabilization Programs in Reforming Socialist Countries: Supply on excess wages that took the general form of progressive taxes on Side vs. Demand Side Factors." Mimeo. Macroeconomics and wage increases above a government-mandated maximum. However, Growth Division, Country Economics Department, World Bank. frequent exemptions undermined the impact on wage levels. Washington, D.C., August 6. The Big Bang package included a measure to control wages by greatly increasing the excess wage tax. The tax was to be levied on firms Frydman, Roman, Stanislaw Wellisz and Grzegorz Kolodko. 1990. that raised wages by more than 30 percent of official inflation in Janu- "Stabilization in Poland: A Progress Report." Mimeo. A revised ver- ary 1990 and 20 percent in February, March and April. Frydman, Well- sion of a paper presented at the May 1990 Conference on Exchange isz and Kolodko (1990) argued that the excess wage tax was a binding Rate Policies of Less Developed Economies and Socialist Econo- ceiling on wage payments even though the government usually did not mies, Berlin. Department of Economics, New York University, New invoke the tax. The penalty arose if wages exceeded the mandated share York, May. of actual inflation. However, firms had to set the level of wages one month in advance based only on a forecast. For the first four months Kornai, Janos. 1979 "Resource-Constrained Versus Demand-Con- of 1990, the wage payments of firms slowly made up for the sigmficant strained Systems." Econometrica 47 (July):802-20. underpayment in January caused by the underprediction of inflation for that month by the Ministry of Finance. 7. The dividend tax was instituted in 1989 to reduce the role of bar- . 1986. Kyklos 39:3-30. gaining in the taxation of firms. It required firms to pay a tax on that portion of their capital stock which was funded centrally as opposed to Lipton, David, and Jeffrey Sachs. 1990. "Creating a Market Economy in that part of capital funded out of retained earnings. Specifically, the Eastern Europe: The Case of Poland." Brookings Papers on Eco- base for this taxis the enterprise's founding fund, or Basic Assets Fund, nomic Activity 1:75-147. defined as the total value of its assets in December 1983. The tax rate was linked to the bank rate of interest for re-financing, set by the Na- Schaffer, Mark. 1990. "State-Owned Enterprises in Poland: Taxation, tional Bank of Poland. The setting of an upper bound for the dividend Subsidization, and Competition Policies." A paper prepared for the payments-25 percent of profits-and the 75 percent reduction in pay- (PHARE) Project, DGII, European Commission. School of Europe- ments allowed for public utilities and agricultural sector firms com- an Studies, University of Sussex, Brighton, England, February. promised the uniform application of this levy. 8. The firms in the sample generally had two main unions, Solidar- Schrenk, Martin. 1990a. "The CMEA System of Trade and Payments: ity and OPZZ (founded by the Communist Party during the period of Today and Tomorrow." Discussion Paper No. 5. Strategic Planning martial law in response to the rise of Solidarity), with roughly equal and Review Department, World Bank. Washington, D.C., January. membership. Often there is a third small, non-party-affiliated union. The share of non-union workers was sometimes as high as 40 percent. . 1990b. "Poland: Reform of the Economic System." Mimeo. The scope of activity by the trade unions is legally confined to approval Socialist Economies Reform Unit, Country Economics Depart- of lay-offs and wage agreements. ment, World Bank. Washington, D.C., March. 173 I 13 Comments on "Life after the Polish 'Big Bang': Representative Episodes of Enterprise Behavior," by Erika A. Jorgenson, Alan Gelb and Inderjit Singh Jan Svejnar This well-motivated paper by Erika A. Jorgenson, Alan for their products and increase (or limit the decline in) Gelb and Inderjit Singh is an impressive analytical re- the prices of their products. This effort to adjust led to port on a visit to nine Polish industrial firms in the sum- the creation of marketing departments, the position of mer of 1990. It contains a lucid discussion of the initial traveling salesman, etc.-i.e., to labor-intensive mea- conditions of the enterprises as well as their responses to sures that were sensible per se and took into account the the major policy changes that took place in early 1990. primacy of job security as an enterprise goal. The firms To summarize, the salient feature of the initial condi- decreased their reliance on middlemen for domestic tions of the firms in the sample is that they entered the sales (an area where they could rapidly develop their own Big Bang period in good shape. Unlike their counter- expertise). They used foreign trade middlemen to stimu- parts in some other countries (e.g., Czechoslovakia and late exports to hard currency areas. Although real wages Hungary), the Polish firms had relatively high profits fell, the ratio of the wage bill to total revenue remained and low debt. They were also subject to a low capital tax, roughly constant. Employment declined by about 5 per- and their wage bills on average accounted for only 13 cent in the first six months of 1990 as casual workers percent of sales. All the enterprises had experience with were laid off and early retirements took place. The credit Western firms and markets. Moreover, managers and squeeze led to an outburst of inter-enterprise credit (in workers effectively shared power in the environment of the form of an increase in payables and receivables); en- poorly defined ownership and control rights. The main terprises expected this growing system of inter-enter- goal of the workers was job security, a goal that the man- prise financing to be sustainable. Finally, all firms agers shared, as the paper implies. The second goal of engaged in an intensive search for foreign partners. workers was the growth of wages, while ownership of en- The authors argue, and document convincingly, that terprise capital came third. the response of the Polish firms to the new environment The fundamental change in economic policies that was very substantive. They also make the case that the took place at the start of 1990 exerted strong downward response was in the "right direction." On the whole, they pressure on enterprise profits as demand fell and the are right on this point. However, it can also be argued cost of inputs (especially those of power and water) rose. that it is necessary to distinguish between adjustments With the old distribution system gone and a new one not that are socially desirable and those that benefit only the yet in place, supplies of inputs became more uncertain, firm. and export orders from the countries in the Council for The fact that there was a significant response by en- Mutual Economic Assistance (CMEA) collapsed. The terprises to the major change in exogenous factors con- elimination of the subsidies and imposition of a credit firms the findings about enterprise behavior in other squeeze constrained the cash flow of enterprises signifi- countries. Most studies in developing and socialist coun- cantly. Finally, the government imposed a stiff incre- tries reveal how enterprises exploit government-im- mental tax on the wage bill to forestall inflationary posed constraints and regulations to their advantage, in increases in wages. the process frequently generating socially inefficient The paper clearly documents the rapid response of outcomes. The present study documents responses to the nine Polish firms to this dramatic change in circum- government measures that were aimed at relaxing the stances. Instead of becoming overwhelmed and para- regulations and constraints. Hence, to a large extent lyzed, the enterprises took measures to generate demand most of the observed enterprise responses (e.g., mea- 175 sures to stimulate demand, find more lucrative markets growth in inter-enterprise credit. Both theory and evi- and use quasi-fixed factors such as labor more intensive- dence from other socialist economies (e.g., Yugoslavia) ly) were efficiency-enhancing from the social stand- suggest that participation by enterprises in this financial point. However, where the government imposed network extends the longevity of failing firms and tends constraints, the firms responded by exploiting these to to develop into a game between a large number of col- their (i.e., the owner-insiders') advantage. The incre- luding enterprises and the government. Insolvency on mental tax on the wage bill is a good example. It stimu- the part of a few firms then threatens to paralyze the en- lated lay-offs of casual workers and early retirements, as tire enterprise sector, a situation that poses a dilemma the insiders strove to increase their wages. Whether this for the government. The government's inability to pre- outcome was socially efficient or inefficient depends on vent the escalation of inter-enterprise credit may subse- the marginal product of labor inside and outside these quently force even conservative governments to bail out firms (the shadow wage), information that is not readily failing firms for fear of the large-scale repercussions of available. Similarly, there is no information on the other individual bankruptcies. likely impacts of this particular wage regulation - the Overall, the paper offers a very valuable microeco- loss of skilled workers and the negative effect on effort nomic diagnosis of the Polish economy in transition. It and work quality. Since one of the salient features of so- covers most major areas of the economic activities of en- cialism was the low level of individual effort, it can be terprises and interprets the main findings in a consistent surmised that the negative impact of the wage regula- conceptual framework. The paper leaves a profound im- tion on productivity could be significant. pression of the significant changes taking place and of It is also worth stressing the potentially very danger- the need to follow future developments closely. ous phenomenon the paper points out - the rapid 176 14 Distortionary Policies and Growth in Socialist Economies, William Easterly This paper discusses several applications of endoge- explaining the large differences in income levels, nor nous growth models in socialist countries. The next sec- should they be given much weight compared with, for tion reviews the evidence on distortions in resource example, stabilization policies (Rodrik 1990).3 allocation in socialist countries, particularly the Soviet The literature on endogenous growth, following on Union. The following section presents a model of how the work of Romer (1986, 1987 a and b, and 1988), Lucas distortions in resource allocation affect growth, discuss- (1988) and Barro (1989a and b), presents models in ing the model under the assumption of a fixed savings which policies can have significant effects on long-run rate, and then considering what saving rate a planner growth. These models variously relax the assumptions of might choose if the distortions are taken as given. In the constant returns to scale (Romer and Lucas), the depen- subsequent section, the model is applied to an example dence on non-reproducible factors (Rebelo 1990) and di- of distorted allocation of capital stocks in the Soviet minishing returns to each factor (Jones and Manuelli Union. The model is then generalized to consider the ef- 1990). Barro (1989a) and Barro and Sala-i-Martin (1990) fects of partial reform in socialist economies. Some con- have discussed how tax rates can distort savings deci- clusions are presented in the final section. sions and lower growth, while the government services financed by those taxes can potentially raise private pro- Introduction ductivity and increase growth. Rebelo (1990) and King and Rebelo (1990) have similarly shown how differences The apparently spectacular failure of the economic in tax rates can translate into large differences in growth system of central planning has led to a reassessment of rates. While much empirical testing of these models re- the system's long-run effects. Although estimates of so- mains to be done, the empirical work discussed in this cialist growth are highly uncertain, it is increasingly literature has tended to confirm strong effects of policy clear that it has been comparatively poor when the high on growth (Barro 1989b and Romer 1989). rate of investment is considered. However, according to the standard neoclassical Distortionary Policies in Socialist Countries growth model developed by Solow (1956), distortionary policies (and all other economic policies) affect only the Although the endogenous growth models strongly level of income and not its rate of growth. Given the ex- support the emphasis given to policies, so far they have istence of non-reproducible factors, constant returns to not paid much attention to the special characteristics of scale and diminishing returns to each factor, steady- policies in socialist countries. The salient fact about state growth can only take place through exogenous these policies is that they cause severe distortions in the technological change.2 On these grounds, the alleged allocation of resources. Prior to recent reforms, condi- long-run growth effects of socialist policy have often tions such as the lack of property rights and production been dismissed (e.g., Lucas 1988). A general conclusion incentives, labor hoarding and labor immobility exces- is that distortionary policies are not that important in sive inventories and unfinished construction, quantita- 177 tive allocation of inputs by planners and severe gives different estimates of per capita growth in four so- misalignment of relative prices severely distorted the re- cialist countries based on comparisons of these coun- source allocation in these countries.4 One special fea- tries with US per capita income growth in 1948 and 1988 ture of the incentive system in socialist economies is the (different estimates of Soviet growth will be discussed in so-called "ratchet effect," whereby improved perfor- the next section). mance only results in an increase in plan targets for fu- The literature identifies a number of distortionary ture performance. practices characteristic of socialist economies. Produc- Table 14-1 shows the growth performance of selected tion is said to be biased toward heavy industry and capi- socialist economies. While their performance does not tal-intensive agriculture, a reflection of the Marxist compare badly with that in the rest of the world, their obsession with large-scale physical capital. The service rates of investment are much higher. Their reasonable sector and housing have received far less emphasis. Table growth performance suggests that the effects of even se- 14-3 confirms this pattern: it shows that the share of vere distortions can be offset by sufficiently high rates of capital devoted to the production of goods (industry and saving. The informal sector is also thought to play an im- agriculture) is far higher in the Soviet Union that in Or- portant role in many of these countries, just as in dis- ganisation for Economic Co-operation and Development torted developing countries, and can help maintain (OECD) economies, while the share of residential dwell- growth. In the Soviet Union, for example, the shadow ings is only half as large. It is highly unlikely that this economy has been estimated at 20 percent of gross do- structure corresponds to the preferences of Soviet con- mestic product (GDP).5 sumers. In a word, socialism produced too many steel Recent developments in Central and Eastern Europe factories and tractors and too few houses. This overem- have led to much questioning of these officially estimat- phasis on heavy industry and capital-intensive agricul- ed growth rates. For example, the Czech economist ture has been blamed for the recent economic Miroslav Hrncir has pointed out the following anomaly. stagnation in the Soviet Union, specifically, for the slow- Per capita growth in Czechoslovakia since the war is of- ing of technological progress and the lack of competi- ficially estimated to have been faster than that in Austria. tiveness in the world economy (Ofer 1987, and see also Yet he estimates that the standard of living in Czechoslo- Gomulka 1986). vakia is now less than half that in Austria, while imme- Another characteristic bias of socialist economies is diately after the war it was roughly equal. Table 14-2 excessive use of inputs. Thumm6 shows how the energy use per unit of output in Council for Mutual Economic Table 14-1. Growth and Investment in Socialist Assistance (CMEA) countries was more than twice as Economies large as that in OECD countries, while the use of trans- port per unit of output was 4 to 10 times larger. The So- Gross domestic Growth viet Union has a ratio of gross output to net output investment 1960-88 considerably larger than that in the United States.7 This (as % of GDP 1988 (percent) use of inputs reflects their subsidization and the soft China 38 6.0b budget constraint facing producers. Poland 33 4.4 The high cost of transport in socialist economies is Hungary (1971-89) 25 3.1 indicative of the inefficient spatial distribution of pro- Algeria 31 3.9 duction. Hewett (1988) has shown how the vertical inte- Yugoslavia 39 4.8 gration of Soviet ministries leads them to ship inputs Soviet Union 33 2.6-5.6 from one of "their" enterprises to another even if there Average low-income 18 4.3is a closer plant of another ministry producing the same (excluding China/India) iu For antrather the mgithe have Average middle-income 25 5.3 input. For administrative reasons the ministries have (Africa, East Asia, also preferred highly concentrated production, which South Asia, EMENA, LAC)' means that sometimes one plant is supplying the entire Average OECD 22 3.4 vast Soviet territory. This monopolization of production a. Gross domestic product. implies that production can easily be disrupted by local a. Gross domestic product. b. Refers to growth in net material product (NMP), 1960-86. incidents. For example, the cigarette shortage in the So- c. 1965-88. viet Union in 1990 is alleged to have been the result in d. EMENA-Europe, Middle East and North Africa Region, World Bank; and large part of ethnic strife that disrupted the production LAC-Latin America and the Caribbean Region, World Bank. Sources: World Bank data except: Poland, World Bank (1990); Yugoslavia, of cigarette filters by the sole producer, located in Arme- growth (Yugoslavia, various issues); and Soviet Union (Ofer 1990). nia. 178 Table 14-2. Measures of Per Capita Income Growth (percent) A. Per capita income: Ratio to U.S. per capita income in the same year Fischer-Gelb physical Exchange indicators rate UN GNP method method method 1948 1988 1980 1988 Czechoslovakia 28.6 51.0 42.0 18.0 Hungary 14.3 44.0 32.0 13.0 Poland 20.6 37.0 27.0 9.0 Bulgaria 9.7 38.0 30.0 13.0 B. Per capita income: Growth per annum Physical Exchange indicators rate GNP method method method 1948-88 1948-80 1948-88 Czechoslovakia 3.48 3.23 0.82 Hungary 4.89 4.58 1.74 Poland 3.49 2.85 -0.11 Bulgaria 5.54 5.67 2.75 Note: Per capita growth in the United States, calculated from the International Monetary Fund's International Financial Statistics data, is as follows: 1948-1980 1.990 percent 1948-1988 1.989 percent Sources: United Nations, Economic Commission for Europe (1948, Table E, p. 235); Fischer and Gelb (1990); GNP method-United States, Central Intelligence Agency (1989); physical indicators method-Erlich (1987); exchange rate method-International Monetary Fund,International Financial Statistics (various issues) and World Bank staff estimates. Table 14-3. Percentage Distribution of Capital Stocks, 1987 Agriculture Industry I+A Dwellings Other D+O I+A+O (I+A)(D+O) (I+A)(D) (A) (I) (D) (0) Soviet Union 14.2 32.2 46.4 18.6 35.0 53.6 81.4 0.87 4.38 Industrial market 5.0 23.4 28.4 35.9 35.6 71.6 64.1 0.40 1.78 economies United States 2.8 22.4 25.2 45.6 29.2 74.8 54.4 0.34 1.19 Australia 3.5 24.4 27.9 27.7 44.4 72.1 72.3 0.39 2.61 Belgium 1.5 20.5 22.0 35.2 42.8 78.0 64.8 0.28 1.84 Finland 7.5 19.9 27.4 33.8 38.8 72.6 66.2 0.38 1.96 Germany, F.R. (1986) 3.6 20.1 23.7 44.2 32.1 76.3 55.8 0.31 1.26 Greece 10.7 21.6 32.3 34.5 33.2 67.7 65.5 0.48 1.90 Norway (1986) 9.7 28.6 38.3 26.4 35.3 61.7 73.6 0.62 2.79 Sweden (1983) 4.1 25.0 29.1 41.0 29.9 70.9 59.0 0.41 1.44 Great Britain 1.8 28.0 29.8 35.1 35.1 70.2 64.9 0.42 1.85 Source: USSR-Soviet data; industrial economies-OECD data, 179 Of course, the phenomenon of pervasive shortages in ing. Throughout the paper, it is assumed that the econ- socialist economies is caused fundamentally by the ab- omy is insulated from the international financial sence of flexible pricing that matches supply and de- markets, although trade in goods is considered. While mand. This situation and the scarcity of retail services the model is highly stylized, the results seem to offer in- imply queueing costs to consumers. Casual observation sight into some of the characteristics of socialist coun- suggests these costs are large. tries discussed above. Socialist countries are also characterized by resourc- es wasted on large stocks of inventories and unfinished Production construction. These conditions reflect the easy credit to enterprises, the soft budget constraint and the efforts of Equation (14-1) shows the production function for enterprises (and consumers) to hedge against shortages the analysis: and cutbacks in resources. Thumm8 shows that the ac- cumulation of inventory was many times higher in Hun- (14-1) Y = (y1KP + 72KP) P gary, Poland, Czechoslovakia and Yugoslavia than it was in the OECD economies. As table 14-4 reveals, the accu- Output Y is a CES function of the two generic types of mulation of inventory in the USSR was also excessive by capital, K1 and K2, with the elasticity of substitution 1/(p international standards: it averaged 2.7 percent of GDP -1). Since no fixed inputs enter the production function, over 1981-88, compared with 0.4 percent in the OECD sustained growth is feasible through the accumulation countries. Unfinished construction in the USSR of physical and human capital. For simplicity's sake, amounted to 83 percent of investment in 1988 and as population is assumed to be fixed throughout the analy- much as 164 percent of investment in the electrical en- sis. This type of production function can be justified as ergy sector. being an asymptotic approximation to one in which fixed inputs become less important at high incomes (Jones and Manuelli 1990; Easterly 1990b). Others have argued At Model of Endogenous Growth with that spill-overs from the accumulation of physical and Distortionary Policies human capital produce a linear relation between output and capital (Romer 1987b). Finally, Rebelo (1990) has This section presents an endogenous growth model shown that sustained growth requires only a core of cap- that attempts to capture some of the special features of ital goods that can be produced without fixed inputs, as the distortionary government policies in socialist coun- in equation (14-1). tries.7 Only the permanent, steady-state effects of such The distortion is defined as the differential between policies are considered. It postulates a model of constant the marginal products (rates of return) of the two types returns to scale in reproducible capital, as in Rebelo of capital, as follows: (1990). Capital is broadly defined to include both physi- cal and human capital. Two types of capital, variously in- ay terpreted as formal and informal sector capital, imported aK and domestic capital, and "heavy" and "light" capital, are (14-2) - defined as producing the single output. Distortion is de- -= et fined as any government policy that causes the marginal - 2 products of the two types of capital to diverge from the optimal resource allocation. The relationship between where X is the exponential rate of the sales tax (or tariff) this rate of distortion and growth is then discussed, with on investment in type 1 capital.8 attention paid to both fixed saving rates and optimal sav- Profit-maximizing producers will rent the two types Table 14-4. Inventories-Flows as a Percent of Gross Domestic Product (current prices) Average 1981 1982 1983 1984 1985 1986 1987 1988 1981-88 OECD 0.1 -0.4 -0.2 1.2 0.3 0.4 0.6 0.6 0.4 USSR 5.8 4.3 4.3 2.9 2.7 0.8 -0.1 1.0 2.7 Sources: OECD-Thumm (1990); and USSR-Soviet data. 180 of capital such that equation (14-2) is satisfied. The im- formed from domestic output at zero installation cost. plication is that the following will be the ratio of type 2 However, they differ in other characteristics such as lo- to type 1 capital (denoted B), derived from equations (14- cation or institutional form of ownership, and these 1) and (14-2): characteristics cause them to enter into production dif- K , ferently and also determine whether they can be hidden (14-3) B = 2 - successfully from the authorities. k1 L~J ep Although the analysis is presented in the form of a sales or import tax on type 1 capital (the most common The distortion X induces more of type 2 capital to be types of taxation in socialist countries), the taxes can be held than is socially optimal. The elasticity of substitu- shown to be equivalent in steady state to a tax on the in- tion [1/(p -1)] determines how strongly the ratio of cap- come from type 1 capital.9 Thus, the analysis would be ital inputs will respond to increases in the distortion. the same for an analysis of a proportional income tax (on The accumulation of the two types of capital is con- both the labor earnings from human capital and the sidered next, first under fixed saving rates, then under profits from physical capital) that is widely but incom- optimal saving behavior. Two interpretations of this dis- pletely evaded. The tax is also equivalent to other gov- tortion in socialist economies are relevant. The r could emnment actions that imply either extra unit costs for measure the ex-post differential in the rates of return investment in the formal sector, such as investment li- that results from non-market allocation of investment censes or other regulatory requirements, or implicit across types of capital, as is the case under central plan- proportional deductions from income, such as govern- ning. Capital is usually allocated according to other cri- ment expropriation of property or a ratcheting upward of teria than its rate of return across alternative uses. For plan targets. The analysis is the same if the different example, the aforementioned excessive investment in types of capital are subject to differential rates of implicit the production of goods and underinvestment in servic- or explicit subsidization/taxation, such as would occur es implies that the latter offers a higher rate of return, with directed credit schemes or quantitative allocation Alternatively, X could measure the implicit tax on formal of foreign exchange. income in socialist economies, where individuals retain little of the additional income from improved perfor- CapitalAccumulation with Fixed Saving Rates mance because of the lack of property rights and afore- mentioned "ratchet effect." Capital in the shadow With a lump-sum rebate of tax revenues, the income economy-where a person can keep much of the reward of the private sector is equal to total output Y As in the from the productive effort-would be represented by K2. original Solow model, it is assumed that a fixed propor- Because of the risk of penalties under the law, the rate of tion, s, of this income is saved. Since it is assumed that return on illegal activity will be much higher than that the economy is closed to inflows of foreign financial cap- in the formal sector. Thus, this simple set-up captures ital, saving will equal investment (I,) in type 1 and type many of the stylized realities of government distortions 2 capital goods: in socialist economies. Other interpretations of the capital types and distor- (14-4) 1 + I2 = sY. tions are also illuminating (although analytically equiv- alent). According to one, it can be assumed that the It is assumed that the rate of depreciation of capital economy produces a single domestic good with capital (a) is the same for both types. made up of imported goods (K,) and the domestic good The growth rate for the economy in a steady state is itself (K2). Both goods are assumed to be traded, and the then given by the following: economy is assumed to be small in international trade. The price ratio of the two goods is thus fixed and can be normalized at unity. The distortion rate X can then be a _ misalignment of the exchange rate, a tariff on the im- ported capital good or an import quota determined by (14-5) g =s (7y + y2BI) P - a. planning (defined as the equivalent tariff rate). L 1 + B Yet another interpretation is that two types of capital are formed out of the single domestic output in an au- It can be shown that growth is unambiguously nega- tarkic economy. The first type, called formal sector cap- tively related to the distortion rate . This finding is in- ital, is subject to a sales tax at the moment of purchase tuitively clear since it is known that output for a given for investment. The second type, informal sector capital, level of capital is maximized at zero distortion. Equation evades the sales tax. The two types of capital can be (14-5) simply represents saving out of output as a ratio 181 to existing capital. This equation is the growth analogue distortion, sufficiently high increases in saving could off- to the standard result in the Solow model that a tax on set the negative growth effects of distortionary policies. capital will reduce the steady-state capital/labor ratio For example, in this simulation the loss in growth from and lower output (see, for example, Atkinson and Stiglitz an arbitrarily large distortion is 5 percentage points, 1980). which could be offset by an increase of 20 percentage It is more illuminating to present a graph of the rela- points in saving. While such high saving imposes a se- tionship between the rate of distortion and growth for vere cost on the population, it is within the range of fea- plausible parameter values (figure 14-1) 10 The relation- sibility. ship approximates a logistic curve, with flat segments at The non-linear shape of the growth-distortion rela- the low and high rates of distortion. In this simulation, tionship also has implications for policy reforms. Elimi- growth approaches a minimum as the distortion X goes nation of small distortions is practically pointless to infinity. Increases in the distortion from zero are not according to this model, since it means just a backwards very costly at first but show increasing costs in terms of movement along the initial flat part of the curve. Simi- lost growth as the distortion grows. This pattern is the larly, modest reductions of very large distortions have dynamic analogue to the well-known static principle of virtually no impact, since the tax rate of an activity that increasing marginal costs of distortions, which in this is at a very low level has little weight. This result implies, case reflects diminishing returns to increased use of type for example, that market reform in socialist countries 2 capital. The flattening of the curve as X increases re- must reach some critical mass to have a growth effect. flects the gradual disappearance of type 1 capital. As the This model supports the idea of the "Big Bang" as in Po- use of type 1 capital approaches zero, the damage caused land in 1990, although this conclusion is only from the by additional increases in the distortion becomes slight. point of view of a permanent gain rather than any tran- The finding of a growth rate always above a certain sitional advantage. minimum even as distortion approaches infinity is intu- The elasticity of substitution plays a critical role in itively plausible in light of the socialist experience dis- this analysis. The results just presented will only hold cussed earlier. The possibility of substituting informal when the elasticity of substitution is strictly greater than capital for formal capital could explain why output and one, the implication being that neither input is essential growth do not fall to catastrophic levels even with al- for production. (The simulation just presented assumed most unlimited distortion in the allocation of resources. an elasticity of 2.) Figure 14-2 shows the consequences This possibility could also explain why high rates of sav- of distortions on growth under alternative elasticities of ing could enable socialist economies to grow respect- substitution. With an elasticity equal to unity, output ably. With finite losses from an arbitrarily large goes to zero as type 1 capital is more heavily penalized, Figure 14-1. Distortion and Growth - Alternative since this input is essential for production. Growth de- Assumptions clines asymptotically to depreciation of the capital stock Growth rate (-a) without replacement. With an elasticity equal to 3, 0.05 on the other hand, growth stays above a minimum of over 2 percent no matter how large the distortion. 0.04 - Figure 14-2 shows that the flat part of the curve at low rates of distortion with inelastic production is much 0.03 - larger than the flat portion with highly elastic produc- tion. The point made earlier about small distortions not 0.02 - being costly is stronger the more inelastic production is. Similarly, the flat part of the curve at high rates of dis- tortion is longer (and the loss in growth smaller) the om1 " .more elastic production is. Modest reform of large dis- - , , , , tortions is more ineffective the greater the elasticity of 0 I substitution is. To summarize, a reform strategy based 0 0.5 1 1.5 2 2.5 3 3.5 4 on this model should be affected by the assumption Exponential rate of distortion about the substitutability of formal and informal capital. If it is believed that substitutability is high, the emphasis .- Exogenous saving rate - Optimal savnng behanior should be on very large reductions in the large distor- tions, since moderate reductions will have little effect. For example, if it is believed that there is substantial re- placement of the formal economy by the underground 182 Figure 14-2. Growth and Distortion -Alternative while the equations of the accumulation of capital are Elasticities of Substitution simply: (exogenous saving rate) Growth (14-8) K1 = I,- aK, 0.04 ~~~~~~~~~~~and 0.02 (14-9) K1 = 1,2 uxK,. 0.01 The condition that investment is irreversible was not 0 explicitly imposed. -0.01 Showing that the planner solving this maximization - .2- problem will allocate investment between the two types such that equation (14-2) continues to hold is a straight- - 0.03 forward matter. That is, t will be equal to the wedge be- -0.04 ~~~~~~~~~~~~~tween the marginal products of the two types of capital. 0 0.5 1 1.5 2 2.5 3 3. 4 4. In steady state, both types of investment and capital, Rate of distortion along with consumption and output, will all grow at the -..... Elasticity of substitution= I - x=3 same rate, which will be given by: Source: Computer simulation by the author. (14-10) g = 2~ a- where r2 is the marginal product of type 2 capital, given economy under socialism, then a drastic market reform bytefloig is needed to make efficient use of formal capital and thus (14-11) y, (y, - + 2) PI to raise growth. If it is believed that substitutability is low, it is pointless to bother with small distortions, but and B is the ratio of K2 to K1 given by equation (14-3). even moderate reductions in large distortions will be ef- Growth is given by the familiar condition that the opti- fective. The growth benefit of reform is greater the more mal growth of consumption is equal to the difference be- inelastic the production structure is. tween the net marginal product of capital and the rate of time preference times the intertemporal elasticity of Capital Accumulation with Optimal Saving Behavior substitution. Since the marginal product of type 2 capi- tal goes down when type 1 capital is taxed more heavily, This section considers the implications of optimal growth is unambiguously a negative function of the dis- saving behavior to see how saving responds to distortion. tortion rate rc. A plot of the growth-distortion relation- Assume that a beneficent planner seeks to maximize the ship for a simulation with the same parameter values as present discounted value of the utility of future con- before is shown in figure 14-1, which also shows the re- sumption: lationship under exogenous saving behavior.i2 c1-n Marginal costs no longer rise with increased distor- (14-6) f;er 1 dt. tion under optimal saving behavior. Even small distor- o ( tions are costly, because they cause both a decrease in Utility is given as an isoelastic function of total con- saving and a static efficiency loss. However, with endog- sumption C where the intertemporal elasticity of substi- enous saving, the growth consequences of very large dis- tution is 1/a. tortions may be less severe than under fixed saving rates. Consumption is given by the excess of income over The reason is that the rate of saving actually increases at investment spending. The investment in type 1 capital is very high rates of distortion. taxed (implicitly or explicitly) at an exponential rate, The reasons for this result are examined below. It is a expQ,r), as before (in other words, the planner is power- straightforward proposition to show that the optimal less to alter the distortions induced by the system but propensity to consume out of total capital (K1 + K2) is can at least alter the rate of saving)": given by: c 1 r r2 (ec"l- (14-7) C = Y-eVI,-I2+T, (14-12) k~_2= (r, -aI 1a) ~+ I 183 The first two terms in this expression are familiar: the ses of the effect of distortions have been carried out in optimal propensity to consume out of wealth (capital) the context of the traditional Solow model, in which dis- will increase with the net rate of return if the intertem- tortions in the allocation of resources only have output poral elasticity of substitution is less than one, while if it effects. Nevertheless, their conclusions can be extended is equal to one, this propensity will be the rate of time in straightforward manner to the endogenous growth preference. The third term gives the amount of the models. The definitive study on the output effect of So- lump-sum transfer of the tax revenues to consumers as viet central planning is found in Desai (1987). She uses a ratio to the capital stock. These revenues are entirely a clever and imaginative method to measure the devia- consumed. As the tax rate increases, optimal consump- tion of Soviet resource allocation from that which would tion declines because the rate of return falls, and the in- pertain under a market economy. She estimates produc- come effect is assumed to outweigh the substitution tion functions for a number of sectors for which data on effect. The consumption decline is partially offset by the sectoral capital and labor are available. She then calcu- consumption of lump-sum tax revenues, but these do lates the loss of output caused by the failure to equate not affect the growth rate, and the lump-sum transfer the marginal rates of substitution of capital and labor disappears as the tax rate increases. The increase in sav- across sectors, as would occur under perfect competi- ing as a share of income in response to the high levels of tion. She estimates a loss in output of 4-10 percent distortion explains why distortion causes less of a fall in across sectors and over time. growth in the optimal saving case (given the assumption One problem with this approach is that it is probably of the simulation of an intertemporal elasticity of less not likely that marginal rates of substitution are equated than one, as is supported by most empirical evidence). even in capitalist economies because of departures from This result gives an unusual perspective on the high perfect competition, government capital and ex-post rate of saving and respectable growth in the socialist miscalculations by producers. Thus, the comparison economies. The high saving rate imposed by planners amounts to one between output in the Soviet Union and could be optimal given the extensive distortions, al- an unreal situation that may not exist in any other econ- though of course the distortions themselves are neither omy. Another approach would be to compare the struc- necessary nor optimal. If the intertemporal elasticity is ture of Soviet resource allocation directly with that in less than one, then optimal consumption will fall with similar market economies, assuming that only the dis- increased distortion-the substitution of present for fu- tortions attributable to central planning induce the dif- ture consumption is outweighed by the income effect of ferences. This assumption, while extreme, still affords a the higher level of distortion. The rise in saving when useful first approximation. the level of distortion is greater implies a less severe de- A second problem with this analysis is that not all of cline in growth. Thus, the intertemporal elasticity of the misallocation of resources at all levels can be mea- substitution is as critical in this analysis as the produc- sured. To be precise, it is also necessary that the misallo- tion elasticity of substitution between the two capital cation of resources between types of capital and types of types is. labor be measured, with each successive level of disag- gregation discovering new sources of distortion and out- Estimating the Growth Effects of Distortions put loss. It is like the famous Russian matryozhka doll, which successively opens up to yield smaller and smaller A simple model such as the one presented in this pa- facsimiles of the same doll. per can be used to estimate the growth effects of severe This section measures the loss of output and growth distortions of resources on actual economies. Although attributable solely to the misallocation of capital across these calculations belong to the "back-of-the-envelope" alternative types. This distortion may well be relatively genre, they can provide a useful first approximation in minor when compared with the other distortions caused advance of more detailed, time-consuming studies. Such by central planning. Nevertheless, it is analyzed here be- calculations are useful because they give an idea of what cause the data are readily available. the likely growth benefit of reforms to correct these dis- Addressing the growth effect of distortions in the So- tortions will be. To illustrate how such calculations can viet Union is complicated by the fact that there are many be made, one involving the misallocation of capital alternative estimates of Soviet growth. Capital accumu- stocks in the Soviet Union, as described earlier, is pre- lation is also difficult to measure, and again there are sented here. several alternative estimates. A crude way to measure Some general remarks about the effects that the dis- the differential growth loss of the Soviet system is to tortions induced by central planning have had on output compare the relationship between growth and invest- and growth in the Soviet Union are in order. Most analy- ment in the Soviet Union with the average for all other 184 economies. It is then possible to calculate how much would take place under a market system where private lower or higher Soviet growth is when compared with producers are maximizing profits. the growth that should be expected from Soviet invest- However, a suboptimal allocation of capital that devi- ment rates, if the average pattern for all other economies ates by t percent from the optimal allocation is observed: holds. Table 14-5 shows the difference between Soviet K, growth and the "expected" growth rate using the alter- (14-16) 2 = (+) _ native estimates. The resulting estimates of the differ- Ks a ences in growth range from plus 0.4 percent to minus It can be calculated that the output (Y ) given by this 2.6 percent. The issue will remain unresolved until a bet- suboptimal allocation of capital will be proportional to ter estimate of Soviet growth is available. The opinion of the optimal level of output as follows: most experts, however, seems to be swinging toward the Y (I + t) -a lower estimates. (14-17) - = 1 A version of the model that illustrates how the growth Y 1 + t (1- a) effect of the misallocation of capital in the Soviet Union Output will always be lower for any deviation t of cap- could be calculated indirectly is presented here. It is as- ital stocks from the optimal allocation. sumed that the common production function for all The growth effect of this deviation from maximum countries is given by: output can be analyzed as follows. In steady state, the growth rate will be given by the common rate of growth (14-13) Y = A (KaK-f `) PH' - 3 of physical and human capital. Thus, the growth rate could be written in terms of the growth of physical cap- where H is human capital (i.e., the capital embodied in ital, where a fixed rate of investment in physical capital, the labor force) and K, and K2 are two types of physical Sk, out of income (presumably determined by the plan- capital among which there is misallocation. The optimal ners) is assumed: allocation of capital, assuming that both are formed y from a single domestic output and thus have the same (14-18) g =S - k . user cost, would be given by the following: 'The decrease in the growth rate associated with the (14-14) k2 = _-C misallocation of capital will be given by: The level of output that results from the optimal allo- (14-19) g -g = Sk L K SK L -- cation of capital would be given by: where g is the rate of growth with the misallocation of (14-15) Y = A [aa (1 - a) l-a] 1K1H1 - capital. 'he effect of the misallocation of capital in the form of overinvestment in the production of goods (in- where K is now all physical capital, or the sum of K, and dustry and agriculture) on growth is estimated to be 0.16 K2. It is assumed that an optimal allocation of capital percent per year (the effect on output is about 2 percent, Table 14-5. Comparison of Actual Soviet Growth with That Expected from Cross-Section Regressions of Gross Domestic Product (GDP) Growth Investment Share of GDP (percent) Differential between Soviet growth and expected growth based Estimated Soviet GDP on investment rate growth rate (1960-85) (Ofer) of 29.3 percent according to source Official 0.4 5.6 PlanEcon Report -0.9 4.3 Central Intelligence Agency -1.8 3.5 Dikhanova -1.8 3.4 Selyunin and Khanin -2.6 2.6 a. 1960-89. Source: Ofer (1987) and (1990), Aslund (1989), Dikhanov (1990) and Vanous (1986). 185 in line with typical estimates of the deadweight losses at- marginal products: tributable to distortions) 13 Lest this level seem small, Y note that over the six decades of central planning the (14-21) - -= e loss from this distortion alone would amount to 10 per- aK2 AK, cent of output. and A calculation focusing on the misallocation of capital ay ay 3 in the form of underinvestment in dwellings (holding (14-22) - -= e other distortions constant) yields a growth effect of 0.29 aK3 aK e percent per year, equivalent to a 19 percent loss in out- The implication is that the ratios of capital stocks with put over six decades.14 These tentative calculations sug- these distortions will be given by: gest that tracking down the loss in growth attributable K2 to distortionary policies could in the end yield quite a (14-23) - = i C2 1e 2, large number. 1 23 and The Problems of Partial Liberalization (14-24) K3 =3_ e_ 3. Two applications of the model can give insight into T t lt a b to the dangers of a partial correction of the distortions in Tin can be c uted asifollows: the highly distorted socialist economies.1s First, as shown earlier, the relationship between distortions and e-2x22 -a3 3 growth is highly non-linear-at high levels of distortion (14-25) t = 1- 2 T3 small reductions in the distortions have almost no effect 1 L2 a3 + ac2e + a3e on growth. A certain minimum level of reform is needed and output will be given as a function of total capital K to pass the threshold where there is a significant pay-off (equal to the sum of the three types) as follows: to growth. The reason is that it is necessary to breathe life into previously moribund sectors or types of capital (14-26) Y = 4) (1 - ') K before much effect on growth takes place. A second application of the model is to consider more where 4D is the ratio of output to capital in the absence than one distortion. Here partial liberalization means of any distortion, defined as: correcting some types of distortion but not others. It is 1a -a well-known in the economic literature (under the head- (14-27) 4D = A (I - a2 - a3) 2 3a2aa3 ing of the analysis of the "second-best" ) that removing some distortions while allowing others to remain can The growth rate can be determined by assuming a create a worse situation than leaving all the distortions fixed saving rate out of income s and a fixed depreciation unchanged.16 It is not surprising that the model present- rate k, so that output growth is given by the rate of in- ed in this paper can be extended to show that what is true crease in the capital stocks as follows: of one-time effects on income is also true for growth rates. (14-28) g = sD (1 -t) - o. Here the example of the privatization of a formerly planned socialist economy is used. The model is general- This simple model can be used to simulate the effects ized to have three types of capital, which under central of partial liberalization, defined as a reduction of one of planning were determined by fiat without regard to their the distortion rates with the other left unchanged. A real relative marginal products. Output is given by: world example will help to motivate such a partial liber- alization. Suppose that the government announces a 1 -a2-3 '2 K3 partial privatization of the economy, such that private (14-20) Y = A K, 2 K~2 K3 firms and households can freely buy and sell capital It is assumed as before that all types of capital can be types 1 and 2 but not type 3, which the state continues formed from a single domestic output. Under a market to own and determine. The partial privatization is equiv- system, the marginal products of the three types of cap- alent to eliminating the distortion T2 but continuing the ital would be equated, since they have the same user distortion t3. That is, households will now be able to al- cost. The distortions induced by central planning can be locate their resources among the two forms of private defined as the deviations between the marginal products capital such that their marginal products are equated in of the capital. There are now two independent deviations order to use them efficiently. The third type of capital to measure, as before, as the exponential wedge between will, however, still be determined by the state, at such a 186 level that T3 does not change by assumption. small, or the distortion t3 is small, the removal of distor- What will be the effect on growth of such a partial lib- tionr, will still raise growth. As figure 14-4 shows, if the eralization? It is possible the liberalization will lower the non-privatized capital is relatively unimportant, with a rate of growth and worsen the distortions."7 This possi- share of only 5 percent, then the partial liberalization bility is greater the higher the share of non-privatized will raise growth. capital is in production and the higher the uncorrected Another interpretation of equation (14-29) is that it distortion is. To illustrate this point, the derivative of the shows the wisdom of reducing the largest distortion total distortion t with respect to the one distortion rate where only one can be changed. If an effort is made to re- T2 can be calculated as follows: duce the smaller distortion, then the other larger distor- (14-29) tion makes a perverse effect more likely. Put another way, the lowest-return activities should not be liberal- ar_ a2 (1 -T) [(l -a2) ( e -et2) + a3( 3- 1) ized while leaving the highest-return activity un- 2 - - +e2 + X3 touched. This conclusion also has an implication for 2 l - a2 - a'3 + a2e + a3e sequencing. If only one distortion can be corrected at a This derivative could be negative (the implication being time, then the larger one should be corrected first. Oth- that reducing T2 will raise the total distortion r) if the erwise there will be a negative supply effect before the rate of the other distortion T3 is high and if the share of second phase of reform. non-privatized capital a3 is high. An illustration of this This model gives quite clear policy counsel-partial possibility is given by a simulation of the model for plau- liberalization that leaves a large portion of the economy sible parameters, shown in figure 14-3.18 In figure 14-3, subject to administrative fiat ma well worsen rowth the share of non-privatized capital is 45 percent. As can rather to improve iat may this posenlgrowth be seen, the relationship between the distortion rate T2 rather than improve it.nA However, this possibility is no and growth, when the other distortion '3 is held con- excuse for not reforming. As noted, perfection is unnec- stant, can be either positive or negative. Beginning at essary: as long as the share of the unliberalized sector or the maximum point A, a reduction of this distortion the uncorrected distortion is relatively "small," less than through partial privatization would worsen the distor- complete liberalization will still lead to increased tions and lower growth by almost 1 percentage point. growth. If a choice must be made between two distor- The reform does not have to be perfect, however. As tions, the largest one should always be reduced first to long as either the share of non-privatized capital is avoid perverse effects from the liberalization. Figure 14-3. Relation of One Distortion to Growth with Figure 14-4. Relation of One Distortion to Growth with Other Distortion Unchanged Other Distortion Unchanged (unprivatized sector 45%) (unprivatized sector 5%) Growth rate Growth rate 0.024 0.022 0.03 - 0.02 o.O,. 0.016 0.014 0.012~~~~~~~~~~~~~~~~00 0 0.5 1 1.5 2 0 0.5 1 1.5 2 Rate of one distortion with other distortion fixed at 1 Rate of one distortion with other distortion fixed at 1 Source: Computer simulation by the author Source: Computer simulation by the author 187 Conclusions short supply at the time, while socks were on the list of critically scarce consumer goods. A stylized model was presented in this paper to show bachev and Yeltsin as quoted in the Washitmmson appointed by Gor- how government-induced distortions can affect the rate 1990. of growth in an endogenous growth model, both with 6. SeeChapter4,UlrichThumm, "WorldBankAdjustmentLend- and without saving effects. The results suggest that pre- ing in Central and Eastern Europe," in this volume. dicting the effect of distortions on growth depends 7. The excessive use of inputs is all the more striking in light of heavily on structural parameters and initial conditions, the downward bias attributable to their lower relative prices in socialist heaviy onstrutura parmeter andinital cnditons, economies. especially the elasticity of substitution of the different 8. op. cit., Thumm, this volume. types of capital, the shares of each type of capital being 9. This section draws on the model presented in Easterly (1990a). distorted, and the initial distortion. While this model is 10. If the income from type 1 capital is what is taxed, then the in- too simple to give detailed policy advice, it suggests come tax rate tequivalent to the exponential sales tax 'r is t =1-exp(-c). some basic principles. The policy-maker should attempt 11. This equivalence is noted by Atkinson and Stiglitz (1980, p. to identify and move along the steeply-sloped portion of 12. The parameter values are s = .2, a = .05, y7 =y2 = .5 and p= .5. the growth-distortion relationship. In the exogenous Sensitivity testing confirms that the shape of the function is the same saving model, this advice means ignoring the small dis- for alternative values except for p, as noted in the text. tortions and making large reductions in the very high 13. The variable T is the lump-sum transfer, ex-post equal to e II levels of initial distortions. If more than one distortion but treated by the consumer as fixed. 14. The additional parameters for the simulation are sigma = 7 (in- exists and only one can be addressed, the largest distor- tertemporal elasticity = .14) and r = .1. The sensitivity of the results to tion should be the one corrected. these parameters is discussed below. While the model is highly stylized, it provides insight 15. The calculation is made with data on capital stocks from 1987. into the need for radical reform in socialist economies- It assumes a capital share of 0.4, a capital/output ratio of 4.33 (the for a "Big Bang." Radical reforms will encounter strong OECD average) and a physical investment share of output of 0.33 (the Soviet investment share in 1987 came from Ofer 1990). political resistance, but so will partial reforms. The ques- 16. The cost of this distortion cannot be added to the previous one, tion is why carry out a politically costly partial reform as there is some overlap between the two since only pair-wise distor- that brings little benefit when a radical reform with not tions are being considered. much higher political costs can revolutionize the econ- 17. Here partial liberalization means a permanent partial change. omy? In addition, this model does not address the desirable phasing of re- forms during the transition. 18. The canonical article is Lipsey and Lancaster (1956). Notes 19. Of course, the privatization will also have favorable effects on the incentives to save and invest, which could well outweigh the possi- 1. This paper is a revised version of the one prepared for the Con- ble negative effects of the partial reform. ference on Adjustment and Growth: Lessons for Eastern Europe, held 20. The parameters are A = 1, a2 = .45, a2 =.45, cx3 = .05 or .45, -2 in Pultusk, Poland, October 4-5, 1990. 1 benefitted substantially from 1 and s = 0.2. comments by my discussant, J. Rostowski, and the conference partici- 21. There is also a political argument against partial reform. It pants and from discussions with Robert Barro, Stanley Fischer, Robert could allow some well-connected private sector operators to capture King, Ross Levine, Sergio Rebelo and Thomas Wolf on related work. the rents from the remaining distortions. This enrichment of a few pri- The research assistance of Piyabha Kongsamut and Maria Christina vate individuals could provoke a political backlash against privatiza- Almero is gratefully acknowledged. The author alone is responsible for tion in general. This situation may explain the widespread hostility any remaining errors. All views expressed are those of the author and toward the private cooperatives in the Soviet Union. not necessarily of the World Bank. 2. More precisely, the standard neoclassical model assumes that References the marginal product of each factor goes to zero as that factor increas- es, with other factors held constant (the Inada condition). Jones and Aslund, Anders. 1989. Gorbachev' Stniggle for Economic Reform. Ith- Manuelli (1990) recently explored the implications of relaxing this as- aca N.Y.: Comell University Press sumption 3. The growth effects of distortionary policies could be preserved Atkinson, A. and J. Stiglitz. 1980. Lectures on Public Economics. New in the Solow model if the focus were on the transition to a steady state. Atk: aw-Hill. Corden (1971), for example, showed strong growth effects of trade pol- York: McGraw-Hill. icy in the Solow transition. However, King and Rebelo (1989) have convincingly demonstrated that the Solow transition can explain very Barro, Robert J. 1989a. "A Cross-Country Study of Growth, Saving and little of the long-run growth rate without generating counterfactual Government." NBER Working Paper No. 2855. National Bureau of implications for the interest rate. Economic Research. Cambridge, Mass. 4. An entertaining example of the misallocation of capital under central planning is given by Hewett (1988, p. 175). Because of a short- . 1989b. "Economic Growth in a Cross Section of Countries." age of certain critical consumer goods, the Tochmash factory in the A paper prepared for the Conference on Human Capital and USSR in 1987 was ordered to produce flashlights, although it was de- Growth, State University of New York (SUNY), Buffalo, NY (May signed to produce machinery for making socks. This machinery was in 1989). 188 Barro, Robert J., and Xavier Sala-i-Martin. 1989. "Economic Growth Lipsey, R.G., and K. Lancaster. 1956. "The General Theory of Second and Convergence Across the United States." National Bureau of Best." Review ofEconomic Studies 24(1)(October):11-32. Economic Research. Cambridge, Mass. Lucas, R.E. 1988. "On the Mechanics of Economic Development." 1990. "Public Finance in Models of Economic Growth." Journal ofMonetary Economics 22(1)(July):3-42. NBER Working Paper No. 3362. National Bureau of Economic Re- Ofer, Gur search. Cambridge, Mass. G 1987. "Soviet Economic Growth: 1928-1985." Joumal of Economic Literature XXV (December):1767-1833. Corden, M. 1971. "The Effects of Trade on the Rate of Growth." In J. Bhagwati, et al., eds., Thade, Balance ofPayments, and Growth: Pa- . 1990. "Macroeconomic Issues of Soviet Reforms." NBER pers in Honor of Charles P Kindleberger. Amsterdam: North-Hol- Macroeconomics Annual 1990. Cambridge, Mass. land Publishing Company. Rebelo, Sergio. 1990. "Long Run Policy Analysis and Long Run Desai, Padma. 1987. The Soviet Economy: Problems and Prospects. Growth." NBER Working Paper 3325. National Bureau of Econom- Oxford: Basil Blackwell. ic Research. Cambridge, Mass. Dikhanov, Yuri. 1990. "Real Growth Rates: 1913-1989." Business in the Rodri, 1990. 'How Should Structural Adjustment Programs Be De- USSR (4)(September). signed?: WorldDevelopment 18(7):933-47. Romer, P.M. 1986. "Increasing Returns and Long-Run Growth." Jour- Easterly, William. 1990a. "Endogenous Growth in Developing Coun- nal of Political Economy 94(5):1002-37. tries with Government-Induced Distortions." A paper presentedthe World Bank Conference on Adjustment Lending, September 12-13, 1987a. "Growth Based on Increasing Returns Due to Spe- 1990. cialization." American Economic Review 77(2)(May):56-62. -i990b. "How Does Growth Begin? Models of Endogenous De- _ _. 1987b. "Crazy Explanations for the Productivity Slowdown." velopment." Mimeo. World Bank. Washington, D.C. In S. Fischer, ed., NBER Macroeconomics Annual. Cambridge, Mass.: MIT Press. Erlich, E. 1987. "Absolute and Relative Economic Development Levels and Their Structure in 1980." Mimeo. Budapest. . 1988. "Endogenous Technological Change." University of Chi- rago, Chicago. May. Fischer, S., and A. Gelb. 1990. "Issues in Socialist Economy Reform." Mimeo. WVorld Bank. Washington, D.C. September. . 1989. "What Determines the Rate of Growth and Technologi- cal Change?" Policy, Planning and Research (PPR) Working Paper Gomulka, Stanislaw. 1986. Growth, Innovation and Reform in Eastern Series, WPS 279 World Bank. Washington, D.C. Europe. Madison, Wisc.: University of Wisconsin Press. Solow, Robert. 1956. "A Contribution to the Theory of Economic Hewett, Ed. A. 1988. Reforming the Soviet Economy. Washington, Growth." Quarterly Journal of Economics 70(February):65-94. D.C.: Brookings Institution, United Nations. Economic Commission for Europe. 1948. Economic International Monetary Fund. Various issues. International Financial Survey ofEurope in 1948. New York: United Nations. Statistics. Washington, D.C.: IMF. United States. Central Intelligence Agency. 1989. Handbook of Eco- nomic Statistics. Central Intelligence Agency. Washington, D.C.:- Jones, Larry, and Rodolfo Manuelli. 1990. "A Convex Model of Equilib- CIA rium Growth." Journal of Political Economy 98(5):1008-38. Vanous, Jan. 1986. "Developments in Soviet and East European Na- King, Robert, and Sergio Rebelo. 1989. "Transitional Dynamics and tional Income, 1950-85." Plan Econ Report II(50-52)(December). Economic Growth in the Neoclassical Mode]." NBER Working Pa- per No. 3185. Cambridge, Mass Yugoslavia. Various issues. Statistical Yearbook of Yugoslavia. - 1990. "Public Policy and Economic Growth: Developing Neo- World Bank. 1990. World Development Report 1990. New York: Oxford classical Implications." Journal of Political Economy (October). University Press. 189 15 Comments on "Distortionary Policies and Growth in Socialist Economies" by William Easterly, Jacek Rostowski The main comments made here address the applica- services for which supply and demand are interrelated. bility of William Easterly's model to centrally adminis- This information could never be collected, let alone tered economies of the Soviet kind in which nearly all transmitted, to all those concerned directly. Similarily, allocative decisions related to production and invest- those making investment decisions (the central plan- ment are made on the basis of administrative orders. The ners) are not motivated to allocate resources optimally. comments do not apply to economies such as those of Instead, they are likely to be affected by various kinds of Hungary, Poland or Yugoslavia at the present time, in ambitions for bureaucratic empire-building. However, which a very large number of these decisions are made even if they were properly motivated, they could not on a market basis and for which Easterly's model is suit- make high quality decisions because they do not have able. Moreover, endogenous growth models, which allow good information on which to base them. Only a price distortions in the allocation of resources to affect not system can provide that information. only the level of output but also the rate of growth of an Imagine an economy that becomes centrally adminis- economy, are a definite improvement over the tradition- tered after a peaceful Communist revolution. At time tO al Solow-type exogenous growth models, which do not all its resources are optimally allocated (or as optimally allow for such an outcome. allocated as the previous capitalist market economy The weakness of Easterly's model is that it does not al- could have allocated them). From time tl on, however, low the growth rate of centrally administered economies two things begin to happen. First, there are changes in to decline over time, as suggested by both theoretical the "state of nature" (final consumer demand, the avail- considerations and historical experience. As Ludwig von ability of raw materials, technology, international prices, Mises and Friedrich von Hayek pointed out over 50 years etc.), so that the original allocation of resources be- ago, the problem with centrally administered economies comes less and less optimal. This process is called "opti- is not just that resources are allocated inefficiently, but mality drift." Even if planners' responses to this drift that the process of allocating them is inefficient. The result in a better allocation of resources than if there had reason is that in such economies the informational and been no response whatsoever (an assumption that is motivational roles of the price system are not allowed to made for the moment), nevertheless the overall alloca- operate. Because of the all-pervasive principal agent tion of resources in the economy will be ever more sub- problems, the people directly in charge of productive re- optimal on average since, as already said, a centrally sources at the level of the factory or shop have little mo- administered economy allocates resources in an inferior tivation to make the best possible use of them. Since the way to that of a market economy, and a larger and larger price system cannot fulfill its informational role, the de- proportion of all resources will have been allocated un- gree of information managers have about the current der the centrally administered economy. During this and future behavior of suppliers and customers is less process the degree of distortion in the economy will have than it would be in a market economy. Hayek in partic- grown, i.e., the situation is as if, in Easterly's terms, the ular stresses this aspect of the problem, pointing out number or magnitude of the tax wedges increases. As the that a single price carries in synthetic form an enormous degree of distortion grows, the long-run rate of growth amount of information regarding production possibili- of the economy should decline. ties and the preferences and expectations of a vast num- The only question is how long the long run is. Clearly, ber of suppliers and consumers of many goods and over time so few of the resources in the economy will re- 191 main that had been allocated under the market economy omniscient entity could always say at any point just what that the effect of the optimality drift will have less and the degree of distortion in the economy is, and, indeed, less impact on the average optimality of resource alloca- identify the ex-post distortions, the planners cannot do tion in the economy and, therefore, less and less impact so. They do not know to what extent they have misallo- on the economy's growth rate, which will settle at some cated any given resources in the past. Therefore they low but positive level. This process may or may not be have to assume that all resources are well-allocated. sufficiently extended over time to account for the ob- However, since that assumption is mistaken, it increases served decline in the growth rates of centrally adminis- the likelihood of present errors. The greater the amount tered economies over their history. In addition, the high of resources is in the economy that have been allocated rates of investment that centrally administered econo- under the centrally administered economy, the greater mies tend to have may even result in an acceleration of the degree of error involved in this Panglossian assump- growth in the early years of the system, a situation that tion. As a result, the quality of resource allocation by further confuses the empirical picture. planners must decline continuously. To illustrate what It is possible, however, to take a more extreme posi- is meant, consider the following sketch of a model, as- tion than the above one on the efficiency of resource al- suming an economy subject to exogenous stochastic location in a centrally administered economy. That more shocks. Let extreme position, which would be consistent with the X'ij = actual allocation of resource i to activityj at reasoning of von Mises and Hayek, would claim that a time t. centrally administered economic system generates no X'ij = optimal allocation of resource i to activity j useful inforrnation whatsoever regarding the allocation at time t. of resources. That claim does not mean that no useful in- formation exists in the system. Quite a lot of useful in- Xij = Ixij - X'ij /Xtij formation is left over from the previous market system, and this information (although it becomes more and X'ij = at + bXX'-'ij + u' where a>O, b>O and u is more out of date with the passage of time) can be useful v to central planners (the key resource allocators, who are randomly distributed. assumed to some extent to be interested in good re- Thus the passage of time (t) causes a deterioration in source allocation within the economy). resource allocation because of the declining relevance of The centrally administered economy, then, has the the information inherited from the previous market- following history. In the early years central planners based system. The direct impact of this effect is repre- have access to a large amount of information inherited sented by the coefficient b, whereby the aggregate value from the previous market system, information that is of past errors in the allocation of resources[,X -'Iii relatively up-to-date. As a result, although the allocation of resources is inferior to that in a market economy, the increases the error in the allocation of any particular re- difference is only one of degree, and if nothing else were hapnig th drf scnai decie above would take... source in the present, because past errors are noise that happening, the drift scenario described above would take reduces the amount of correct information available to place. However, the critical point (one that is not prop- . . . erly taken into account in the drift scenario) is that un- der this system the market is not constantly "verifying" rors accumulate, ,X'ij and grow with time, whereas the allocation of resources, i.e., the system is not contin- ii uously generating new information that allows the opti- in a market economy new errors in the allocation of par- mality of previous allocations of resources to be assessed ticular resources by particular actors are offset by im- as a market economy does. As a result, two processes in- provements in the allocation of the same resources by crease the amount of "noise" or "informational entropy" other actors or in the allocation of other resources. in the system and make it ever harder for planners to Thus, in a market economy make good decisions regarding the allocation of resourc- es. The first is that the information inherited from the X'ij - cX' 'ij + u' where cO. long run, the output of a centrally administered economy UY is zero (also, resources are totally randomly allocated). What is more, the increase in "informational noise" in the This implies that not only would growth rates decline dur- centrally administered economy has no natural upper ing the life of the centrally administered economy, but that bound, so that the lower bound for the amount of correct they would ultimately become negative and would remain information in the system is zero. This is because not all negative until output reached zero. Historically one is un- useful information is lost when particular resources previ- likely to observe such a phenomenon, since such a system ously allocated under the market system are reallocated would become politically unacceptable to both the elites under the centrally administered economy. Given that ini- and the population long before zero output was reached. It tially they are allocated in a system in which a lot of cor- is worth noting, however, that the history of the USSR rect information exists (inherited from the market since 1975 is not at variance with such a model. economy), they are likely to be allocated in such a way that the amount of information embodied in their alloca- Notes tion does not decline to zero immediately. However, each subsequent reallocation of those resources is likely to re- 1. I would like to thank Stanislaw Wellisz for his very useful com- duce the amount of useful information embodied in their ments and encouragement. allocation. Thus the process of "information noise" in the economy can go on forever. 193 Part VI Adjustment and Reforms: Financial Market, Foreign Trade, Fiscal Sector 16 Financial Aspects of Socialist Economies: From Inflation to Reform' Guillermo A. Calvo This paper focuses on some of the financial issues in- nature of monetary and credit policy in the medium volved in the market-oriented economic transforma- term are also discussed briefly. The need for bank super- tions in socialist countries. The central theme is the vision and the dangers of playing with short-term inter- tension between "bad" credit policy-that which gives est rates to postpone a crisis are covered. rise to inflationary spirals-and "good" credit policy- Some final remarks are presented in the last section. that which prevents unnecessary credit crunches and contributes to the efficient allocation of resources. Bad Credit: A Money Machine The next section explores the implications of bad credit by means of a simple model in which credit is used The conventional fiscal deficit as such does not play a to accumulate inventories (which are essentially waste- central role in inflation in the socialist countries. In- ful). The model is capable of depicting a situation in stead, credit is the pivotal factor. This section presents an which inflation and inventories are excessive-a depic- equilibrium model of credit-driven inflation. To simplify, tion that appears to be in line with the data emerging the possibility of transitory disequilibria is omitted (al- from several pro-reform socialist countries. This type of though they may give rise to relevant phenomena in pre- inflation is somewhat different from that associated with reform socialist countries, such as queues).2 fiscal deficits. To drive this distinction home, the credit! Assume that inventories-driven process is called a "money machine." The third section studies the implications of trying to (16-1) A = PkZ, deactivate the money machine by (1) increasing the in- terest rate on bank credit, (2) setting domestic credit tar- where M, z and Pk denote the money supply, new inven- gets and (3) removing the subsidies for inventory goods. tories and nominal price of the stock, k, of inventories, The first two measures are shown to be potentially suc- respectively. Equation (16-1) says that the central bank cessful, while the last one emerges as likely to fail and extends credit to firms to purchase (newly produced) in- even to be counterproductive. ventories3 and that the supply of money expands only The fourth section examines the role of good credit. because of that financial operation. For the time being, The main theme is that the market is likely to provide credit is assumed to be granted at zero interest.4 Equa- good credit at optimal levels only under rather stringent tion (16-1) depicts a typical money machine in pre-re- conditions. For example, it is shown that the market form socialist countries. could fail to supply efficient levels of credit if the domes- There are two types of industries: producers of inven- tic credit markets are segmented or if there are problems tory; and producers for consumption. The outputs of with the credibility of policy. new inventories and of consumption goods, denoted by Financial policy is the subject of the fifth section. The i and y, respectively, are assumed to be constant over focus is on policies that help the market achieve efficient time. The consumption sector is assumed to utilize a y solutions. One message is that monetary policy should units of inventories per unit of time (a > 0).5 Therefore, be the main instrument used to accommodate the initial at full capacity utilization, the accumulation of invento- increases in prices after price decontrol. The section dis- ries, k, is given by cusses why that accommodation should not be thought as a reactivation of the old money machine. The role and (16-2) k = z- ay - 8k, 197 where 6 (> 0) stands for the constant rate of depreciation collateral constraint (equation 16-3) is binding, then M of the inventories. = Pkk (i.e., the money supply equals the nominal value At each point in time there is a "collateral" constraint of the stock of inventories); thus, NIM = ilk, as stated in whereby total bank credit cannot exceed the value of the expression (16-6), can be obtained immediately from the inventories. Formally, latter equation and the money machine equation (16-1). In other words, the flow of new inventories governs the (16-3) M < Pkk, flow of new money, while the stock of money is directly related to the stock of inventories.6 or, defining p as equal to Pk/P, where P is the price of In particular, from equations (16-2) and (16-6), at consumption goods, steady state where k = 0, (16-4) m y employed in inventory-type industries. ml 1 - aIylz In a more realistic model, a large number of firms will Domestic Credit Target produce goods that end up in the inventory of some oth- er firm(s). Therefore, the rise in bank interest rates may Now consider the case in which im is set high enough lead to a generalized temporary loss in output. This out- to make the accumulation of inventory unattractive to come is, in principle, welcome. However, in a realistic firms. Since the collateral would be nil, there would be set-up, firms need credit for reasons other than just the no stock of credit outstanding, and, under the above ex- accumulation of inventory. Thus, the sudden fall in in- treme assumptions, money would disappear from the ventory prices may cut their access to the credit market system. However, it is possible to extend the model to (e.g., because they did not comply with the collateral the case in which the money stock has an exogenous, constraint)-a situation that would give rise to a typical central-bank determined, component. In this case the credit crunch.9 system would be fully anchored by the monetary target. Notice that for that kind of regime to hold, it is neces- The Price of Inventories sary only to generate slightly positive real interest rates for inventory. Based on the above reasoning, as the econ- Recall that the model presented in the second section omy enters such a domestic credit-target regime, the is not capable of determining nominal prices. Thus, if rate of inflation is capable of taking a sudden dive toward the authorities set Pk at a higher level, although all the level compatible with a (supposedly much lower) nominal magnitudes will take an equiproportional up- nominal target. ward jump, the equilibrium rate of inflation (under the It should be noted, however, that success in stopping assumptions of the second section) will remain the inflation is accompanied by a sharp fall in the demand same. Hence, an upward revision of the price of invento- for inventories. This situation has several implications. ries is ineffective as an anti-inflationary device. If the inventories are internationally tradable, (1) such a Suppose, instead, that the central planner exogenous- drop in demand will bring about a sharp trade surplus. ly determines the relative price of inventories in terms of However, to the extent inventories are not fully tradable, consumption goods. Let this exogenous relative price be 199 denoted by p, and assume that interest on bank credit is societies. For example, inter-firm credit has blossomed nil (as in the second section). With expression (16-4), it in countries such as Poland and Hungary, and trade can easily be verified that if p3 is larger than its equilibri- credit appears to be indispensable for modem interna- um value in the model presented in the second section, tional trade. then the collateral constraint is never binding, and firms How can bad and good credit be differentiated? This will be credit-rationed (a realistic feature of pre-reform question is perhaps one of the central issues facing the socialist economies). Hence, according to equation (16- monetary authorities in the newly market-oriented so- 5), if the monetary authority is capable of insuring that cialist economies. The trade-off is clear. (1) Although firms use credit only to buy new inventory goods (as tight central bank credit has a good chance of stopping specified in equation 16-1), the rate of inflation is given the inflationary spiral, in the process credit could turn by out to be too small, and a serious (and inefficient) fall in output could follow. On the other hand, (2) while an ac- (16-11) rc = p/lm. commodative credit policy results in full employment, inflation could be driven to levels that are unacceptably Note that the rate of inflation given by equation (16- high. 11) is larger than the value given by the right-hand-side expression in equation (16-7). The reason is that by in- The Best of Both Worlds? creasing the price of inventories, the credit needed to fi- nance the production of inventories becomes larger, a In some scenarios the above-mentioned trade-off is phenomenon that results in a larger expansion of the non-existent. Such is the case, for example, in a regime money supply and thus in higher inflation.10 of fixed exchange rates with free trade and perfect inter- This scenario provides an interesting example of mis- national mobility of capital. The latter implies that the guided policy. In this economy inflation is partly fueled domestic interest rate equals that of the international by an over-accumulation of inventories. However, if the markets. Thus, from the point of view of firms, credit is authorities attack the problem by increasing the relative granted as in the pre-reform regime. There exists a given price of inventories, more inflation will follow, even interest rate at which credit is liberally granted if solven- when the monetary authority simultaneously resorts to cy conditions are satisfied (e.g., the type of collateral credit rationing. The basic reason is, once again, that the condition assumed in the second section). The main dif- higher relative price of inventories gives rise to a higher ference with the model of that section, however, is that rate of growth of the money supply. domestic prices are tightly linked to their international counterparts. Therefore, the rate of inflation mimics the Good Credit international one, the implication being equality be- tween domestic and international real interest rates. Bad credit gives rise to inflation. This scenario is, of Moreover, since the latter are normally not negative, this course, not the only form of, or role for, credit. The mon- regime will exhibit efficient stocks of inventories." etary authority could, for example, expand the supply of This regime requires a relatively efficient financial money at an exogenous rate and let the interest rate be system. Suppose, for example, that households have per- determined "by the market." In such a case, bank credit fect access to the international capital markets'2 but is determined by the money multiplier, and in principle firms depend only on domestic financing. If the domes- low inflation can coexist with bank credit. A difficulty tic financial system is efficient, the household sector can with this kind of money supply rule policy is that the re- easily accommodate the greater demand for credit by sulting supply of money could be "too small" relative to firms. This process works as follows. The increased de- prices, and output could suffer. If prices are perfectly mand for credit tends to raise domestic interest rates flexible, this nuisance may turn out to be minor. Howev- above intemational levels, a situation that makes it at- er, if wages are downward inflexible, or if the exchange tractive for households to exchange foreign for domestic rate is fixed (as in some post-reform socialist countries), currency and deposit the proceeds in some domestic fi- then a relatively small real supply of money can give rise nancial intermediary. The flow of these funds will con- to a credit crunch and consequent loss in output. tinue until domestic and international interest rates are Too little credit is therefore a possibility. This eventu- equalized. In tum, the funds are funnelled to the firms ality is, of course, never true in the stylized model in the sector in the form of new credit.13 However, if the finan- second section because credit is assumed to play no pro- cial system is not well-developed-as seems the case in ductive role. In practice, however, credit appears to be most socialist economies-intemational funds may be essential for the operation of both capitalist and socialist too slow to come or potential lenders could call for much 200 higher domestic interest rates. Thus, a credit crunch "bad" one, lenders expect that a credit crunch will devel- could be hard to avoid. op and that the next government will confiscate part of their bank savings. The good outcome is an equilibrium, Imperfect Financial Markets because if everybody believes the policy will be contin- ued, funds will flow into the banking system at the The above discussion illustrates how easy it could be slightest sign of a differential between domestic and in- for an economy to fall into credit crunch difficulties. ternational interest rates. Hence, a credit squeeze will This subsection discusses other reasons for the failure of never happen. On the other hand, the bad outcome is the credit market. also an equilibrium, because if lenders fear the develop- Consider the realistic case where, in the pre-reform ment of a credit crunch, they will be afraid to put their period, no firm is allowed to go bankrupt. For example, savings into bank accounts. Therefore, if the stabiliza- if a firm develops financial troubles, the central bank tion program exhibits credit crunch conditions at the steps in as the lender of last resort. This situation, it is beginning, no funds will be forthcoming, and the credit believed here, is one important reason that inter-firm crunch will actually take place. Notice that the bad out- credit is such a common and relevant feature of the fi- come is an equilibrium independent of whether or not nancial regimes of pre-reform socialist economies. The the lenders are right in expecting that a long enough central bank offers full insurance, the implication being credit crunch will give rise to a policy change. that firms can lend to one another with no risk of de- fault.'4 Imagine now that the authorities announce that Financial Policy the central bank will stay away from the credit insurance business and that that policy announcement has some The central message of the previous discussion is that credibility. How will the "market" react? while a passive credit policy may give rise to inflation, Operation under a full insurance system implies, the stoppage of official credit flows may be associated most likely, that lenders have had little incentive to col- with output losses that are not called for by efficiency- lect information about borrowers. Thus, the market type considerations. A credit crunch could develop for starts off with very little information. Consequently, the reasons ranging from the lack of adequate financial in- removal of a lender of last resort has implications that go termediaries to just the expectation on the part of poten- beyond monetary policy-it has microeconomic infor- tial lenders that a credit crunch could occur. mational aspects as well. Thus, in the short run a reduc- tion in official credit to firms is unlikely to be matched Early Stages of Reform by an expansion of private credit (including, in particu- lar, inter-firm credit). Eventually, private credit may Typically, a market-oriented reform in the socialist emerge full bloom, but there is likely to be a period of economies starts with a radical program of price liberal- credit shortage during the transition that is not trivial. ization. The rationale is that free prices are indispens- The duration of this transition is likely to depend on the able for efficient resource allocation. pace, breadth and depth of the financial reform. In this respect, the recent Polish stabilization plan Another source of imperfection in the credit market suggests some useful lessons. A salient characteristic of is less-than-complete credibility of policy. There are the Polish experiment is that real credit to socialized many reasons why a policy announcement is not fully firms (in terms of the cost of production) appears to have credible (see Calvo 1989), not the least of which is that fallen by more than 30 percent. This decline suggests the politician who makes the policy announcement is that a credit crunch could partly account for the esti- likely to have a finite-not to say short-political life, mated 20-30 percent drop in the output of the socialized particularly in cases involving radical economic reform. sector. The case of Poland is very illuminating, because Suppose, for instance, that the policy is expected to be households hold a substantial amount of savings in dol- changed if it leads to a credit squeeze (and resulting loss lar-denominated assets and, as shown by Calvo and Cori- in output) for a period longer than a certain critical one. celli (1990), few of the assets for which there are official Assume, in addition, that if there is a change in the pol- records (i.e., dollar-denominated bank deposits) have icy regime, potential lenders expect that the govemment been converted into zloty accounts. This case resembles will go back to the old ways of paying negative real rates the one discussed in the third section, where a credit of return on bank savings (through, for example, partial crunch develops even when the economy as a whole has confiscation of deposits). There are at least two possible access to intemational funds. equilibrium outcomes. In the "good" outcome, lenders The Polish experiment supports to some degree the are confident of the success of the present policy. In the view that price reforms in socialist economies may have 201 to be accompanied by an aggressive expansion of credit prices and the exchange rate rise without a substantial at the beginning of the program-particularly when the loss of reserves. Attempts to prevent the loss in reserves price reform leads to an initial drastic increase in the by, for example, increasing the interest rates on bank de- price of some key factors of production. A contrary view, posits are likely to backfire (Calvo 1990a). however, is that such a credit expansion might not deac- Consider now the case in which credit is insufficient tivate the money machine because that type of credit re- to sustain full employment. The view postulated here is sponds to price rises and thus is like the bad credit that that belated attempts to provide more central bank cred- originally set the money machine in motion. it could be dangerous because they may be taken as a sig- A closer look at this issue reveals that the two points nal that the monetary authorities have resorted to of view are not mutually contradictory. In fact, both offer reactivating the old money machine. Thus, prices may relevant insights. Care is necessary when distinguishing start to rise as in the pre-reform regime, a situation that between stocks and flows. High inflation is fueled by in turn will generate another credit crunch unless the large flows of money into the system, not by a possibly authorities come back later with even greater official large infusion of money at the time the prices of raw ma- credit, and so on. One alternative that deserves attention terials go up (i.e., by a stock change in the money supply is to lower the costs of production. This measure could at the beginning of the program). The latter avoids the be implemented by reinstating some input subsidies or, credit crunch and cannot per se generate inflation. in some cases, by appreciating the domestic currency. Hence, the first view, which says that an initial credit! Since this type of policy goes to the heart of the credit money expansion may be beneficial, is correct, but at the crunch policy, it is likely to boost production. Further- same time the second view is also right in warning more, if the country has been facing a sizable loss in out- against continuing this money accommodation policy in put, the stimulus to output could go a long way toward the future. Indeed, the main message of the second point alleviating the government's budget deficit that this pol- of view is that the authorities should make absolutely icy may entail. clear that the initial credit accommodation is a one-time event. Firms and individuals should not expect credit! TheMedium Term: Sequencingandlnterest-ratePolicy money accommodation to be repeated in the future, un- less the system is subject to a new exogenous price-level The Latin American experience strongly suggests shock (or any other exogenous shock against which leaving financial liberalization to the later stages of a re- monetary accommodation is ex ante agreed, and clearly form program. In this respect, the central observation is announced, to be desirable). that early financial liberalization may magnify the dis- A lesson of the Polish stabilization experiment is that tortions. These distortions may be the result of: (1) in- a reform that entails a large increase in the prices of complete dismantling of the commodity taxes or some key raw materials should perhaps be accompanied subsidies; (2) market imperfections, such as the pres- by a comparable, one-time increase in central bank cred- ence of monopolistic firms-a relevant consideration for it to firms. Furthermore, the central bank could help reforming socialist economies; or, perhaps more subtle tidy up the initial conditions by substituting central but not less relevant, (3) government policy announce- bank credit for inter-firm credit. Central bank credit may ments that lack total credibility with the private sector. be liberally granted and can even take the form of a The following remarks focus on the credibility-related transfer to the firms sector. The higher supply of money distortions, since they are likely to be relatively less fa- so generated may not fuel further inflation to the extent miliar to the reader. that the firms keep a higher stock of monetary balances Credibility distortions were mentioned earlier in con- to carry out their ordinary transactions. nection with the credit crunch difficulties, although in A difficulty with the proposed solution to the initial those instances a free financial system may play a wel- credit crunch problem is that before the reform takes fare-enhancing role. There are, however, cases in which place there is little or no information about the post-re- the interaction between the lack of credibility and the form demand for credit. Hence, errors are inevitable. free financial markets is counterproductive. Suppose What is the best policy response if errors are detected? there is incomplete credibility about the success of the Following are some tentative thoughts. anti-inflationary program and that, as a consequence, First, if the initial expansion of credit proves exces- the rate of devaluation expected by the private sector is sive, prices will overshoot, or greater than programmed larger than the one contemplated in the program. The reserves will be lost (if the monetary authority inter- latter implies that the nominal rates of interest on de- venes in the foreign exchange market). In this instance, posits and loans will tend to be higher than if there were a good case can be made for biting the bullet and letting full credibility. At the same time, however, the domestic 202 prices of tradable goods will tend to be governed by the array looms. Therefore, the probability and depth of actual exchange rate, the implication being that their systemic financial crises may be greatly enhanced unless rate of inflation may follow closely the actual rate of de- the financial institutions are well-supervised. valuation rather than the expected one. Since, as argued Finally, it is advised that special caution be taken above, the nominal interest rate reflects the expected about paying interest on bank reserves. Several Latin rate of devaluation (which, by assumption, is higher American countries have fallen into this trap. Remuner- than the actual rate), it follows that the (ex-post) real in- ated bank reserves allow banks to pay interest on depos- terest rate on tradable goods will tend to be higher than its, even if the deposits are never lent. This practice if the stabilization program was fully credible. How high provides the central bank with a new instrument with the ex-post real interest rate is will depend on the gap be- which to stave off currency runs or to counteract less tween expected and actual devaluation, i.e., the credibil- dramatic portfolio shifts that would result in higher ity gap. This gap brings about unplanned redistributions black market premia. Furthermore, a rise in deposit in- of wealth from borrowers to savers. If firms are net bor- terest rates induces an increase in the demand for cen- rowers, the credibility gap implies a redistribution from tral bank liabilities that, at least for some time, may help firms to households. As the previous discussion sug- reduce the fiscal deficit. Unfortunately, however, the fis- gests, this redistribution is particularly harmful to firms cal deficit actually gets worse, or public debt rises fur- producing tradable goods, some of which may be driven ther, as the interest on reserves has to be paid. Moreover, into bankruptcy that is unrelated to any efficiency con- quelling a crisis by means of this interest rate instru- sideration. ment is likely to result in even more critical difficulties This example shows that financial intermediation in the future (see Calvo 1990b). could have harmful effects on the economy when the policies do not enjoy full credibility, even though other Final Remarks better known distortions have been eliminated. This sit- uation is very unfortunate because reforming socialists Credit is central to the functioning of market econo- economies-given the novelty and extreme nature of mies. It is the vehicle by which the surplus of one agent their reforms-are likely to have serious credibility can be utilized for production or for consumption by problems, at the same time that, as argued above, they other agents in the economy. It is essential to the effi- could greatly benefit from access to credit to relieve cient allocation of resources. However, by their very na- credit crunches. ture, credit contracts entail one party transferring Reforming socialist economies are notorious for their resources to the other-and receiving in exchange only lack of financial services. Thus, despite the above-men- a promise of future transfers in the opposite direction. tioned credibility complications, it is posited here that fi- Thus, credit markets require trust and, above all, well- nancial institutions should be developed. However, the defined property rights. credibility issue suggests that the financial policy should By all accounts, the credit markets in reforming so- be aimed at preventing the magnification of distortions. cialist economies are still heavily dependent on the cen- This objective may be aided by implementing conserva- tral bank. Even in cases in which inter-firm credit has tive banking and financial policies. mushroomed, the central bank has insured much of In that respect, it seems advisable to provide incen- those loans by operating as a lender of last resort. Thus, tives for longer run financial deposits and loans, since in those countries the trustworthiness and validity of long-term assets may help reduce the probability of fi- property rights necessary for the functioning of credit nancial crises (see Calvo 1990b). To avoid the syndrome markets have still to be tested. In the short run, enter- of high interest rates, loans and deposits could be in- prises in reforming socialist countries will find them- dexed to the "market" exchange rate.16 The indexation selves still heavily dependent on central bank credit. In could be phased out later as credibility is established. Fi- fact, such a dependence could be more pronounced dur- nally, the financial institutions have to be closely super- ing the early stages of the reform. The reason is that the vised. There is a worldwide tendency for the fisc withdrawal of the central bank as lender of last resort (including the central bank) to bail financial institutions may substantially increase the riskiness of the inter-firm out when they face so-called systemic problems. This loan market. practice tends to remove the incentives for financial in- Consequently, the first stages of a market-oriented stitutions to react to perceived global financial prob- reform program are likely to be characterized by a credit lems. In addition, the "free insurance policy" implicitly shortage. At first sight, this condition may appear para- granted by the fisc may actually embolden financial in- doxical because the typical complaint of many of these stitutions to take even larger risks when a systemic dis- economies-particularly those that suffer from high in- 203 flation-is that credit is excessive. These two opposing 3. In practice, bank credit for purchasing "old" inventories is also roles of credit, and their mutual tension, provided the a likely possibility. Allowing for that type of credit would give rise to an backgrund,tothe present paper. even more powerful money machine than the one discussed in this sec- aaCKgrOUnd to tne present paper. tion. However, such an extension of the model is not discussed here be- This paper looked at a simple example of what was la- cause it is not pertinent to the central points. beled bad credit. Such credit, which is granted by the 4. Incentives for firms to "demand" this type of credit are dealt central bank, gives rise to both inflation and the misallo- with later in this section. cation of resources. In the example presented here, the 5. To simplify the exposition, the economy is assumed to be closed misallocation took the form of an overaccumulation of to international trade. However, in a later discussion this assumption is relaxed. inventories. However, as pointed out at the outset of this 6. The price of inventories, Pk, is involved in these relationships. section and as was discussed in the paper, there is also However, it washes out completely because it enters multiplicatively in good credit. It may not, however, materialize in the both relationships. short run to the extent necessary for a successful pro- 7. Equation (16-7) implies that the rate of inflation declines as the of reform. flow of inventories, 2, increases. This result is somewhat counterintu- gram ol relorm. itive because the accumulation of inventories is the main factor behind In particular, the paper emphasized the importance the money machine. However, as the discussion around equation (16- of working capital-the relevance of ensuring that there 6) shows, although inflation is an increasing function of the flow of is enough liquidity in the hands of enterprises to con- money, it is also inversely related to the stock. For this reason, in the duct their regular operations. In this regard, it was ar- present model, 7t is equal to 2/k, and not just to z From equation (16- gued that the removal of input subsidies could generate 2), at steady state (i.e., k = 0), kIz = (1 - 07)/o1. Hence, at steady state the ratio i/k declines as the flow of inventories, 2, increases, a phenom- a serious liquidity crunch that-in view of the character- enon that explains the above (possibly) counterintuitive result, It will istic imperfections in the credit market in the first stages be shown below, however, that the association between inflation and of a reform program-could not be offset by the private the flow of inventories would be positive if the central planner fixed the sector. As a result, the productive sector could be choked relative price of inventories with respect to consumption, p. off just because there are not enough financial instru 8. Notice that absolute prices, P and Pk, are not determined by the model. This situation is a well-known feature of interest rate-based ments to carry out the productive transfer of resources. monetary policy (see Sargent and Wallace 1975 and Calvo 1983). In the In addition, this paper showed that during the first stag- present context an easy remedy is to assume, realistically, that the cen- es of a market-oriented reform, the credit market could tral planner chooses, for example, the initial nominal price of con- be further weakened by the sometimes enormous reallo- sumption (i.e., the nominal price of consumption at some initial time cation of resources required by efficiency. This possibili- 0, say). Mterwards the price of consumption is determined by the infla- tion equation (16-7), while that of inventories is given by equation (16- ty was illustrated by the sharp fall in demand for 5) inventories during the early stages of the plan when the 9. This issue will be explored later. earlier money machine induced an overaccumulation of 10. Under the present assumption, the conventional result in so- inventories. cialist economies-That inflation is an increasing function of the flow The major general point of this paper is that financial of inventories-is recovered. 11. Furthermore, bankruptcy for not satisfying the credit con- markets are unlikely to spring to life any time soon after straint (equation 16-4) would be avoided if the inventories were inter- the initiation of market-oriented reform and to contrib- nationally tradable. ute to a much better allocation of resources. Further, 12. This case is not as extreme as it sounds: in several socialist the paper suggests that if monetary/credit policy relies economies households have a sizable share of their portfolios in the too heavily on a credit crunch, while inflation may be form of foreign exchange. Thus, borrowing from the international markets can be closely emulated by, for instance, households lending slowed, it will be at the cost of equally, if not more, dis- to firms, with the latter using the proceeds to buy capital goods abroad. turbing economic problems, such as deep and protracted The country as a whole lowers its foreign financial assets, a condition recession. that is equivalent to the country increasing its international financial liabilities (i.e., international borrowing). Notes 13. See Frenkel and Razin (1987) for a fuller exposition of this case. 14. Pre-reform firms have few investment opportunities aside 1. I am grateful to Fabrizio Coricelli for his useful comments. The from holding inventories. Hence, risk-free inter-firm loans may be at- views expressed in the paper do not necessarily represent those of the tractive even though the inflation and interest rate controls could re- International Monetary Fund. sult in negative real interest rates. 2. Although important to a full description of the inflationary pro- 15. The cases of Argentina and Chile during the late 1970s and ear- cess in socialist economies, a focus on rationing equilibria may lead to ly 1980s provide clear-cut illustrations of this type of impasse. See, for a failure to see some of the fundamental monetary forces behind infla- example, Calvo (1986) and Edwards (1985). tion. 16. The same applies to government debt obligations (see Calvo 1990a and b). 204 References Calvo, Guillermo A., and Fabrizio Coricelli. 1990. "Stagflationary Ef- fects of Stabilization Programs in Reforming Socialist Countries: Calvo, Guillermo A. 1983. "Staggered Prices in a Utility-Maximizing Supply Side vs Demand Side Factors." International Monetary Framework." Joumal of Monetary Economics 12:383-98. Fund and World Bank. Washington, D.C. August. . 1986. "Fractured Liberalism: Argentina under Mart!nez de Edwards, Sebastian. 1985. "Stabilization with Liberalization: An Eval- Hoz." Economic Development and Cultural Change (April):511- uation of Ten Years of Chile's Experiment with Free-Market Poli- 33 cies, 1973-83." Economic Development and Cultural Change (January):223-54. 1989. "Incredible Reforms." In G. Calvo, R. Findlay, P. Kouri and J. Braga de Macedo, eds., Debt, Stabilization and Development. Frenkel, Jacob A., and Assaf Razin 1987. "The Mundell-Fleming Model Oxford: Basil Blackwell, Inc. a Quarter Century Later: A Unified Exposition." International Mon- etary Fund, Staff Papers 34 (4)(December):567-620. 1990a. "Are High Interest Rates Effective for Stopping High Inflation?" International Monetary Fund. Washington, D.C. Febru- Sargent, Thomas J., and Neil Wallace. 1975. "Rational Expectations, ary. , the Optimal Monetary Instrument, and the Optimal Money Supply Rule." Joumal of Political Economy 83:241-54. 1990b. "Credibility Crises and Economic Policy." Interna- tional Monetary Fund. Washington, D.C. January. 205 17 Adjustment, Trade Reform and Competitiveness- The Polish Experience Jan W Bossak The first section of this paper reviews the primary de- Outward-Oriented Development: The velopment strategies pursued in recent decades. It Emerging Orthodoxy makes the point that the inward-looking strategies pur- sued in Latin America have generally been unsuccessful, The Success of Outward-Oriented Development while the outward-looking ones found in some East Strategies Asian countries have led to significant and stable eco- nomic growth. Many economists are urging the Central In the 1980s, after two to three decades of rapid and Eastern European countries to heed this lesson and growth, Latin America, Africa, the Middle East and Cen- to pursue export-led growth. The section concludes by tral and Eastern Europe registered a decline in economic defining what an outward-oriented strategy consists of. development. The growth of the previous decades had all The following section looks at the challenge of adjust- too often been founded on development strategies that ing an economy while maintaining growth. It presents had failed to emphasize economic efficiency and interna- some characteristics of successful growth-oriented poli- tional competitiveness and that had relied heavily on fi- cy packages, emphasizing both the need for internal nancing from abroad (World Bank 1989). While in the commitment by governments to achieving stabilization majority of the countries in Latin America in the 1980s, and adjustment with the least possible disruption to restrictive, deflationary domestic policies and real devalu- growth and the role that the developed countries must ations reduced imports and increased exports, a situation play in furthering the adjustment effort. A key issue is that often led to trade surpluses, their macroeconomic ef- the inability of countries to develop the required invest- forts to improve the balance of payments resulted in re- ment resources because of their huge debt burdens. cession and a decline in per capita incomes. After several A number of the issues raised by adjustment in two years of trying to adjust their economies, most of the areas-the importance of being clear as to the purpose heavily indebted countries not only failed to resume of adjustment and the appropriate content of an adjust- growth but also to reduce inflation and foreign debt. ment program and other factors related to successful In East and Southeast Asia, in contrast, the newly implementation-are reviewed in the third section. The industrialized economies generally pursued sound recent policy reforms are described next, followed by a macroeconomic policies and tried to maintain the discussion of the foreign debt problem and the need for competitiveness of their economies, especially in terms of assistance by lenders and the developed world. exports. Generally they adapted well to the shocks of the The main conclusion of the paper, presented in the fi- 1970s and early 1980s. To a large extent, their success was nal section, is that the countries of Central and Eastern a result of their maintaining high levels of savings, Europe need to implement a long-term policy that ad- investment and export promotion. In the 1980s, when they dresses the structural issues and emphasizes exports, tried to improve their fiscal balances in combination with a and they have to stand behind the effort. However, it is reduction in their investment ratios, they did not unlikely that any policy package, no matter what the de- experience a substantial decline in the rate of growth of gree of governmental commitment, can succeed without total factor productivity and technological progress. As a debt relief and an inflow of external funds. result of changes in relative prices, levels of spending and 207 profitability, their international competitive capacity oriented export-led growth, export promotion and liberal- increased, and they experienced substantial export ization are not synonymous. An outward-oriented strategy surpluses. implies liberalization of the trade and payment regimes in As noted, none of the adjustment efforts undertaken order to align domestic with international prices and to en- in Latin America produced results comparable to those courage both exports and imports by setting a realistic ex- in the successful East Asian countries. However, recent- change rate and removing import restrictions. According ly those Latin American nations that pursued sounder to Bhagwati (p. 285), an outward-oriented strategy is main- fiscal policies recorded a moderate improvement (Chile, ly a matter of setting price incentives in such a way as to en- Mexico and Colombia). Some of the countries that sure that the home market does not become more lucrative strengthened their fiscal balance also started to liberal- than the foreign market. ize their foreign trade and used devaluations in the real An outward-oriented strategy has two basic themes: exchange rate to improve their export price competitive- liberalization; and an export development strategy. Lib- ness, increase the volume of exports and restrict im- eralization is in fact a laissez-faire and non-intervention- ports. ist policy involving the eventual removal of all major The widely acknowledged success of outward-orient- economic distortions related to government activity not ed development and adjustment has drawn the attention only in foreign trade and payments but also in the do- of many Central and Eastern European countries. Ac- mestic economic system and policies. Liberalization is cording to Sachs, a new orthodoxy is emerging that links the opposite of protectionism. In an extreme case, it re- recovery in a debtor country to a shift to outward-orient- jects the idea of active trade and industrial policies and ed development strategies designed to produce export- assumes the superiority of exchange rate adjustment led growth. Followers of this new orthodoxy describe the and macroeconomic policies. policy content of these strategies as consisting of: It is doubtful that there is a universal recipe for solv- ing current and long-term development problems * liberalization of the trade regime through liberalization. Countries are at different stages * real exchange rate devaluations and unification of of economic development and face different economic the exchange rate conditions. They vary not only in their level of per capita * privatization of state enterprises income but also in the level of development of their mar- * maintenance of a small budget ket institutions and forces. Therefore, many advocates of * reduction in government intervention in the the liberal approach also see the need for country-specif- economy ic program designs, within the context of an outward- * deregulation and oriented strategy. * demonopolization and the fostering of competi- Adjustment programs are also not synonymous with tion. macroeconomic policy, liberalization and a package of government non-interventionism. To the contrary, ad- According to Khan (1990), the International Mone- justment programs are part of a conscious state policy in tary Fund (IMF) and the World Bank have been attaching which liberalization of the economic system and policy ever greater importance to the promotion of exports, plays a crucial but not exclusive role in creating a sound mainly by advocating real exchange rate devaluations to basis for future balanced and effective long-term growth. raise the profitability of exports and supplies. To support Advocates of liberalization are cautious about the or- liberalization efforts, the World Bank is gradually in- der of liberalization and the coordination of liberaliza- creasing the role of structural adjustment loans (SALs) tion and macroeconomic stabilization policies. An and sectoral adjustment loans (SECALs). Both institu- important issue is the way in which liberalization is im- tions are cooperating with interested countries in work- plemented: step-by-step or via radical, comprehensive ing out medium-term adjustment programs to support therapy. The most difficult and controversial question is outward-oriented structural adjustment. how quickly to liberalize the capital market. Because premature liberalization is dangerous, it is often argued Some Definitions that liberalization of the capital account should be virtu- ally the last step in a liberalization program. The outward-oriented developing economies, especially An export promotion or export development strategy those of East Asia, have certainly outperformed the econo- differs from liberalization in that it gives a clear priority mies of Latin America. Therefore, it is plausible to link to aggressive government promotion of exports while of- much of this superior performance to their development ten delaying the liberalization of imports and subse- strategies and the role of export policy. However, outward- quent adjustment of domestic to international prices 208 and liberalization of capital movements, which are intro- succeeded both in restoring adequate net investments duced only gradually following a build-up of internation- and in resuming growth. The Brady plan indicates that al reserves. Those countries that intend to pursue an at last the US government and creditor banks alike have outward-oriented strategy but are burdened by a large concluded that debt reduction has to be part of a solu- foreign debt and are afraid of suddenly exposing their in- tion to the debt crisis (Antowska and Malecki). A debt re- efficient, inflexible economies to international competi- duction initiative has official support and funding from tion often prefer an export development strategy. the IMF and the World Bank, provided the countries in- An export development strategy often involves con- volved undertake effective adjustment programs. siderable government activity. This intervention can be A key question, however, is what "effective adjust- of a great importance in establishing the credibility and ment programs" means. According to Barber Conable, commitment of the government, which are vital in in- president of the World Bank, countries must undertake ducing investors to risk investing in the development of stabilization efforts within the framework of an adjust- export industries. Typically, the strategy calls for state ment program that permits sustainable growth. A suc- encouragement of savings, investment and technologi- cessful adjustment program is one that achieves an cal progress. According to Michel Camdessus, the man- appropriate macroeconomic balance, that simulta- aging director of the IMF, the countries that are best able neously raises the level of output obtainable from exist- to protect growth during the difficult adjustment period ing resources, and that discourages excessive spending are those that maintain a strong export performance, and inefficient investments that are not growth-enhanc- that keep domestic savings and investment from falling ing. To be successful, growth-oriented programs have to sharply and that are able to share the adjustment burden overcome such problems as distorted factor and com- between increased supply and reduced aggregate de- modity prices, inefficient public sector enterprises and mand. This package requires not only a set of macroeco- protectionism extended to inefficient and uncompetitive nomic policies but also structural ones (including industries. In addition to a more efficient allocative elements of liberalization and industrial policy) that en- mechanism, which is critical to achieving sustainable hance international competitive capacity. growth and the development of competitive industries, Export development strategies should eventually lead domestic savings must be raised and additional resourc- a country out of external disequilibrium and import re- es obtained from abroad to diminish the cost of servicing strictions and on to liberalization. The success stories of the foreign debt. Japan, Korea, Taiwan and Malaysia obviously confirm The right kind of adjustment will not take place by this observation. Moreover, their experience indicates chance. It requires that a government implement a set of that export development strategies may be efficiently macroeconomic and structural policies that encourage pursued with active government trade and industrial saving, sound investment, cost-saving technology, fi- policies and in some cases in the presence of regulated nancial stability and exports. It is also important that financial and capital markets (Bossak 1990). It also indi- creditors assist the countries by providing adequate fi- cates that successful development may be furthered by nancing, maintaining open and growing markets and high quality government, macroeconomic and systemic fostering appropriate exchange and interest rates. reforms and structural policies (Bossak 1990). The key issue in the pursuit of adjustment and growth is to find a combination of macroeconomic policies that Adjustment with Sustainable Growth: can attain the stabilization objectives while also support- The Challenge ing structural adjustment with the least disruption to growth. A key point is that stabilization alone does not According to Camdessus, a central mission of the guarantee growth. Moreover, the specific policy package Fund is to support adjustment for durable growth. In designed to induce structural change and growth will countries in which inflation and balance-of-payments vary from country to country. crises are major problems, it is extremely difficult to Greater economic efficiency and improved productiv- achieve adjustment with growth. Before 1982, about 2 ity of investments require the elimination of macroeco- percent of the annual gross national product (GNP) of nomic distortions, as well as measures to improve the the highly indebted countries consisted of resources efficiency of the public sector. Constantine Michalopu- from abroad. Since then, these countries have been los, senior economic adviser in the World Bank's Policy transferring roughly 3 percent of their GNP abroad a and Review Department, is of the opinion that two sets year (World Bank 1989). Domestic savings would need to of policies are important to stimulate a shift in resources rise by 5 percent of GNP to offset this change in net and an increase in the supply of tradables: a macroeco- transfers. Despite extraordinary efforts, no country has nomic policy mix; and other policies aimed at promoting 209 the mobilization of resources, an increase in the efficien- be combined harmoniously to respond in the most ratio- cy of their use and the restructuring of production in fa- nal and efficient way to the changing environment. vor of tradables. If the thesis that reforms must be tailored to individ- ual cases is accepted, it is important to recognize that ad- Some Issues Raised by Adjustment justment of a given system to new developments, or a transformation, is not the same as reform. Reform The Purpose ofAdjustment means modifications of a system without changing its core elements and the logic of its functioning. Transfor- A country's adjustment program can also be seen as a mation means moving from an old system to a new one response to the ever- changing environment in which a (Bienkowski and Bossak 1990). In the case of the Central given society seeks ways and means to achieve economic and Eastern European countries, the adjustment is be- progress. Adjustment may be required in response to ex- ing accompanied not by reform but by a fundamental ternal and internal conditions. These conditions may be transformation into a new economic order. Transforma- unexpected, sudden events or the result of long-term tion means not only dismantling of the old systems but processes. They may involve not only new economic building new ones in their place. phenomena but also fundamental political changes, as in Because the process takes time and is politically high- Central Europe. It is of fundamental importance to diag- ly sensitive, a serious question arises over what ways and nose accurately why adjustment is necessary as a basis means to use at the time of the systemic transformation for designing the adjustment program. That is, any ad- to secure efficient structural adjustment. Should real justment program must be based on the answer to the structural adjustment be postponed until the systemic fundamental question-adjustment for what purpose? transformation is completed, or should the two be pur- The program must then be directed toward that end. sued in parallel by temporarily substituting state inter- Typically, the objectives of economic adjustment in- vention for the market mechanism? clude: Another important question is whether the macro- economic stabilization measures should be followed by a • price stabilization transformation of the economic system (including liber- * reduction or elimination of the current account alization of the trade regime) that is comprehensive deficit or achievement of a surplus to service and reduce enough to overcome the following: the foreign debt e increase in economic efficiency and the level of * recession, which to a large extent is caused by the technology macroeconomic stabilization measures * greater flexibility and adaptability in order to se- * the structural problems that result in a misalloca- cure a dynamic comparative advantage and to make the tion of resources economy less vulnerable to external shocks and * technological backwardness e more savings and investment. * monopolies and limited competition a underdevelopment of the market infrastructure All these objectives are components of a competitive and economy capable of achieving sustainable, long-term, * the debt overhang. balanced and open growth and of securing gains from the international division of labor (see figure 17.1). The response of Central and Eastern Europe to ab- sorption and switching measures has been less elastic The Implementation ofAdjustment than that in market economies because of supply rigidi- ties. The institutions, economic agents and professions The causes for and objectives of adjustment have im- that are taken for granted in market economies have to portant implications for what should be done. Only be re-established and developed. Human capital and when this information is clearly understood should the management have to be upgraded. To resolve these prob- policy mix, the role of macroeconomic policy, the scope lems, governments need to adopt a set of macroeconom- of the needed reforms of the economic system, and the ic measures and structural policies that address them industrial and trade policies be considered. Further inte- appropriately using market mechanisms. The set of mea- gral parts of these considerations are who will be respon- sures should be aimed at mobilizing resources and sav- sible, how the adjustment programawillbeimplemented, ings and channeling them into the most productive and what the role of the state and market forces will be. uses, support cost-saving technology, increase total fac- With respect to the latter, a key point is how the two can tor productivity, ensure factor mobility and high eco- 210 Figure 17.1. Basic Components of the Competitive Strength of a National Economy - r 4 Openness of the Trade and External Factors A Payment System Internal Factors | Domestic 1International Rto |Equilibrium 1 -Competitive Strength Exchang Economic Pblcies Mobilization of Economic Efficiency Econotic Flexibility Resources, Human * and and Energy and Knowledge Technological Progress Innovativeness Soure: Bossak (1984). nomic flexibility, encourage the development of highly (1) liberalization of virtually all prices to clear de- efficient and competitive export-oriented industries and mand and supply in the market and remove the price secure financial stability. distortions to the rational allocation of resources. (2) introduction of the internal convertibility of the Policy Reforms under the New Polish Regime zloty and the setting of a unified, undervalued rate of exchange for the zloty, abolition of the preferences fa- In a report prepared by the United Nations Develop- voring exporters and introduction of new customs ment Programme and the World Bank, the main conclu- tariffs. sion was that macroeconomic disequilibrium and (3) restrictive income policies, based on a barrier-type systemic problems seriously constrain trade policy. In- tax levied on wages and increases in salaries above the ternal macroimbalances in the form of shortages, excess established coefficient. demand and liquidity were seen as the main factors in- (4) restrictive monetary policies aimed at reducing hibiting the expansion of exports. This report confirmed the financial liquidity of enterprises and promoting the results of earlier studies prepared at the Foreign stabilization and growth in zloty saving deposits. Trade Research Institute and World Economy Research (5) restrictive fiscal policies that limit spending, in- Institute that the main factors inhibiting trade perfor- crease revenues and press for surpluses in para-bud- mance are systemic in character and are associated with getary funds. macroeconomic disequilibrium; that is, they are outside of trade policy. Parallel with efforts to stop inflation and stabilize the The stabilization program and systemic reforms un- economy, the Polish government undertook measures dertaken by the Solidarity-led government beginning in to transform the economic system. First and foremost January 1990 have changed the situation radically. The among them were liberalization of the regime and macroeconomic policy contains five main elements: reform of the financial and budgetary systems. These 211 systemic measures were followed by the introduction of modernization, commitment to the company and job se- new legislation aimed at clearing the way for the curity. In addition, the growing number of unemployed transformation of ownership and development of the and the development of a labor market will have an im- wider range of market economy institutions and portant impact on industrial relations. mechanisms needed to establish the institutional The privatization of state-owned enterprises will be infrastructure for the factor markets. either voluntary, in which case the initiative comes from One aim of the transformation of the economic sys- the enterprise, or obligatory, in which case it comes tem was to change the basic relations among the main from the treasury, typically triggered by the enterprise's actors in the economic system: the state, enterprises continuing losses or failure to pay a dividend. Small in- and labor. The intent was to change labor-management dustrial and commercial enterprises will be sold or relations at the workplace and to create new types of re- leased. Non-viable enterprises without prospects will be lations between enterprises. The Solidarity-led govern- liquidated. Distressed but seemingly viable enterprises ment is firmly committed to enterprise restructuring, will be restructured with the support of a state agency. privatization and private sector development. Without This process will lead either to their sale or transforma- dominant private ownership of the means of production, tion into a joint-stock company. the factor markets cannot develop, and market forces Parallel with these activities, a demonopolization and therefore cannot lead to an efficient allocation and use of deregulation program will be implemented to foster resources. competition. Special attention is to be given both to the The Memorandum of Development Policy published demonopolization of wholesale and retail trade, agricul- in September 1990 assumes major changes in the own- tural trade and the distribution sector and to the dereg- ership structure of the Polish economy through a pro- ulation of telecommunications, banking, insurance and gram of privatization of state-owned enterprises, transport. promotion of the formation of new private companies Development of the private sector by encouraging and a radical improvement in performance. Legislative new entries and promoting investments, including by initiatives have been undertaken to amend, among oth- foreign capital, will be supported by the removal of the ers, the Civil Code, the Civil Procedure Law, the Law on remaining legal barriers, the provision of institutional Land Management and Real Estate Expropriations, the and infrastructural support and access to credit and Housing Law and the Law on Financial Management of technical assistance. Success in this field is closely State Owned Enterprises. All those amendments aim to linked to an improvement in the banking and credit sec- reduce the legal and regulatory obstacles to the expan- tor and to growth-enhancing tax reforms. sion of private sector activity. An important element of enterprise restructuring, The government regards privatization as a key ele- privatization and private sector development is the ment in the transition to a market economy and an im- founding of a capital market and the development of a portant stimulus for balanced economic growth. The modem banking system. Law on Privatization of State Owned Enterprises pro- The structural reform and transformation program vides a general framework for privatization. The main includes reform of the trade and currency and exchange economic objectives of privatization are to encourage rate regimes. The managed trade system, which is based savings and more rational, economic use of resources, to on licensing and predominantly centrally allocated for- ensure the allocation of resources along the lines of eign exchange, with a complicated system of subsidies, comparative advantage, to stimulate technological, or- taxes and countervailing duties, has been transformed ganizational and managerial progress and to raise total into a liberal, transparent, market-based system that factor productivity. Transformation of state-owned en- opens the country to foreign competition and world terprises into joint-stock companies owned by the trea- price signals. The government is committed to main- sury, institutional and private investors will enable the taining this open trade regime and an appropriate ex- basic factors of a capital market to be introduced. Estab- change rate to ensure the international price lishing this market is an important prerequisite for en- competitiveness of Poland's tradables. The liberal, open hancing flexibility and competition and therefore for trade system is to complement the liberal price-setting achieving greater allocative mobility and more efficient system and provide the competitive pressure to combat use of resources. the monopolistic position of many enterprises. The gov- Privatization, private sector development and enter- ernment assumes that an open trade system will support prise restructuring will change attitudes and human be- efficient enterprise restructuring and the allocation of havior, especially toward savings, innovation, risk- resources according to comparative advantage. taking, work discipline, the role of the labor unions, 212 Liberalization of the trade system has opened the way technological progress. According to Nasilowski, with- to an almost complete elimination of quantitive restric- out debt forgiveness Poland will have to achieve a 5 per- tions and the administrative allocation of foreign ex- cent rate of growth in national income to service the change. Moreover, the government has abolished import debt and keep real incomes stable. licensing for practically all traded products. There is To attract foreign capital, Poland needs, beyond the some licensing of exports to ensure an adequate domes- measures already taken, to develop a sound vision for tic supply of products whose prices are still state-con- how to solve the debt problem and to foster the develop- trolled and to prevent the re-export of some products ment of competitive industries and modem infrastruc- imported under special arrangements with the Council ture. The assumption that the less restrictive on Mutual Economic Assistance (CMEA). macroeconomic policies and systemic reforms alone will Liberalization of the trade regime has been accompa- lead to growth in excess of 5 percent seems unrealistic. nied by reform of the customs tariff. The average tariff It seems particularly unrealistic given the Persian Gulf for manufactured goods is around 12 percent, with lower and Soviet oil crises, which constitute very serious exter- tariffs for imported capital goods. The main objective of nal challenges to the fragile democracy and pro-market the tariff reform has been to stimulate efficiency by low- liberal economic transition and stability in Poland (and ering the level of effective protection. in Central Europe generally). Both crises create supply The government has also decided to liberalize the bottlenecks and lead to a forced transfer of domestic sav- transfer of dividends for foreign investors by simplifying ings abroad that inflates local prices and crowds out pri- the procedures for profit repatriation and making invest- vate investment. ments more attractive. The direct losses the Gulf conflict has caused Poland and the other Central and Eastern European economies The Foreign Debt Problem seem to be overshadowed by the indirect ones. The Cen- tral Planning Board of Poland estimates that Poland The short- and medium-term macroeconomic policy stands to lose US$2.5 billion from Iraq's invasion of Ku- makes price stabilization and achievement of an export wait, with $1.7 billion of that consisting of broken con- surplus priorities at the expense of the rate of growth tracts or the elimination of likely contracts. It also and employment. The main objective of the so-called includes $170 million worth of assets that Polish civil structural reform and systemic transformation program engineering contractors and private citizens left behind is to establish a market economy. It is assumed that price in Iraq and Kuwait and $500 million in non-payments of stabilization and market forces in particular will ensure Iraq's debt to Poland. On top of these losses comes the greater efficiency and competitiveness and enhance eco- new oil price hike: it is estimated that in 1991 alone the nomic growth. The role of the state will be limited to additional bill for oil will cost Poland over $2.5 billion. macroeconomic management and support for the undis- The inflationary effects of this external development will torted functioning of market forces. be dramatic, irrespective of whether the prices of crude The question is whether the macroeconomic policies slide back to their pre-Gulf crisis level. and market forces alone are capable of solving such real The switch to dollars and world prices in Poland's structural problems as the significant foreign debt, the trade with the Soviet Union in 1991 will cost an addition- inefficiency of mining and heavy industries, and the al US$1.5-2.5 billion, depending on the level of world oil technological gap. The foreign debt problem in particu- prices. The unification of Germany will result in a con- lar has to be taken seriously. There are two options. One traction in exports of 35-40 percent in terms of Poland's is to assume a liberal market economy system and pres- trade with the former German Democratic Republic. sure for substantial debt forgiveness. The other is to as- These external shocks will not allow for less restric- sume a market economy with active, positive industrial tive macroeconomic policies. Therefore, it is highly and trade policies aimed at securing the expansion of ex- doubtful that the liberal systemic reforms alone, accom- ports necessary to service and repay the principal of the panied by restrictive fiscal, monetary and income poli- debt. cies, can overcome the deep structural misallocation of The optimists assume that the Western democracies resources and dramatic switch to new, more demanding will support the extraordinary effort to introduce a liber- markets. Despite the exceptional adjustment efforts, al economy by a substantial debt reduction on the part of without substantial debt relief, some net inflow of for- official and commercial creditors. They also assume that eign capital and an active structural adjustment policy, the macroeconomic policy and systemic reforms will Poland may enter a long and painful recession, accompa- create the foundation for a viable economy that will en- nied by political turmoil and stalled negotiations, on the able a gradual acceleration in productivity, growth and debt problem. Poland's disequilibrium is a deep-seated 213 structural one that requires not only a contraction-and may undermine the development of a viable export sec- switching across sectors-in demand and systemic re- tor. With the progress on price stabilization and systemic forms, but also positive economic policies that will help transformations, a structural policy is needed that ad- overcome the weakness of the supply side, the low pro- dresses the real structural problems and that strength- ductivity, the technological, organizational and manage- ens export capacity and competitiveness based on rial backwardness, the misallocation of resources, and improved productivity, quality and elasticity of export the distorted industrial organizational structure (domi- supply. nated by relatively large state enterprises and underde- In order simultaneously to achieve success in short- veloped medium and small ones that are necessary for a term stabilization and lay the basis for long-term eco- competitive environment and greater economic flexibil- nomic growth, a set of policies has to be implemented ity). harmoniously. This set needs to involve macroeconomic policies, systemic reforms and active, positive economic Conclusion policies tuned to the mobilization of savings, release of hidden layers of energy and innovation and the efficient The conclusion is that Poland will not be able to over- allocation of resources along the lines of dynamic com- come the economic stagnation and to develop a signifi- parative advantage. cant and efficient competitive export base unless it The main message of this paper is that those coun- implements a long-term policy that addresses the major tries in Central and Eastern Europe that are able to structural problems comprehensively, unless the gov- achieve macroeconomic balance, to promote efficiency, ernment is actively involved in its implementing that to protect the growth of efficient industries during the policy and unless it receives simultaneous substantial difficult time of adjustment and to encourage export ex- debt relief accompanied by a net inflow of foreign sav- pansion based on a set of macroeconomic and structur- ings. al adjustment policies have the highest chance of The elimination of the hyper-inflation and initiation succeeding, as long as their efforts are seriously assisted of the systemic reforms are indisputable achievements of by foreign creditors and supported by their own societ- the Balcerowicz adjustment program. However, to be ies. sound, the stabilization needs to achieve greater eco- nomic efficiency.' The increase in the export surplus as a Notes result of the recession and the switching effect of the de- valuation in the real exchange rate of the zloty are tran- 1. A real adjustment policy corrects the disequilibrium, while a sitory. In fact, the decline in output and lack of progress quasi-adjustment program only represses it. in improving economic efficiency and technology are clear evidence of the weakness and transitory character References of the export surplus achieved in 1990. It is true that re- strictive income, fiscal and monetary policies, coupled Antowska, J., and Z. Malecki. 'The Brady Plan.' Institute of Finance, with a real devaluation of the zloty, resulted in a sharp arsaw. decline in imports and relatively higher profitability of Bhagwati, J. "Outward Orientation. Trade Issues." In V. Corbo, M. Gold- exports. However, gradual progress with the price stabi- stein and M. Khan, eds., Growth-Oriented Adjustment Programs. lization will increase aggregate demand and import de- Washington, D.C.: International Monetary Fund and World Bank. mand and absorb some exportables. The export surplus will decline and eventually may turn into a deficit. The Bienkowski, W., and J. Bossak. 1990. "Adjustment and Competitive- macroeconomic price stabilization measures and real ness." A paper presented at the IREX-PAN Conference. exchange rate devaluation have increased the competi- Bossak, J. 1984. "Socio-economic Factors Behind Competitive tiveness of exports at the expense of the absorption of a Strength of Japanese Economy." (CSPS), Warsaw. higher share of domestic savings and of a declining na- tional product. Therefore, the export surplus is not only .1990. Japonia. Strategia rozwoju w punkcie zwrothnym. a product of the recession, but it is also undercutting the Warsaw: PWN. stabilization efforts. The development of exports should play a more posi- Khan, M. 1990. "The Macroeconomic Effects of Fund-Supported Ad- tive role, especially in the structural adjustment and justment Programs." IMF Staff Papers. International Monetary economic recovery. At a time when the fixed nominal Fund, Washington, D.C. June. rate of exchange plays the role of nominal anchor, exces- Nasilowski, M. In Polityka Finansowa, Nier6wnowaga i Stabilizacja. sively restrictive income, fiscal and monetary policies Warsaw: Instytut Finans6w. 214 Sachs, J. "Trade and Exchange Rate Policies in Growth-Oriented Ad- United Nations Development Programme and World Bank. "Poland, justment Programs." In V. Corbo, M. Goldstein and M. Khan, eds., Policies for Trade Promotion." Growth-Oriented Adjustment Programs. Washington, D.C.: Inter- national Monetary Fund and World Bank. World Bank. 1989. World Development Report 1989. Washington, D.C.: World Bank. 215 18 Comments on "Adjustment, Trade Reform and Competitiveness- The Polish Experience" by Jan W Bossak Gdbor Oblath Although the notes that follow go beyond Jan W. which require the withdrawal of government from sever- Bossak's paper, most of the points raised are closely re- al activities-is necessary to revitalize these economies. lated to the problems he discusses. What the government should do is assist in the creation The major message of the paper-that to restore via- of the markets and institutions that help markets to ble economies in Central and Eastern Europe, policies function rather than continuing with the same public designed to achieve stabilization and external adjust- activities and trying to pick the winners in international ment in the very short run are inadequate-is very apro- competition. The inevitable change in the orientation of pos. The reason is that the systemic transformation trade policy, i.e., the opening up of these economies to taking place in several Central and Eastern European the world market, involves larger adjustment costs, in countries, which has few, if any, historical precedents, relative terms, than has been the case in less inward- involves fundamental changes in the basic institutions looking countries. Although, because of previous or and workings of both the economy and society. This kind present price controls, comparisons of this sort (i.e., of transformation is different from changing the orien- comparisons related to effective protection) are difficult tation of a country's trade regime from inward-looking to make, several signs indicate that the economies of to outward-looking or from protectionist to liberal or Central and Eastern Europe have been far more closed in from import-substituting to export-promoting. (As not- their trade policy than most developing countries are or ed by Bossak, these terms, although often used inter- have been. changeably, may describe different aspects of a foreign An important reason for, and an indication of, this trade system.) Instead, the challenge facing the Central stronger inward-orientation relates to the extremely im- and Eastern European countries wishing to replace their portant role of the Soviet (and Council for Mutual Eco- inherited socialist systems with market economies en- nomic Assistance [CMEAI) markets for the manufactures tails a far more difficult undertaking than the above- of the smaller Central and Eastem European countries. mentioned switches in trade orientation entail. These markets have been virtually separate from the The above statement does not imply that internation- world market, involving completely different prices and al experience related to policies aimed at changing the standards than those that apply to world trade. A large trade orientation of various countries is irrelevant for part of Central and Eastem Europe's manufacturing in- Central and Eastern Europe. To the contrary. It does im- dustry has been designed for this (presently shrinking) ply, however, that due consideration must be given to market. The major part of the goods produced by indus- the underlying differences in basic institutions. The di- tries oriented toward the CMEA market are "home goods" rect applicability to Central and Eastern Europe of in the trade policy sense-they cannot be switched to Southeast Asia's positive experience with "structural world markets, at least not without a large extra cost (in policies" and the kind of government involvement that the form of significantly lower prices). seems to have worked there give rise to profound skepti- This orientation of trade toward the CMEA market is cism; implementing similar policies in Central and East- the origin of the most serious trade problem the small ern Europe is not feasible. One of the problems Central Central and Eastern European countries face nowadays. and Eastern Europe faces is the governments' deep in- Beginning in January 1991, the CMEA will switch over to volvement in the day-to-day life of the economies of the convertible currency payments among the Soviet Union region. Substantial deregulation and liberalization- and other Central and Eastern European partners. It is 217 not at all clear how serious the short-term costs of this At this point the issue raised by Bossak concerning change will be. Even before the mid-1990 Gulf crisis and short- and long-term adjustment becomes relevant. increase in oil prices, it was estimated that the costs in With the details passed over here, there is a danger that Hungary would be between US$1 and $1.5 billion. Even devaluation in itself might reinforce the traditional and the approximate magnitude of the short-term costs gives inefficient structure of exports. Investment is what is re- an idea of the extent and implications of the protection ally needed, both domestic and foreign, to improve these offered by the CMEA market. countries' non-price competitiveness, not just the sell- Evidence of the special but strong inward orientation ing of low quality products at ever-decreasing prices. If and protected nature of Central and Eastern Europe's structural policies are understood as measures involving economies is the very large difference between purchas- strong incentives for, and promotion of, export-oriented ing power parities (PPPs) on the one hand and market or investments, then the application of structural policies commercial exchange rates on the other. The extent of is fundamentally appropriate, but only in this sense. this difference does not fit into the Balassa explanation For reasons mentioned above-namely, the inflation- based on relative income levels and relative prices of ary repercussions of large devaluations and doubts about traded goods in terms of non-traded goods. It can, as in- a swift and healthy supply response-the best course terpreted here, be explained only by the fact that the would be for Hungary to continue with a gradual, in- quality of tradables (including services related to trad- stead of a drastic and quick, trade liberalization policy, ables) in Central and Eastern Europe is much lower than accompanied by a transitory increase in the tariff for lib- it is in the West. This disparity indicates that tradables eralized products. This tariff increase would be less infla- were strongly protected. When exporting to the West, it tionary than changes in the exchange rate leading to turns out that because of differences in quality, tradables similar effects, and it could be removed gradually. fetch a lower foreign trade price than that implied by Bossak's concern over the debt problem of Poland is PPP comparisons. This fact is reflected in the commer- very real; other countries in transforming Central and cial (official) exchange rate in both Poland and Hungary. Eastern Europe face a similar problem. This concern is This point leads to the issue of exchange rate policy in closely related to the points discussed at the beginning adjustment strategy. While strongly opposing the popular of these comments-that an exceptionally comprehen- ideas often raised by past and present Central and Eastern sive transformation has to take place that alters (and not European policy-makers, according to whom the ex- simply reforms) the economic systems of Central and change rate should be the instrument of anti-inflationary Eastern Europe. If the West is interested in the success policy, this discussant has serious doubts about the ratio- of these changes, it should assist the transformation by nality and efficiency of large devaluations to promote ex- lifting the pressure to achieve day-to-day, i.e., very short- ports in the Central and Eastern European context. term, continuous improvements in the trade balances Consider Hungary's experience and problems, and not the with the heavily indebted, transforming Central and issue of seeking the "nominal anchor" for stabilization Eastern European countries. These countries trans- purposes. In Hungary, large devaluations may trigger the ferred very large sums of net resources in the 1980s, acceleration of inflation because of the low efficiency of while their debt grew continually. For example, Hunga- monetary policy. The problem of arrears (i.e., non-pay- ry's externa; debt increased more than twofold in the ments or inter-enterprise forced credits) substantially di- 1980s, while a net resource transfer abroad of almost minishes the efficiency of monetary policy, especially in US$3 billion (roughly equal to the net inflow of resourc- periods of monetary tightening. A devaluation cannot, es in the 1970s) took place. The conditions imposed by however, work without efficient monetary control. There- the International Monetary Fund call for significant in- fore, without establishing a healthy monetary system, creases of this transfer in the coming years-the period large devaluations might, instead of having real effects, that seems most critical to the transformation. The basic contribute to a significant acceleration of inflation. difficulty is that continuing a short-run trade surplus The other problem is related to supply response proper necessitates heavy restrictions on the domestic econo- and the kind of supply response required for adjustment. my, which in turn lead to a decline in gross domestic Clearly, the precondition for adjustment is a durable product (GDP). Thus, the burden of the debt service in- change in the real exchange rate: that change depends on creases. A vicious circle might evolve in which, because monetary policy, as indicated. However, even if a durable of the ever-increasing restrictions, very little adjustment decline in the real exchange rate can be achieved, the takes place, and the necessary surplus in the trade bal- question of how swift and what kind of export increase can ance is achieved at continually growing costs in terms of be expected remains. a falling GDP. 218 These dangers are extremely serious. To be sure, designed, and it is clear the debt overhang is constrain- some kind of unconditional debt forgiveness is not rec- ing their implementation, it would be in the interest of ommended. Debt relief, in itself, cannot solve the prob- the West as well as a generous act to lighten the burden lems of Central and Eastern Europe. However, if serious, of the debt service. This point is the other very important well-designed and promising programs for revitalizing message in Bossak's paper. He is completely correct in any of the Central and Eastern European economies are this conclusion. 219 19 Fiscal Issues in Economies in Transition Vito Tanzi l Restructuring and the Institutional moval of food subsidies. Workers and managers will Requirements oppose the privatization of their enterprises for fear of losing their jobs or power. Those who have benefited Recently, the countries of Eastern and Central Europe from free, or almost free, housing will oppose privatiza- have abandoned the economic and political ideology that tion of the housing stock. In short, existing institutions determined their economic policies and institutions over have constituencies and lobbies that, although they may the past four decades (and longer for the Soviet Union). welcome the change to a market economy in the ab- They have started a process of reform and transformation stract in the belief that it will generate Western stan- aimed at replacing central planning with a largely market- dards of living, will oppose the immediate negative costs based economic system similar to that common in the of the changes. During the transition, the costs of re- West. The strongest forces pushing for this transforma- structuring will be current and visible while the benefits tion are the widely recognized failures of the Marxist- will be future and uncertain. Doubts will inevitably arise Stalinist ideology and the growing awareness of the large as to the wisdom and the speed of the change. These differences in the standards of living between centrally doubts will pose additional obstacles to the introduction planned and market-oriented European countries. These of the required policies. If the resistance of the opposi- differences cannot be justified on the basis of cultural and tion is strong-and recent events imply it will be-exist- historical factors or differences in resource endowment. ing institutions, or at least some of them, may hang on Half a century ago, for example, the populations of Czech- for some time. This situation will increase the cost of the oslovakia and Hungary had incomes similar to or perhaps transition, raise uncertainty and delay realization of the even higher than those of Austria. Today, the Austrians full benefits of the economic transformation. have much higher standards of living. The Soviet Union Second, the creation of new institutions requires has enormous natural resources compared with the West time, some more than others. At some point, when many European countries but a much lower per capita income. of the old institutions have been dismantled or at least The transformation to a market economy will require rendered ineffective and the new ones have not yet been the dismantling of many institutions and the establish- fully installed, a kind of institutional vacuum will arise. ment of new and often very different ones. If the trans- Both economically and politically this period will pose formation could be instantaneous, it would be easy. Once the greatest difficulties. the new institutions were in place, the countries could A third and related point is that the establishment of channel their energies toward reducing, if not eliminat- each institution will have its own dynamics. That is, the ing, the income gap that separates them from the West. time needed to create an institution, say, a new tax sys- Unfortunately, the transformation is likely to be difficult tem, will depend on technical requirements more than and time-consuming for several reasons. First, while the on the need to have that institution in place. The realis- populations of these countries may favor the idea of a tic sequencing is likely to be different from the ideal one. market-oriented economy, abandonment of existing in- The staggered introduction of institutional changes will stitutions will have, especially during the transition, ma- create uncertainty, since at each step the possibility that jor impacts on particular groups, who are likely to the next step may not be taken will influence decisions oppose changes that will affect them negatively. For ex- by individuals and enterprises. ample, pensioners and many others will oppose the re- 221 Fourth, and an important point, is the interdepen- the right legal, accounting, financial, managerial and dence of institutions and skills. Modern economies are other skills required by Western-style market economies somewhat like ecological systems: their various compo- are not available. The implication is that the institutions nents-institutions, industries, professional skills, to be created, especially during the transition, must be etc.-support and feed each other. Each industry or in- simple so as not to make excessive demands for particu- stitution depends on the existence of other industries lar skills. Some observers have argued that the high ed- and institutions. Take, for example, the tax system or the ucational levels of the populations of Central and financial system. To function effectively in a modern Eastern Europe will facilitate the transition to a market market economy, they need well-developed legal frame- economy. Unfortunately, this assumption is only partly works as well as particular accounting, legal and mana- true: a high level of education does not guarantee that gerial skills. They will require that information be the required skills are available or can be easily acquired. processed and presented in certain ways. Economic rela- The skills that are not available and that cannot be ac- tions in a market economy are based on legal, account- quired within the countries given the absence of teach- ing and financial frameworks, such as contract law, ers are exactly the ones that the newly transformed corporate law, bankruptcy law, financial laws and regu- market economies will need the most. lations, and tax laws and regulations, that have devel- oped over many years, in some cases over decades and Restructuring and the Fiscal Institutions centuries. Some contracts and obligations are implicit, but they are still generally observed in a modern market General Background economy because society has come to accept them. Over the years, the advanced Western economies have The Eastern and Central European countries have achieved some sort of "ecological balance" among their entered the transition period with levels of government institutions and available skills. The demand for certain expenditures ranging from around 50 percent to over 70 institutions and skills has led to their development, at percent of their gross domestic product (GDP) or in the times after periods of difficulties and the commission of case of Czechoslovakia, of their net material product (ta- errors. A balance between required and available skills ble 19.1). These are extremely high levels, especially in and institutions is unlikely to exist or to develop quickly light of the relatively low per capita incomes in these in a country that in a short period has transformed itself countries. These high expenditures are the result of sub- from a centrally planned to a market economy. For ex- sidies for families and enterprises and of government fi- ample, accounting skills will be lacking, and it will nancing of the preponderance of public enterprise therefore be difficult to prepare balance sheets and in- investments. The privatization of public enterprises and come and loss statements for newly privatized enterpris- the removal, or at least the reduction, of subsidies would es. This situation in turn will hinder the introduction reduce these public expenditures. However, other fac- and enforcement of a modern corporate income tax. tors will push in the opposite direction, especially during Privatization will be hampered by the lack of Western- the transition. style managers with business school backgrounds. The transition and restructuring will inevitably in- The lack of these skills will prevent the smooth and crease unemployment significantly, as some public enter- quick establishment and functioning of Western-style prises are closed or are forced to restructure their institutions. It would be naive to assume that Western production toward goods that are sufficiently competitive institutions can be copied and transplanted as long as and of good enough quality to be sold in a market econo- Table 19.1. General Government Expenditures of Central and Eastern European Economies (percent of GDP) 1982 1983 1984 1985 1986 1987 1988 1989 Bulgaria 54.7 53.7 51.9 55.2 61.4 64.7 63.7 61.9 Czechoslovakiaa 67.0 67.4 67.4 69.8 69.8 71.2 74.1 76.6 Hungary 61.2 61.9 59.4 61.1 64.9 61.6 63.0 59.0 Poland 52.2 47.6 48.4 48.2 49.7 47.7 48.0 48.1 Romania n.a. n.a. n.a. n.a. n.a. n.a. n.a. n.a. USSR 49.3 48.6 48.9 49.8 52.2 52.3 53.3 51.4 n.a. Not available a. Percent of net material product. Source: Various national sources. 222 my and in an open environment. If the economies are dy- nally, substantial sums will have to be spent to create the namic and if political or social obstacles do not prevent new institutions needed in a market economy, such as labor mobility, the period of unemployment may not be central banks, commercial banks, tax administrations, too long for many workers. In particular, private sector budget offices, new faculties in the universities and so activities and especially the expansion of the labor-inten- forth. Some funds may be gained from the dismantling sive service sector, which is now very undeveloped, could of the planning and other offices (internal security) re- generate substantial demand for labor. Still, the costs of quired by a planned economy but not by a market econ- adjustment in terms of unemployment could be large, as omy. Nevertheless, at least in the short run the pressure the recent experience of Poland, the former German Dem- for greater public spending will be substantial. ocratic Republic and other countries shows. Regardless of The bottom line, then, is that for several years the how dynamic these economies prove to be, unemploy- countries of Eastern and Central Europe will continue to ment will rise. As a consequence, the government will require levels of public expenditure far higher than what need to play a role in providing unemployment compen- would be expected from international comparisons, giv- sation and retraining those who lose their jobs. Such mea- en their per capita incomes. These high levels will make sures will have an impact on public spending. it essential to scrutinize closely the allocation of the ex- Removal of the subsidies for food, lodging, energy, penditures and the efficiency with which they are used. transportation, cultural and recreational activities, pub- They will also require that government revenue remain lic utilities and so forth will reduce public spending. high. These requirements have implications for budget- However, it will also sharply reduce the real incomes of ing and for tax policy and administration. vulnerable groups such as pensioners, invalids and chil- dren. Thus, there will be a need to adjust pensions and Budgeting other entitlements. Once again, there are spending im- plications. Within this same area it is important to men- In Western countries the budget office is legally re- tion that the public enterprises have been responsible for sponsible for preparing the budget, for making basic de- important social security and welfare functions. Implicit cisions as to the level of total spending and its allocation in the privatization of these enterprises, or even the among the various ministries and functions, and for en- pressure to make them more competitive, is that the suring that the objectives sought by the budget are being state will need to assume some of these functions. achieved at the lowest cost. In these countries, while the In part because of the importance of consumer subsi- budget office determines total spending on the basis of dies, money wages have been very low in most of these macroeconomic considerations, its allocation is at least countries. Removal of the subsidies will create strong influenced by requests from the spending units. pressures to increase money wages. Workers in these In centrally planned economies, the budget office has countries will make increasing use of a weapon widely been far less important; its role has been subordinated to used by Western workers-the right to strike. Should that of the planning office. It has had little if any control government workers get an increase in money wages, over productive expenditures, i.e., investment spending, public spending will go up. which has been determined by the plan, and it has had lit- The need to spend more for some services such as ed- tle control over important extra-budgetary accounts for ucation and health will also create pressure for greater military spending, national security and the like. The ba- public expenditures. In some of these countries the stu- sic decisions on total spending and to a certain extent on dent/teacher ratio is far higher than in the West, and the resource allocation have been made by the top levels of quality of health services is very low. In many of these the bureaucracy, in long-term plans, with little or no in- countries, low levels of spending for operations and fluence by the spending agencies. Thus, the yearly bud- maintenance have left the social infrastructure in very gets have been little more than instruments to promote poor condition. Restoration of this infrastructure to a and implement the objectives of the plan rather than reasonable state as well as the provision of services (such those of the spending agencies. Another problem with this as telephones) that are not now available will create ad- budgetary system relates to the classifications used. No ditional pressure for public spending. distinction has been made between ordinary revenues and It is unlikely that all enterprises will be privatized. It borrowed funds. Thus, excessive borrowing has showed is improbable that the public enterprises will become ef- up as a budgetary surplus. On the other hand, too sharp a ficient enough in a relatively short time to turn a profit, distinction has been made between the productive and that, given the changing economic situation. Thus, they non-productive sectors (all others). Considering how inef- will continue to need subsidies. Some subsidies may be ficiently capital has been used, this terminology has made directed toward enterprises that are restructuring. Fi- a mockery of the concept of efficiency. In addition, little 223 attention has been paid to cash management; no cost has planned economies, they can be thought of either as tax- been imputed for maintaining idle cash. Finally, there of- es charged on the sales made by state enterprises to re- ten has been no specific budget law, or it has been largely tail shops or simply as the difference between the prices ignored. charged to consumers (minus a small profit margin for The shift toward a market-oriented economic system the retailers) and the prices received by producers. Turn- has made clear that these budgetary procedures and over taxes are not charged on sales between producers. practices must be replaced by a more decentralized, Since both the producer prices and the retail prices are transparent and efficiency-oriented system. Under this fixed by the department of prices and thus are known, new system, the level of spending for the budget year these taxes acquire qualities similar to those of specific must be more closely related than in the past to projec- taxes.3 That is, the tax revenue can be calculated once the tions of the behavior of revenue and the economy in gen- volume of sales is known. In cases where, for social or eral over the budget year. The allocation of total planning reasons, the government has wanted to subsi- spending must be based more on the submissions of the dize the consumption of particular products, these taxes spending agencies. Labels such as productive and non- have become negative and thus amount to subsidies. productive must not determine the priority of spending Turnover taxes have at times been applied with hun- categories. The budget office must have staff trained to dreds or even thousands of rates. This situation would be assess the merit of requests by the spending agencies, it a nightmare for Western-style tax administrations, must have a greater say in determining the size of total which have difficulty coping with a few rates for value- spending, and it must be able to provide more useful added taxes (VATs). Nevertheless, governments in Cen- classifications of spending categories. On the basis of tral and Eastern Europe have collected from the turn- this information it should be possible to classify spend- over taxes revenue far greater than that Western ing according to economic or functional category. Cash countries have collected with VATs.4 management must also be improved, with recognition of The method of collection has been very simple and the opportunity cost of keeping cash idle. highly effective. All payments between enterprises and In sum, the budget office must have skills and man- between enterprises and retailers have been made power it does not now have. Only if the budgetary func- through credit transfers within state banks. Cash pay- tion is greatly improved will the budget office be able to ments have not been allowed, and there have been no play a role similar to that of the budget offices in indus- other forms of payment. Since the prices were adminis- trial countries. At the same time, an effective budget of- tratively fixed and known to the banks, they could calcu- fice cannot be developed overnight. Apart from the cost, late the tax liability on the basis of the volume of sales. it requires individuals with skills not now available, as Thus, the taxes were transfers from the accounts of the well as well-thought-out legislation. It also requires that retailers to the government's account. In other words, political obstacles to the reorganization of that office be the banks would reduce the balances of the retailers by removed. This reorganization inevitably implies a loss of the required amount and credit part of that amount to political power for many groups that used to have great- the enterprise and part to the state as a tax payment. er discretion over on the use of public resources. These This simplified description highlights the great de- groups will attempt to prevent budgetary reform. pendence of the turnover tax on two elements-price controls and the method of payment. Moreover, with Tax Policy and Administration this system there is no need for a Western-style tax ad- ministration. The taxes are simply transfers of a part of The transformation of the economies of Eastern and the gross revenues of enterprises before they receive Central Europe from command to a market orientation those revenues. This system could not survive the liber- has profound implications for their tax systems. Tradi- alization of prices and methods of payment. When prices tionally these countries have collected substantial reve- are no longer fixed and are thus no longer objectively nue from taxes, particularly three of them-turnover, known to the banks, and when payments can be made profits and payroll. Although the specifics of the applica- through different channels and by different means (cash, tion of these taxes differ from country to country, the ba- checks, credit and so forth), it is no longer possible to sic principles are broadly similar. The following rely on this system. discussion focuses on those principles to show the po- Hungary, Czechoslovakia, Poland and Romania need tential effect of the reforms on revenue. The differences to reform their turnover taxes as a consequence of the among the countries are ignored.2 price liberalization and financial reforms. Hungary has Turnover Taxes. Given the traditional way in which led the way in freeing most prices and in reforming the turnover taxes have been applied in the centrally tax system. In 1988, it introduced a VAT with two rates- 224 15 percent and 25 percent-that replaced various turn- Price liberalization and privatization will require over taxes. These rates are very high but are necessitated profound changes in the way profits are taxed. They will by the decision to exempt large sectors from the tax and also require clearly determined ex ante rules. They will by the need to maintain high levels of revenue. Prepara- further require that profits be better measured to ap- tion for the introduction of the VAT took more than two proximate the economic concept.7 These skills required years during which time new administrative mecha- to achieve these determinations (accounting and mana- nisms were set up. The change was facilitated by the fact gerial skills) are now lacking and will take some time to that, as in most centrally planned economies, much of develop. Once the switch to a Western-type taxation of the production takes place in a relatively small number profits is made, the role of the tax administration will of large enterprises. With liberalization and privatiza- also change, since it will be responsible for determining tion, and with the growth of small enterprises and ser- the accuracy of the tax declarations and will need to have vices, this situation will undoubtedly change, and the ability to challenge taxpayers when their declara- Hungary will face greater problems maintaining current tions are suspect. rates. During the transition the revenue from this tax is Czechoslovakia is planning to replace its many turn- likely to fall and will remain much lower even after the over taxes with just four, which would have rates of 0 transition is completed. It should be recalled that the percent, 12 percent, 22 percent, and 32 percent, as an in- revenue from taxes on profits in the West is only a small termediate step. The new rates would be levied on the ba- proportion of the revenue collected from this tax in the sis of wholesale prices, to take effect January 1, 1991. It centrally planned economies. eliminated the negative turnover taxes on food in July Payroll Taxes. The third major source of revenue- 1990 by raising retail prices, and it was planning to in- payroll taxes-is similar to that in the West. The taxes troduce similar changes for energy prices in December are collected directly from the enterprises as a propor- 1990. Special excise taxes will be levied on particular tion of the payroll. Since the payroll is often met by cash products (alcohol, tobacco, gasoline, coffee and passen- transfers from the banks, they can simply withhold the ger cars) to maintain the level of revenue. These changes proportion required by the tax. Once again, this system will facilitate the liberalization of prices. However, major is efficient and simple; it does not require a tax adminis- changes will need to be made to the tax administration tration to handle the collection, especially where the since the traditional system for collecting turnover taxes payroll tax rate is proportional. will no longer be feasible. Without these changes, tax Other Taxes. Some of the remaining taxes are largely revenue could fall substantially. presumptive ones, as, for example, the taxes on land and In January 1990 Poland reduced the rates of its turn- on private activities. These taxes are often negotiated ex over tax to 14 and is contemplating the introduction of a ante with the taxpayers and, once they are determined, VAT. require little if any checking on the part of the tax ad- Tax on Profits. The taxes levied on profits in centrally ministration. There has been little or no attempt to base planned economies have had little in common with them on objective criteria of income or profits. those used in the West. One difficulty is that without proper accounting it is difficult to define profit, especial- Concluding Remarks ly in the case of socialized enterprises.5 The definition of profit is distorted by the price controls, the extremely A few basic conclusions can be drawn from the brief low interest rates paid by the enterprises, the distorted discussion in the previous sections. First, the centrally rules of amortization that often assume lives of assets planned economies have been very effective in develop- and buildings much longer than is the case in the West ing revenue systems that, at low administrative (but not (and much longer than the actual economic lives of the efficiency) costs, have succeeded in raising levels of rev- assets), the valuation of inventories, and the discretion- enue comparable to those in the most advanced industri- ary nature of what expenses enterprises can claim as al countries. They have achieved this success, however, costs. In these countries, the taxes on the profits of state at high efficiency costs, since the taxes probably made a enterprises have not been based on an ex ante and objec- distorted economy even more distorted,8 and they prob- tive, even if distorted, concept of profit. Rather, the con- ably greatly reduced the incentives of the enterprises to cept of profit has often been arbitrary, such as a fixed become more efficient. This point applies particularly to proportion of the producer's turnover, and often the en- the taxation of profits. The remarkable feature of this terprise has negotiated the taxes based on that arbitrary overall system of taxation is that no real tax administra- concept so that they amount to ex post taxes.6 tion was needed, so that many of these countries have 225 never developed one in the Western sense. By Western A fourth conclusion is that the transition from a cen- standards, and with respect to the needs of market econ- trally planned to a Westem-style tax system will need to omies, the existing tax administrations appear some- include an "inter-regnum" period in which one tax system what primitive. will have been abandoned and the new one will not yet A second basic conclusion is the great fragility of this have been fully implemented. Thus, temporary sources of revenue system with respect to the changes required to revenue will need to be developed to prevent revenue from transform these countries into market economies. The falling too fast and fiscal deficits from becoming too large. transformation will largely destroy the basis for the The experience of several of these countries confirms turnover taxes and, to a lesser extent, the profit taxes. that tax revenue will fall once they start the transition. These countries must pay a lot of attention to the cre- For example, in China the ratio of revenue to gross na- ation, almost from scratch, of a new and modern tax sys- tional product fell from 34.4 percent in 1978 to 20.4 per- tem that will almost surely yield less revenue than the cent in 1988. In Hungary the ratio of revenue to GDP fell one it will replace. The implication is that the level of from almost 60 percent in 1983 to less than 50 percent public spending must be sharply reduced to prevent ma- in 1989, in Poland from 43.4 percent in 1982 to around jor macroeconomic problems. Large fiscal deficits have 30 percent in 1989, and in the Soviet Union from 47 per- already appeared in some of these countries, especially cent in 1985 to 41 percent in 1989. the Soviet Union, where tax revenue has fallen in part as In this interregnum period, excise taxes can play an a consequence of the ongoing reforms. These deficits important role in generating revenue, and simplified need to be contained and reduced if they are not to create versions of the turnover tax can help until they are re- major economic imbalances and problems (such as high placed by full-fledged VATs. Imports can also contribute rates of inflation, balance-of-payments difficulties and to the revenue needs, perhaps through an across-the- higher unemployment) that will make the populations board minimum tax imposed at the time the exchange question the wisdom of the change toward a market rates are adjusted. Western-style income taxes on indi- economy. In fact, fiscal deficits are more dangerous in viduals must be introduced, although they may not be these countries because of the difficulty of covering the immediately productive. shortfalls through sources other than inflationary fi- Finally, policy-makers in these countries must resist nancing. the temptation to see the fiscal instruments in the same A third conclusion is that the existing tax administra- light as they saw the plan, that is, as a way to replace the tions, which are much smaller and less powerful than judgment of the market with their own. Although equity those in the West, must be substantially expanded and has always been an important objective of fiscal policy, strengthened and must be given greater powers. This this objective must be pursued mainly by using the fiscal change is particularly important if, as argued by Kornai instruments to moderate the distribution of income that (1990, p. 118), "Nowadays, people in general consider it a would result from the free play of the market.9 laudable act, rather than something to be ashamed of, if Fiscal policy must not interfere with the choices of someone defrauds the state...." Staff with the skills re- the market by promoting certain activities over others. quired to administer the new taxes will need to be trained, That lesson should have been learned from the experi- perhaps with foreign assistance. New techniques of ad- ences of developing and developed countries alike. ministration will need to be developed that reflect local conditions. 226 Appendix 1. Statistical Tables Table 19-1-1. General Government Tax Revenue of Central and Eastern European Economies, 1986 (in percent of GDP) Czecho- Yugos- Bulgaria slovakia Hungary Poland Romaniaa lavia USSR OECD Profit taxes 18.7 11.1 11.0 13.5 6.4 16.0 3.0 25.1 Income taxes 4.0 0.8 3.8 6.0 8.6 3.9 12.1 Turnover taxes 16.5 15.9 16.7 11.6 9.8 6.8 11.5 11.5 Social security contributions 9.7 5.9 16.6 10.3 7.2 8.1 3.3 9.3 Customs duties 0.5 0.7 3.5 2.8 0.9 1.8 7.2 n.a. Other 0.2 1.0 3.5 4.1 n.a. 5.0 3.8 2.2 Total 49.6 48.6 52.5 43.6 37.4 36.7 45.7 38.1 n.a. Not available. a. 1988. Sources: National sources for all countries. For OECD, Revenue Statistics of OECD Member Countries (Paris, 1988). Table 19-1-2. Bulgaria: Level and Composition of Table l9-1-4. Hungary: Level and Composition of Tax Revenue Tax Revenue (in percent of GDP) (in percent of GDP) 1985 1986 1987 1988 1989/ 1984 1985 1986 1987 1988 1989 Tax revenue 46.1 49.6 46.3 47.4 49.3 Tax revenue 52.6 49.9 52.2 52.4 54.0 49. Profit tax 19.2 18.7 18.9 21.2 23.3 Profit tax 12.6 9.6 11.1 11.9 8.4 7.0 Income tax 4.0 4.0 4.0 4.0 4.1 Income tax 0.9 0.9 0.8 0.8 4.7 5.5 Turnover tax 12.6 16.5 13.3 11.6 11.2 Turnover tax 17.9 16.7 16.7 16.9 22.6 17.6 Customs duties 0.4 0.5 0.4 0.8 0.8 Customs duties 3.4 3.1 3.5 2.8 3.0 4.0 Social security Social security contributions 9.7 9.7 9.5 9.5 9.6 contributions 11.7 15.7 16.6 15.5 14.4 14.3 Other 0.2 0.2 0.2 0.3 0.3 Other 6.0 3.9 3.5 4.5 0.8 0.5 a. Preliminary or estimated. Sources: National data and staff estimates. Sources: National data. Table 19-1-5. Poland: Level and Composition of Table 19-1-3. Czechoslovakia: Level and Compo- Tax Revenuea sition of Tax Revenue (in percent of GDP) (in percent of GDP) (in_______percent___________of_____GDP)______ 1985 1986 1987 1988 1989 1985 1986 1987 1988 1989 Tax revenue 43.4 43.6 40.5 40.2 36 Tax revenue 50.7 48. 56 5 - l507 54.9 Profit tax 10.7 11.0 11.0 12.1 10.2 Profit tax Income tax 3.7 3.8 3.6 3.5 3.4 26.9 25.1 26.2 25.1 21.4 Turnover tax 12.3 11.6 10.6 10.8 8.9 Income tax Customs duties 2.8 2.8 2.4 2.0 1.1 Turnover taxa 16.3 15.9 15.6 15.8 17.6 Social security Customs duties 0.5 0.7 1.4 2.2 1.0 contributions 10.6 10.3 9.5 7.9 8.7 Social security Otherc 3.3 4.1 3.4 3.9 4.5 contributions 5.9 5.9 6.3 6.7 15.0 Other 1.0 1.0 1.0 1.0 n.a. a. State budget and social insurance funds. utner___________1.0________ 1.0_______ 1.0__________________n.a.______ b. Includes dividend payments based on enterprise capital. a. Includes intemal market differential. c. Real estate tax, excess wage tax, and tax payments by the nonsocialized sec- Sources: National data. tor and the population, charges on stock revaluation, and temporary levies. Sources: National data. 227 Table 19-1-6. Romania: Level and Composition Notes of Tax Revenuea 1. The views expressed in this paper are personal ones and may or 1988 1989 may not reflect official positions of the International Monetary Fund. The assistance received from Lazlo Garamfalvi is much appreciated. Tax revenue .37.4 42.4 2. For a good discussion of the tax systems of these countries, see Profit tax 13.5 11.1 Gray (1990). Income tax 6.0 6.8 3. However, discretionary price adjustments have been frequent. Turnover tax 9.8 14.7 Thus, effective tax rates can change considerably over time. Still, the Customs duties 0.9 1.0 new prices are known. Social security contributions 7.2 8.7 4. In Western countries it has been very difficult to achieve reve- nues from VATs greater than 10 percent of GDP. The centrally planned a. In Romania there are as yet no official estimates of gross domestic prod- economies have reached or exceeded 15 percent with the turnover tax- ucts. Unofficial ones have been used. Sources: National data. es. 5. A World Bank mission that visited nine enterprises in Poland was given eight different definitions of profit. Table 19-1-7. USSR: Level and Composition 6. Furthermore, the formal rates have been very high. 7 Additionally, the tax rates will need to be much lower and to fall of Tax Revenue into line with those that prevail in the rest of the world. (in percent of GDP) 8. Second-best arguments make this a qualified conclusion. 1985 1986 1987 1988 1989? 9. It is very likely that the distribution of income will become less equal during the transition. The governments should do their best to Tax revenue 43.4 A2.0 41.5 38.8 38.5 moderate the situation, but only in ways that minimize the damage to Profit tax 15.2 16.0 15.3 13.5 12.3 efficiency. Income tax 3.9 3.9 3.9 4.1 4.5 Turnover tax 12.6 11.5 11.4 11.5 11.9 References Customs duties 8.4 7,2 7.5 6.3 6.3 Social security Gray, Cheryl W. 1990. "'ax Systems in the Reforming Socialist Econo- contributions 3.3 3.3 3.4 3.4 3.5 mies of Europe." World Bank Working Paper, WPS 501. Washing- ton, D.C. a. Preliminary or estimated. .Sources: National data. Kornai, Janos. 1990. The Road to a Free Economy. New York: W.W. Table 19-1-8. Yugoslavia: Level and Composition Norton and Company. of Tax Revenue Ofer, Gur. 1990. "Macroeconomic Issues of Soviet Reforms." Mimeo. (in percent of GDP) March 9-10. 1985 1986 1987 1988 1989u Tax revenue 33.1 36 7 34.5 33.2 34.7 Profit tax 5.4 6.4 5.2 4.9 5.7 Income tax 7.6 8.6 8.5 7.1 9.7 Turnover tax 6.6 6.8 6.0 6.9 5.3 Customs duties 2.6 1.8 2.5 3.1 1.8 Social security contributions 6.3 8.1 8.3 7.3 8.9 Other 4.6 5.0 4.0 3.9 3.3 a. Estimates. Source: Statistical Yearbook of Yugoslavia, 1990. 228 Part VII Panel on Privatization 20 Privatization in Eastern Europe: The Case of Poland David Lipton and Jeffrey Sachs' The transformation of the Eastern European econo- intermediaries can be gauged from the fact that institu- mies into market economies requires comprehensive ac- tional investors now hold more than half the value of tion on three fronts: macroeconomic stabilization; shares in the United Kingdom, Italy and Japan, as well as liberalization of economic activity; and privatization of more than half the value of the New York Stock Ex- state-owned enterprises.2 Each of these is a monumental change.4 The economic challenge, then, is to combine task. Nonetheless, privatization stands out as the most the redefinition of property rights with the creation of difficult and novel of the three, both conceptually and vital financial market institutions. The political chal- politically. There are enormous challenges in transfer- lenge is also awesome: to design a mechanism for creat- ring state-owned property-which constitutes around ing private property rights that can win broad, lasting 90 percent of the industrial capital in Eastern Europe- social approval (and prevent special interests from para- to private hands in a manner that is rapid, equitable and lyzing the process through a fight for the spoils). fiscally sound and that accomplishes two fundamental Ironically, the rush of investment bankers and West- goals: the efficient operation of the resulting private en- em experts who have been proposing privatization strat- terprises; and the development of efficient capital mar- egies in Eastern Europe has not addressed the real needs kets. of privatization. Not surprisingly, the bankers have fo- The task of privatization in Eastern Europe is not cused almost exclusively on a /?nr-by-firm strategy not widely understood in the West, partly because mislead- unlike the programs in which they have participated in ing analogies have been made to privatization in other other parts of the world. In Eastern Europe, however, parts of the world.3 In a typical country that has recently the likelihood is that this customized approach would "privatized" some state enterprises, only a handful of bog down for political, economic and financial reasons firms-perhaps up to a few dozen-have been sold by well before a significant proportion of state firms are ac- the government to the private sector. These sales may tually privatized. have made an important economic difference in some In this paper, the authors review the enormous scope sectors, but they have not involved a fundamental trans- of the privatization task, and suggest means for a rapid, formation of the economy. The amount of capital trans- efficient and equitable transformation of state property formed through privatization has generally been a small into private property. The focus will be Poland, where proportion of total business capital and national income, the authors serve as economic advisors. Some of the and the economies typically had large, private industrial problems discussed are more urgent in Poland than else- and financial sectors before the privatization. where. Particularly problematic, for Poland, is the fact In Eastern Europe, privatization is a very different that the workers' councils are powerfully organized in task, involving nothing less than the complete redefini- many enterprises and are fighting for worker self-man- tion of property rights for literally thousands of enter- agement and against privatization. Nonetheless, most of prises. Privatization means the creation anew of the the analysis of privatization in Poland applies through- basic institutions of a market financial system, includ- out Eastern Europe. ing: corporate governance of managers; equity owner- The authors offer one disclaimer at the outset. Even ship; stock exchanges; and a variety of financial though rapid privatization is favored, it is doubtful that intermediaries, including pension funds, mutual funds privatization will produce immediate, large increases in and investment trusts. The importance of such financial productivity or managerial efficiency. The real gains 231 from private ownership will take years to manifest them- tion of shares to the population. Other important and in- selves-the length of time needed for managers to be fluential pieces are those by Hinds (1990) and Frydman upgraded, supervisory boards to gain experience, stock and Rapaczynski (1990). Other researchers advocating markets to improve the quality of valuation of enterpris- rapid privatization through a free distribution of shares es, and real economic restructuring to take place. None- include Olivier Blanchard and Richard Layard, Fernando theless, the authors believe that to enjoy those Saldanha and Branko Milanovic and Anders Aslund. The enormous long-term gains, it is necessary to proceed president of the Hungarian-American Enterprise Fund, rapidly and comprehensively on creating a privately Alexander C. Tomlinson, has also proposed rapid privati- owned, corporate-based economy in Eastern Europe. zation through the free distribution of shares (Tomlin- Since privatization is such a vast topic and cannot be son 1990). The Economist (July 21, 1990), too, has treated comprehensively in a single paper, it is impor- advocated rapid privatization in this manner. tant to spell out what is not discussed here. The prob- For other analysts, privatization should take place at lems of "small-scale" privatization of retail shops and a more measured pace to ensure the development of ef- smaller industrial enterprises are not addressed.5 The so- fective enterprise ownership. In the view of Kornai cially complex question of restitution of the property, or (1990, p. 82), the apparatus of the state: reprivatization as it is called in Eastern Europe, that was nationalized in the 1940s through 1960s is not touched is obliged to handle the wealth it was entrusted with on. The ground-up development of the private sector in carefully until a new owner appears who can guarantee Eastem Europe, which is at least as important, and in a safer and more efficient guardianship. The point is not Kornai's view, even more important, than the privatiza- to hand out the property, but rather to place it in the tion of state enterprises is not discussed.6 Finally, rela- hands of a really better owner. tively little is said about the crucial question of managerial compensation, which can provide a vital link Komai believes that to bring about effective between the interests of managers and the interests of ownership, the enterprises should be sold carefully and the owners of the enterprises. mostly on a one-by-one basis, rather than freely distributed to the public. The Debate over the Pace of Privatization The authors are sympathetic to the concems that Kornai raises. As he stresses, the privatization strategy Perhaps the central debate about privatization in should focus on establishing effective ownership and Eastern Europe concerns the feasible pace of an effective corporate governance, rather than on simply transferring privatization strategy. Some authors believe that effi- nominal ownership to the private sector. The secondary ciency gains from private ownership and private capital market for corporate control, which might operate, for markets are so overwhelming that the process must be example, through takeover bids on a stock market, will speeded up as much as possible. Advocates of rapid priva- not be reliable enough to ensure efficient matches tization are typically confident that even if quick privati- between enterprises and owners. Nonetheless, the zation initially leads to an inappropriate distribution of authors believe that privatization can proceed faster than ownership with, for example, too diffuse ownership, or a one-by-one sale of enterprises would dictate. The firms in the wrong hands, then the capital markets will process can be accelerated, through the free distribution encourage a reshuffling of ownership through takeovers, of shares in a manner consistent with the development of mergers and buy-outs so that there is a proper matching effective ownership. of owners and firms. For these analysts, privatization One reason that rapid privatization is stressed here is should be undertaken as rapidly as possible to reap the that the current pattem of ownership in Eastem Europe full benefits of private ownership. is itself prone to massive inefficiencies. The potential costs Many authors who take this point of view have re- of overly rapid privatization must be traded off with the cently called for a free distribution of the enterprises high costs of maintaining the present system in which into private hands as a way to speed privatization.7 In Po- state-owned enterprises lack clear incentives (or actually land, the landmark study calling for a free distribution of have perverse incentives) in the face of the market forces shares to speed privatization by Lewandowski and now being introduced in Eastem Europe. However, the Szomberg (1989). In Czechoslovakia, the Minister of Fi- authors' reasons go further than that, to the politics of nance, Vaclav Klaus, and his deputy, Dusan Triska, have privatization. In their opinion, the real risk in Eastem been outspoken advocates of rapid privatization through Europe is not that the privatization will be less than the sale of enterprise shares for vouchers. In Hungary, optimal, but that it will be paralyzed entirely. Unless Liska has been an outspoken advocate of a free distribu- hundreds of large firms in each country are quickly 232 brought into the privatization, the political battle over ful of privatizations, Hungary has entered into a heated privatization will soon lead to a stalemate in the entire internal debate over political accusations that the priva- process, with the devastating long-term result that little tization process is selling Hungarian assets too cheaply privatization takes place at all. to the rest of the world. The first director of the Hungar- Consider the current political situation. While the ian State Property Agency was fired in part because of state enterprises are presumably owned by the state, the the controversial flotation of shares in the tourist agency various components of ownership-including the rights Ibusz on the Austrian stock exchange. In Poland, an in- to use the property, to benefit from financial returns, to creasingly powerful coalition of interests that favor benefit from financial returns and to dispose of the prop- worker management is already organizing itself against erty-are in fact jointly in a shifting and imprecise way, widespread privatization. among managers, workers and the state. In this situa- The authors propose a series of detailed steps to ad- tion, workers and managers have incentives to wrest the dress the various pitfalls of privatization. At the outset, it income and assets of the firm away from the state, often is recommended that hundreds of the largest enterprises in ways that are highly inefficient and politically explo- be converted into Treasury-owned joint-stock companies sive. to ensure that the enterprise sector is indeed transformed If property rights are not clearly defined in the near into corporate form. Right from the start, some firms future, enormous energy will likely be wasted in a bruis- could be managed on a case-by-case basis (for instance, ing fight over property rights. As in Argentina, for exam- where a private bidder comes forward). Most firms, how- ple, the nation's political and social energy will be spent ever, would follow a special track emphasizing the rapid defining the rules of the game, rather than operating ef- distribution of shares. A portion of shares would be given ficiently within stable rules. Groups of workers and man- at a low price or at no charge to workers, and another por- agers, both at the enterprise level and at the political tion of shares would be transferred rapidly and free of level, will surely try to strengthen their claims (both le- gal and implicit) over the enterprises in ways that will charge to various financial intermediaries (such as mutu- further confuse the ownership structure. Their efforts al funds, pension funds and commercial banks). Shares in simply block the state's political and legal capacity to these intermediaries will in turn be distributed or sold to - ~~~~~~~~~~~~~households. Finally, the government would retain a por- privatize the firms at a later date. Insiders such as influ- tonshares Fieach tee and would radua sell ential bureaucrats and state managers might well use tion of shares of each enterpnse and would gradually sell their positions to grab an inordinate share of the state them off as a block to "core investors" who are to take a property. Such actions could undermine the public con- key role in management of the enterprise.8 In this way, it sensus to proceed with privatization. is hoped that rapid privatization through free distribution Workers' desires to block privatization might also in- can be combined with the advantages of case-by-case crease rapidly in the near future if, as expected, unem- sales. ployment rates rise sharply in Eastern Europe. Workers might assume, with some justification, that their job Current Ownership Patterns in Poland tenure will be undercut by a privatization of their firm. Even if workers in a particular enterprise do not actually The government of Poland aspires to an ownership block privatization, they may attempt to bargain with structure like that of Western Europe. First, the bulk of the government, demanding, for example, a cut in the the productive sector should be held in private hands. enterprise's debts, or various guarantees on employment Second, the ownership of large enterprises would be pre- levels as their "price" for letting the privatization go for- dominantly in corporate form, with shares held by ward. If the government becomes enmeshed in case-by- households, financial intermediaries and other non-fi- case bargaining, there will be no end in sight, given that nancial firms. Third, financial markets should be devel- hundreds of large enterprises must be privatized. oped to facilitate the trading of shares. These fears of paralysis are not hypothetical. In al- As can be seen in table 20-1, the proportion of enter- most all countries where privatizations have been at- prise capital in state hands in Poland and the rest of East- tempted, there have been major political obstacles to the ern Europe is vastly above the proportion in Western case-by-case approach. These obstacles are likely to de- Europe.9 Nowhere in Western Europe does the share of velop more rapidly in Eastern Europe given the thinness state ownership currently exceed 20 percent. Even the or, in some cases, non-existence of capital markets, the Social Democratic regimes in Scandinavia-which serve difficulties of valuation, the likelihood that privatiza- as a role model for many politicians in Eastern Europe- tions will be followed by lay-offs and the sheer number of have very little state capital, and almost none in the in- firms that must be privatized. Already, after just a hand- dustrial sector.10 233 Table 20-1. Size of the State Sector, Measured by subjected to strong international competition resulting Output and Employment, Selected Countries from a policy of low tariffs and free trade, the problem of (percent) industrial monopoly is rather small even in sectors that Country Output Employment are dominated by a few firms. Indeed, the Polish Ministry Command economies of Finance recently reviewed the competitive environ- ment facing the 500 largest industrial enterprises and Czechoslovakia (1986) 97.0 ... concluded that there are few cases, if any, where con- East Germany (1982) 96.5 94.2 cerns over monopoly power should stand in the way of Soviet Union (1985) 96.0 - rapid privatization. Poland (1985) 81.7 71.5 To some extent, the figure of 3,177 state-owned in- China (1984) 73.6 ... dustrial enterprises exaggerates the task of privatization, Hungary (1984) 65.2 69.9 since output and employment are concentrated in the Market eco sa top few hundred firms. The size distribution of firms by Market economies employment, output and earnings is shown in table 20- 2. Note that the top 500 firms, ranked by sales, account Austria (1978-79) 16.5 14.6 for 40 percent of employment, 66 percent of sales and 68 Austral (1978-79) 14.5 13.0 percent of new income in the socialized industrial sec- Italy (1982) 14.0 15.0 tor.13 If most of the top 500 firms can be privatized, then Sweden ... 10.6 an important part of the industrial sector, as measured Finland ... 10.0 by employment, sales and net income, will have been United Kingdom (1978) 11.1 8.2 privatized. Germany, F.R. (1982) 10.7 7.8 It is useful to divide the task of privatization into Portugal (1976) 9.7 ... three categories depending on the size of the firms. Denmark (1974) 6.3 5.0 Large-scale privatization refers to firms that are too Greece (1979) 6.1 ... large to be sold or distributed to a single Polish buyer. Norway ... 6.0 For these firms, privatization will require that owner- Spain (1979) 4.1 ... ship be vested in several groups of owners or sold to a Netherlands (1971-73) 3.6 8.0 large foreign buyer. Medium-sized privatization refers to United States (1983) 1.3 1.8 firms in which a majority stake can be sold to a single a. Excludes government services, but includes state-owned enterprises en- buyer, either through a management-worker buy-out or gaged in commercial activities. Source: Milanovic (1989), Tables 1.4 and 1.7. in a sale to an outside group. Small-scale privatization refers to very small firms, mainly in retail trade but also to some extent in the industrial sector. In these cases State-owned Industrial Firms in Poland privatization will be accomplished mainly by transfer- ring ownership to the firms' workers through a lease or Poland had 3,177 state enterprises in the industrial an outright purchase. sector as of 1988.11 The country aims to privatize most or As a working assumption, the Polish authorities are all of these firms, so that the ownership structure in in- presently focusing on the 500 largest industrial firms as dustry will approach Western Europe's. As in Western the universe for large-scale privatization, although the Europe, privatization is likely to come first in industry privatization program will simultaneously go forward on and small-scale services and then later, and perhaps to a small-, medium- and large-scale enterprises. Since lesser extent, in utilities, communications and railway large-scale privatization is the most difficult kind of transport. Privatization is typically more complicated in privatization, and since such privatization will involve these latter cases, since the output markets in these sec- the bulk of the industrial sector, the focus in this paper tors tend to be less competitive than in industry. There- is on these firms. fore, in those cases privatization must be accompanied by a regulatory apparatus for controlling the pricing and Legal and Political Aspects of Ownership output decisions of the privatized firms.'2 While some observers have worried that privatization Many of the deepest problems of privatization arise in Eastern Europe should be delayed until after wide- from the ill-defined current ownership structure in Po- spread demonopolization has occurred, the authors do land and in the rest of Eastern Europe.'4 In Western not believe that monopoly power is an urgent problem in thought, ownership entails several rights over a re- the industrial sector. Since most Polish industry is now source: the right to determine its use; the right to the 234 Table 20-2. Poland: Size Distribution of State Enterprises, 1990 Salesa Net Income a Employmentb Number Billions Percent Billions Percent Thousands Percent of of of of od of of enterprises US$ total US$ total workers total Top 100 18.1 39 2.9 43 711 . 18 Top 200 23.1 49 3.6 53 1,036 26 Top 300 26.5 57 4.0 59 1,261 31 Top 400 29.1 62 4.4 65 1,461 36 Top 500 30.9 66 4.6 68 1,612 40 Total of socialized industrial sector 46.8 100 6.8 100 4,051 100 Note: Firms are ranked by sales in the Lista 500 and Informacja Statystyczna. a. Data are annualized from January-June 1990. b. Employment refers to 1989 employment levels. c. Sixteen firms have been dropped from the top 500 firms. The remaining 484 firms are used as a proxy. Source: Lista 500 and Irnformacja Statystyczna. earnings of the resource; and the right to dispose of the enterprises did blur many of the ownership rights previ- resource. In Eastern Europe, however, those rights are ously exercised by the central bureaucracy. now rather vaguely distributed among the workers, managers and state bureaucracies. Workers' Councils Under classical central planning, the state main- tained all ownership rights. The enterprise had a "found- It also raised the question of which group was to as- ing organ," usually a branch ministry, which maintained sume these rights. In all countries, the formal answer formal oversight of the enterprise. The enterprise's func- was that the workers of the enterprise exercised respon- tions, of course, were largely determined by the central sibility for the enterprise, usually through a collective plan. It was during the process of reform Communism assembly of all the enterprise workers (such as an annual (in Yugoslavia after 1948, Hungary after 1968, Poland af- shareholders' meeting) and through a representative ter 1980 and the Soviet Union after 1985) that the situa- workers' council, which was to govern the enterprise be- tion became highly muddled. tween meetings of the general assembly. Part of the process of reform Communism in all these A workers' council in Poland is composed of 15 work- countries was the granting of increased operational and ers who are elected by secret ballot for two-year terms. financial autonomy to the enterprise.15 In Poland, these The councils are charged with "the information of the reforms were adopted in September 1981, with the Law enterprise's annual plans, investment decisions, changes on State Enterprises and the Law on Self-Management of in the scope of enterprise activities, acceptance and ap- the State Enterprises' Employees, just before the impo- proval of the annual balance and decisions on merger sition of martial law. In theory, the enterprises were to and enterprise dissolution" (see Sajko 1987, p. 1372). In become largely self-managing and self-financing, with most enterprises, the workers' councils have been given less reliance on transfers from the state. Investment the power to appoint the manager. In enterprises of "pri- spending was to come increasingly from retained earn- mary importance" the workers' councils have been given ings rather than from a budgetary allocation. The enter- veto power over the choice made by the government. prises were to develop more of their own work plans, Note that the development of the workers' council with less top-down direction from the bureaucracy. In was not the only logical alternative for Eastern Europe, this way, enterprises were to become increasingly sub- although it was universally taken to be the one most ject to market forces. consistent with Marxist ideology. Another possibility, for It is well-known that these decentralizing reforms in example, could have been the classic state-owned enter- Poland, and elsewhere in Eastern Europe, were at best prise in the British tradition, which is governed by an in- partial, and failed to create hard budget constraints for dependent board of directors appointed by the the enterprises."6 Under reform communism enterprises government.'7 The board of directors possesses some op- escaped from central planning into an endless series of erational independence from the government and ap- negotiations with the bureaucracy over taxes, subsidies, points the management, approves its plans and monitors prices and output rather than into a true market envi- its performance. ronment. Nonetheless, the devolution of power upon the 235 Interestingly, privatization would now be much easi- As the room for maneuver increased for the enterpris- er had enterprises taken the form of British state enter- es, the problems arising from the muddled ownership prises. In that case, the enterprise would already be in structure became acute. Workers and the managers the corporate form, and privatization would merely in- gained control over the firm but lacked clear title to the volve the transfer of control from the board of directors assets of the firm; they therefore had an overwhelming to the new private owners. In the case of worker-man- incentive to try to appropriate the capital income of the aged firms, however, privatization requires a disenfran- firm and to strip the firm of its assets. As Lewandowski chisement of the workers' council at the same time that and Szomberg (1989, p. 258) put it, "There is still no one corporate governance is vested in a new board of direc- to lose anything from the decapitalization of state prop- tors. erty." For example, with increased freedom to set wages, In practice, the workers' council played a lesser role, the workers' councils pressured managers to raise wages even under the communist system, than set forth in the to absorb an increasing amount of the firms' cash flow. legislation. While workers' councils were set up in Liberalization in 1987-88 therefore ushered in an enor- roughly 90 percent of Poland's state enterprises, it is mous real wage explosion and a wage-price spiral.'9 commonly estimated that they have been fully effective The possibilities for asset stripping became even more in only about 15 percent of the enterprises, al though dramatic when various property laws were changed dur- these have typically been among the larger enterprises ing 1988-89, in the final period of Communist rule. (Rudolf 1988, p. 424). In many other firms, where work- First, the private sector was liberalized, and state firms ers' councils were not effectively organized, the state en- were allowed to do business with the private sector. Sec- terprise managers actedfwith considerable independence ond, joint ventures with foreign partners were encour- terpisemanger actd wth onsderale ndeendnce aged. As a result of the changes, managers quickly not only from the workers but from the state bureaucra- discov reduay o te state propert byimkin cy as well. discovered ways to appropriate state property by making c-as well.,sweetheart deals with an outside partner, in a process that quickly became known as "spontaneous privatiza- tion." Lewandowski and Szomber called the process "le- The Communist party, when it was in power, also gal parasitism." played a key role in management of the enterprises. Not Consider the case of joint ventures. The enterprises only was the choice of managers heavily influenced by were given enormous freedom to enter into joint-ven- the "nomenklatura system," in which the party nomi- ture arrangements with little central control. Many state nated politically acceptable directors, but also important managers, particularly in firms without strong workers' decisions for the enterprise were referred to the "enter- councils, quickly realized that they could make deals prise collective body," which was composed of the first with foreign partners. They would grant the foreign partner a highly favorable stake in the enterprise, and in secretary of the party, together with the managig direc- return the foreign partner would grant the manager an and the chairman of the workers' council (Rudolf 1988 attractive position in the new venture. The manager ef- an the c fectively traded the state property for personal gain. p. 426). Even more egregious cases ensued in which the state After the collapse of Poland's Communist rme,sthe enterprises entered into contracts with newly estab- independence of the enterprises increased enormously. lished private firms in which the manager had a personal Of course, the nomenklatura system collapsed, as did stake. The manager might then lease to the private firm the direct intervention of the party in enterprise mat- the plant and machinery of the state enterprise at highly ters. In many enterprises, the workers' councils gained favorable terms. The profits of the state enterprise would new strength and ousted enterprise managers. In other thereby be transferred to the private firm. These abuses cases where the workers' council remained weak, the became recognized by the general public in 1989 and led state managers inherited the powers ceded by the bu- to an enormous storm of protest. The public realized reaucracy and the party. Although in March 1990 the that during the Communist-managed transition to a rights of the workers' councils were reaffirmed, empha- market economy, the Communist-appointed managers sizing that "the council must approve the decision by the were appropriating the best property for themselves. The enterprise to sell assets, to form or to enter into an exist- public's revulsion against spontaneous privatization was ing trade company, to buy shares of stock in corpora- intense and surely helped to speed the demise of the old tions, or to sell off shares in those corporations," there regime. have been few legal changes in the power of the workers' Since the start of 1990, Poland's industrial enterpris- councils.1" es have had almost complete freedom with regard to 236 pricing, production and international trade decisions. most legal scholars in Poland assert that the enterprise Thus the autonomy of the enterprises has been greatly fund as well as the state fund belong to the state. enhanced. At the same time, administrative means have been introduced to cope, however imperfectly, with the Pnratization and Worker Management anomalous ownership structure. Wage controls have been put in place to prevent enterprises from appropri- The shift to a Western European ownership structure ating their own profits, and indeed profitability of the in- will require that enterprise governance be removed from dustrial sector increased markedly during 1990, despite the workers' councils and managers and placed squarely a domestic recession. in a supervisory board (or board of directors) controlled 20 While the government's oversight of joint ventures by the owners of the enterprise. In essence, privatiza- has been strengthened, there is anecdotal evidence to tion requires first that certain ownership rights now suggest that at least some joint-venture arrangements vested in the enterprises, and particularly in the work- still serve the self-interest of managers. In addition, ers' councils, be eliminated so that the property rights while new conflict of interest laws have been put in place can be transferred to the real owners. to prohibit spontaneous privatizations through sweet- The transfer of power to private owners poses a signif- heart deals, the effectiveness of these regulations is not icant political challenge to the governments of Eastern yet clear. The government has also undertaken to review Europe, since support for worker control and worker and reverse cases of conflicts of interest that occurred in ownership remains powerful in many enterprises and in the past. some political circles. Not only was the worker manage- In pasome ways, however, recent reforms have madeit ment ideology validated in the years of struggle against In some ways, however, recent reforms have made it cnrllni u tas n elo nzds harder, not easier, to define real claimants to the assets central planng, but it also enjoys well-orgamzed sup- '. residual income of a firm. On the one side, the gov- port in enterprises that have active workers' councils. arndent residualwincom a firm. On the on se,thergov These councils now aim to strengthen their control and eem ine s awn a shar e b.detw a nterprise perhaps to win outright ownership of the firms. In enter- receive s ewearni andfer m the state budget.E ntprises nw prises facing cutbacks in employment, many workers receive fewer transfers from the state budget and are al- view their workers' councils as the best hope for avoid- lowed to retain more of their earnings after paying vani- ing lay-offs and protecting workers' interests, if not the ous corporate taxes, but they are not free to distribute interests if not tha those earnings through wage payments without incur- interests of the capital. thoseearnigs though age aymens witout icur- It is clear that for political reasons, and perhaps for ring substantial penalties under the excess wage tax law. ethical reasons as well, that workers will have to be par- In addition, because there are clearly no private share- tially compensated for the transfer of control from the holders to whom the profits can be distributed, the resid- workers' councils to private owners. After all, workers ual income remains with the firm. There, the retained stand to lose property rights because of privatization. earnings can be used for only two purposes: to amass fi- The standard suggestion is that workers should be al- nancial wealth; or to make investment expenditures. lowed to purchase a limited proportion of shares (20 per- It can be expected that unless the claimants of the re- cent according to the Polish Privatization Law) in their sidual income are quickly defined through the privatiza- enterprise at a discounted price when the firm is priva- tion, management and workers will, through indirect tized. The practical question is whether this concession means, find ways to enrich themselves. Some of the prof- will be politically palatable, at least enough so to permit its might be enjoyed through a reduced work effort or the privatization to move forward smoothly. It is possible through low-return investments, such as in new factory that even with the concession of favorable purchase cafeterias, for example, that benefit the incumbents of terms for shares, some workers' groups will fight to slow the firm. the privatization to assert their claims for complete In yet a final facet of the ownership rights debate, worker control. most legal analysts in Poland believe that the state re- There are strong reasons, on grounds of equity as well mains the holder of enterprise capital, but even this as efficiency, for rejecting the workers' strong demands point has become muddled. The book value of the capital to deliver greater firm ownership to them. First, workers of a firm has been divided between the "founder's fund," were never really granted ownership of their enterprises which represents the initial capitalization of the firm, in the Communist reform because doing so would have and the "enterprise fund," which reflects the book value created vast and unjustifiable inequalities in income and of retained earnings. Some managers and workers' wealth. Only 4 million of Poland's 18 million workers are groups have asserted that the enterprise fund should be actually employed in the state-owned industrial sector, viewed as already belonging to the enterprise, while so that less than one-fourth of the workers would receive 237 any benefits from the full distribution of ownership to the objections of workers' groups, who rejected the the industrial workers. Moreover, some workers are in state's claim to sole ownership of the enterprises. highly profitable firms while others are in loss-making The Mazowiecki government then went ahead with firms. Thus, to give the workers ownership of their re- the Act on Privatization, which in many ways reflects the spective firms would be highly capricious in its distribu- confused state of affairs. The new law allows for the tional consequences. transformation of state enterprises into treasury-owned From an efficiency standpoint, it makes little sense joint-stock companies under the direction of a new Min- for workers to own their own enterprises (or to lease the istry of Property Transformation, and also provides that capital in their enterprises), except in the case of small, the process should be approved by the state enterprise labor-intensive operations. Worker ownership or labor- manager, the workers' council and the founding organ. management tends to cut firms out of the capital mar- Thus, the enterprises are given a veto. The state, howev- kets. Outside investors tend to shun enterprises in which er, retains a trump card, since the Prime Minister, on the workers have a controlling interest, since 3they can act motion of the Minister of Ownership Transformation, opportunistically to absorb all of the income of the firm can order the transformation of an enterprise. In the in the form of wage compensation.21 Moreover, it makes end, there remains the delicate balancing act between little sense for workers to hold most of their equity the interests of workers, managers and the state. wealth in their enterprise, since on diversification The act has several other key provisos but in general grounds they should hold financial capital and human provides enough flexibility that the government has a capital in separate investments. very wide range of options. Most of the legislation seems Ironically, many supporters of US employee stock to envision a standard case-by-case privatization. Several ownership plans (ESOPs) have traveled to Eastern Eu- articles in the act detail how shares are to be sold to the rope recently to organize support for worker ownership. public, with the workers getting a discount of half-price The effect in many ways has been pernicious. In the 23 O a f Unite Stats, th ESOPtendsto prmote modet ex on up to 20 percent of the shares. Once again, flexibility UJnited States, the ESOP tends to promote a modest ex- is built into the act by allowing the Council of Ministers tent of worker ownership, almost always below the 20 (h eutive body ofith goernment) o Mitiste percent threshold envisioned by the Polish authorities.22 (the executive body of the government) to permit State percent threshold esed bte Plish austrite Treasury shares to be sold in a different manner to those firms in the United States that employ more than 1,500 specified in the act. The act also provides that shares firmers in the ch UnitedpStes thatwempl more than 1,50t might be distributed for vouchers, or "scrip," issued by worktheshares.Therearestillfewer in which employsore 5 eret the government but does not specify how that is to be ofn e hajoreys.theae stil ewerpin. white worksa done. The Deputy Prime Minister publicly stated during count for a tiny proportion of the industrial labor force. the debate on the law that the free distribution of shares, Yet, in Poland, the US ESOPs have been known to sup- through one mechanism or another, would play some port the political case for more complete worker man- role in the privatization. agement and ownership. During the period between corporatization of the firm and its sale, the enterprise is to be governed by a su- ThePolish Privatization Law pervisory board, of which two-thirds are appointed by the government and one-third elected by the workers. A In June 1990, the Polish Parliament overwhelmingly maximum of two years can pass between the time of cor- passed the new Act on the Privatization of State-Owned poratization and privatization (defined as the point in Enterprises, which is designed to be the guiding legisla- which the state's holdings fall to less than half of the tion for privatization. The act does not embrace a single shares). strategy for privatization but rather sets a general frame- The law also restricts the role of foreign investors, work that could accommodate a variety of strategies. who are allowed, in the aggregate, to purchase up to 10 The history of the legislation is enlightening. In the percent of the shares of an enterprise without restriction fall of 1989, just after the Solidarity-led government as- but who must obtain government approval for owner- cended to power, it tried to introduce a Law on Transfor- ship of more than 10 percent. The Mazowiecki govern- mation of State Enterprises, which would have ment had stressed that it would be liberal in granting converted state enterprises into joint-stock companies access to foreign investors, although it emphasized that with the treasury as the sole owner (this process is usu- the process must be controlled to prevent abuses and to ally termed "corporatization," or sometimes "commer- monitor the extent of foreign participation in the econ- cialization" ). The Transformation Law stalled in the omy. Since the law builds in so much flexibility, it is Parliament and was ultimately withdrawn as a result of clear that politics, and policy choices, rather than the le- 238 Table 20-3. United States: Employee Ownership of Large Industrial Firms Persons Company Business employed Ownership greater than 50 percent Science Applications Research and development 10,000 Parsons Corporation Engineering 10,000 Amsted Industries Manufacturing 8,300 Weirton Steel Steel manufacturing 8,100 Avondale Shipyards Shipbuilding 7,500 W.L. Gore Associates High-tech manufacturing 5,000 Simmons Company Furniture manufacturing 4,900 Republic Engineered Steel Steel manufacturing 4,900 Graybar Electric Electrical equipment manufacturing 4,700 Treasure Chest Advertising Printing 4,000 National Steel & Shipbuilding Shipbuilding 4,000 Stebbins Engineering Engineering 4,000 CH2M Hill Engineering 2,300 Northwestern Steel and Wire Steel Manufacturing 2,300 McLouth Steel Steel manufacturing 2,000 Cranston Print Works Textile printing 1,800 Okonite Company Wire and cable manufacturing 1,700 North American Rayon Rayon manufacturing 1,500 Total 87,000 Percent of total manufacturing employmenta 0.4 Ownership between 25 percent and 50 percent Phillips Petroleum Petroleum 28,400 PMC Corporation Industrial manufacturing 28,000 Tyson Foods Chicken processing 25,000 Ashland Oil Oil refining 22,800 USG Construction material 22,000 Colt Industries Industrial products 19,700 Olin Corp. Chemicals/defense 17,000 Hallmark Cards Greeting cards 15,521 Lowe's Companies, Inc. Lumber and hardware 14,700 CBI Industries Energy and manufacturing 11,400 Stone and Webster Engineering 10,000 Harcourt Brace Jovanovich Publishing 8,000 Herman Miller Inc. Furniture manufacturing 5,000 Dentsply International Dental equipment 3,500 Applied Power, Inc. Automotive equipment 3,400 Swank, Inc. Leather goods 3,100 Anderson Corporation Window manufacturing 2,800 Tyler Corp. Cast iron pipes 2,700 H.K. Porter Manufacturing 2,600 Nationwide Automotive Auto supplies 2,200 U.S. Sugar Sugar manufacturing 2,100 Granite Construction Co. Highway and heavy construction 2,000 Anthony Industries Recreation products 2,000 CF&I Steel Steel manufacturing 2,000 Quad/Graphics Printing 2,000 Ormet Inc. Aluminum manufacturing 1,700 E-Systems Electronics 1,500 Total 261,121 Percent of total manufacturing employmenta 1.4 a. Total manufacturing employment in US equals 19,167,000 employees. Source: National center for Employee Ownership (1989) and U.S. Labor Department. 239 gal constraints of the new privatization act, will deter- tory institutions such as the Securities and Exchange mine the pace and strategy of privatization. Commission, which pursue investigations of malfea- In August and September of 1990, Deputy Prime Min- sance and which require various forms of disclosure that ister Balcerowicz publicly outlined the shape of the are widely analyzed by the investment community. privatization program to be developed by the Mazowiec- The basic proposal described in the next section, is ki government. The program would not be based on the that the ownership of most enterprises should be divided case-by-case sale of medium- and large-scale enterprises up into tranches among various groups and financial in- through initial public offerings. Instead, privatization stitutions that will each have an incentive to monitor the would be accelerated by means of several techniques, in- enterprise and promote sound management. In the pro- cluding, among other things, the discount sales of posal, a portion of the shares is sold or given to workers; shares to workers and a voucher system to place shares another part is transferred to pension funds and com- in the hands of the public. mercial banks; another part is transferred mainly to mu- tual funds that in turn will be owned by individual Corporate Governance and Financial households; and another part is sold in a block to a "core Intermediaries investor" group that commits both to hold a substantial proportion of the shares for several years and to manage A privatization program in Eastern Europe must do the firm. more than simply return enterprise ownership to private hands. The government should also strive to create an A Strategy for Achieving Effective Corporate ownership structure that will effectively oversee the Govemnance management of the newly privatized assets. That is, the There are two essential tasks in establishing effective government should foster an effective structure of cor- governance of the productive capital now in state hands. porate governance. Moreover, the government should The first, more urgent, task is to introduce a provisional encourage the development of financial intermediaries system of corporate goverance that can monitor a that will be important both for monitoring enterprises firm's management and prevent both managers and and for allowing the private sector to diversify the risks workers from squandering its capital income and capital of property ownership. assets before full privatization takes place. The second, Creation of adequate long-term oversight of manage- longer term task is to foster a structure of ownership in ment is strongly emphasized for two main reasons. First, which the new private owners will be in a strong position there can be little confidence in the current managerial to manage their newly acquired assets. class in Eastern Europe. Many managers owe their posi- A vital first step to privatization is the conversion of tions to their Communist party allegiances rather than state enterprises into corporate form (their corporatiza- to their technical competence. Further, many compe- tion), to concentrate the property rights of the enter- tent managers won their positions because of their engi- prise in a corporate board of directors appointed by the neering expertise, which was crucial in a planned owners. Inevitably, given a realistic timetable for any economy, rather than their ability to navigate the enter- privatization scheme, the initial boards of almost all en- prise in the uncharted waters of an open market econo- terprises will have to be appointed by the government, my. Thus, an enormous effort will have to be made to with the subsequent boards to be appointed by private evaluate current managers and to train and promote owners as they emerge during the privatization. To re- new managers. duce the enormous administrative burden of creating a Second, Eastern Europe lacks many of the individual large number of corporate boards, the task should ini- and institutional actors that are normally involved in tially be focused on the 500 largest firms. corporate governance in the West. Therefore, special The urgency of corporatization results from the com- care must be taken to assure that at least some institu- plete inadequacy of the current structure of governance, tions capable of effective corporate governance are cre- in which the manager is completely unmonitored in ated. In particular, unlike in normal market economies, most firms or is monitored by workers' councils in oth- Eastern European countries cannot rely on corporate ers. It is stressed here that the lack of effective gover- oversight by: the original families and entrepreneurs nance has already produced enormous social, political that established an enterprise; outside directors that and economic problems that are likely to continue or have a long involvement with an enterprise and under- worsen despite attempts at administrative controls. stand its history and corporate culture; a vast financial Corporatization was one of the first steps in the re- press and investment analysis sector that investigates, cent German economic unification. The Treuhandan- reports and evaluates management behavior; or regula- stalt, the state property trust, mandated the conversion 240 of all 8,000 state enterprises into corporate form by ad- which hundreds of thousands or millions of owners each ministrative decree. So far, the Poles, Hungarians and retain a small number of shares. Most ownership of the Yugoslavs have avoided this step for fear of running into large enterprises should be held by intermediary firms political conflict with the powerful workers' councils such as pension funds, mutual funds or commercial (known as "enterprise councils" in Hungary). These gov- banks, or by large owners with concentrated stakes. This ernments also fear being accused of politicizing the principle also conforms to the idea that small investors economy by attempting to reconcentrate control in the should hold diversified portfolios, through mutual funds hands of the state. perhaps, rather than shares in a single enterprise. The political risks of corporatization pale in compar- Second, the privatization should be designed to foster ison with the potential economic gains. If the principle at least one significant non-financial investor in each is not accepted early in the privatization, managers and major industrial enterprise. This investor would hold workers' councils together with the state will likely end around 20 percent of the shares and would create a "sta- up bargaining over the terms for privatizing on a firm- ble core" of ownership of the firm, to use a concept de- by-firm basis, thus paralyzing the process. Moreover, veloped in the French privatization. In the privatizations without early corporatization there will be no effective of the 1980s, the French believed that their capital mar- way of replacing the most incompetent managers in en- kets were too thin to rely primarily on public placements terprises that have weak or non-existent workers' coun- as the dominant method of privatization. They also cils. Since the boards of directors will monitor joint- lacked the investment banks that had guided public venture arrangements and other major corporate deci- placements in the United Kingdom. Most importantly, sions, the ability to forestall "spontaneous privatization" there was concern that no single owner or ownership will be greatly enhanced. group would emerge with a significant stake in each en- At the same time, the risk of politicizing the process terprise. through appointing boards of directors can be signifi- Therefore, the French devised a scheme known as the cantly reduced if, first, the task is largely and visibly sub- "stable core." This involved soliciting a bid from a single contracted to professional institutions, such as a investor, or group of investors working together, to buy consortium of Polish and international management ad- 20 percent or more of an enterprise. This group commit- visory firms, and, second, the initial board of directors is ted a sum of money, prepared a management proposal made provisional until the enterprise is at least partially and submitted its financial bid. It also committed to hold privatized (at which point, the new owners would as- its shares for at least five years. After review of the bids, sume some of the responsibility for forming the board of the French government designated a winner to serve as directors). In fact, Polish company law requires that the the stable core. The winning bid was selected not only on first board of directors be constituted for one year, while the share price offered, but also on the financial strength subsequent boards of directors be constituted for a of the bidder, its reputation and experience, and the three-year tenure. quality of its management plan.2" The long-term challenge of the privatization program In Eastern European privatization plans, as in the is to create a structure of ownership in which the owners French plan, governments would attempt to market a 20 have effective control over the enterprises. For example, percent block of shares in each enterprise to an invest- if ownership of the enterprises is too widely dispersed, ment group that could be domestic, foreign or mixed. the individual owners will have very little incentive to This group would be vested with significant representa- monitor the management.24 Moreover, it will be useful tion on the corporate board in return for a requirement to match firms with appropriate owners in the privatiza- that the investor group hold the enterprise for a speci- tion itself, rather than relying on subsequent trading to fied period of time, perhaps three to five years. establish the "right" owners for the firm. The market for Furthermore, the Eastern European countries should corporate control through takeovers is highly flawed, create a legal and institutional environment in which fi- with significant externalities and asymmetries of infor- nancial intermediaries play a more active oversight role mation. Therefore, the market cannot be relied upon to than is typical in the United States and United Kingdom. do a good job in matching potential owners and firms: The need for oversight by financial intermediaries re- many efficient takeovers may never be achieved, and sults from the lack of other institutions or individuals in many inefficient takeovers may be consummated.25 Eastern Europe that can be relied upon to help oversee These concerns lead to three principles for establish- an enterprise's corporate management. Thus, a great ef- ing effective long-term control over privatized firms. fort will be needed to economize on information that First, the privatization should avoid creating an atomis- will be vital in corporate governance. Since banks, pen- tic ownership structure for the large enterprises, in sion funds and mutual funds will have such information, 241 these institutions should also be assigned a major role in The Role of the Stock Market in Privatization the governance. As a first step toward strengthening the hand of the fi- It is doubtful that the stock exchange can or should nancial intermediaries, the Eastern European econo- play a major role in the privatization or more generally mies should aim to develop universal banking, as in in the development of financial markets in Eastern Eu- Germany and Japan,27 where the commercial banks hold rope in the next few years. While each country will surely stakes in corporate assets and play an active role in the develop a stock exchange, the liquidity of the new ex- oversight of the enterprises.28 The new banks should changes will be low, and their capacity to raise corporate place representatives on the corporate boards of direc- capital or serve as a market for corporate control will be tors and strengthen their capacities to participate in the highly circumscribed. Moreover, information concern- restructuring of troubled firms. Of course, qualified ing the fundamental valuation of firms will continue to banks cannot be created at once, but it will be far easier be limited for many years, since firms will be operating to build up the operational capacity of a dozen large for the first time in a market environment. Thus, it can banks (perhaps through management contracts with be expected that the markets will be subject to extreme foreign banks) than to rely on the decentralized over- volatility and will tend toward insider trading, not only sight of thousands of individual enterprises.29 because asymmetries of information will be pronounced, As a second step toward strengthening the oversight but also because the policing of the exchanges will be by financial intermediaries, newly-created mutual funds imperfect in the first few years. Nor are there likely to be and pension funds should be encouraged to appoint rep- many firms with a capital value that is large enough to resentatives to the boards of directors of enterprises in support many institutional investors and small share- which they hold shares, and to create the institutional holders. capacity to monitor closely a large number of firms. In general, the continental European economies rely When the mutual funds and pension funds are initially much less heavily on stock markets than do the United licensed, for example, they could be required to present Kingdom and the United States for raising and trading plans detailing how they will appoint directors to corpo- corporate wealth and for gaining corporate control.32 rate boards and how they will develop the expertise to One sign of the smaller role of the stock market in West- monitor these companies. One possibility is that they ern Europe is the low number of listed companies in might subcontract the management oversight to an in- most European countries. Austria, for example, has only ternational management consulting firm. 81 listed Austrian firms; Finland, 78; Norway, 122; Swe- The proposal that mutual funds and pension funds be den, 135; and Switzerland, 177. In addition, hostile take- so actively involved in corporate oversight may seem un- overs in these countries scarcely occur. There is little usual, but it should be noted that several recent analysts evidence to suggest, however, that this relative dearth of have argued that the same change should apply in the stock market activity has hindered effective corporate United States.? One proposal in the US context is for the governance: many observers feel that the opposite is institutional investors to appoint a new class of profes- true-that active stock market trading encourages a sional, outside directors who would serve full-time on a short-term bias in managerial decisions. Given the diffi- small number of boards of directors (perhaps six boards culties of establishing an active stock exchange for a per person) for corporations held by the investors.3' large number of firms, it seems safe to recommend that These new directors would constitute a "critical mass" Eastern Europe follow the West European lead, at least on each board, but would serve together with traditional, initially, and downplay the institutional role of the stock outside directors (such as chief executive officers [CEOs] exchange. of other companies) and inside directors. The proposal for Poland is more natural in view of the fact that the A Strategy for Privatization newly developed financial intermediaries will have a much smaller task with regard to portfolio management At the outset a basic question was posed: how can a than is usual in the United States. The reason is that the very large number of enterprises be privatized quickly market for corporate shares will be considerably smaller while at the same time establishing an effective struc- and less liquid there than in the United States. Thus, ture of corporate governance? It is argued here that a they will have the time to focus somewhat more on cor- large role must be played by new institutional investors porate governance and rather less on portfolio manage- and a relatively small role by the stock market. Follow- ment. ing is a description of a strategy for privatization that 242 might be used for transferring ownership and control to cial preparation. They tend to be time-consuming under private owners. normal circumstances and would be far slower in East- One key point in this strategy is that much of the ern European, where valuation of firms is nearly impos- privatization should be accomplished through a free dis- sible and where the financial infrastructure for initial tribution of shares to various groups, including workers, public offerings does not yet exist. Second, the financial pension funds and mutual funds, rather than through capital currently in the hands of the public is a small the sale of shares in an initial public offering, which was fraction of the value of the enterprises to be privatized, standard practice in the well-known British privatiza- assuming a reasonable market interest rate (and thus a tions. In Eastern Europe, the free distribution of shares reasonable price/earnings ratio for the firm). Thus, any helps sidestep the difficult, costly and time-consuming attempt to sell a large proportion of the enterprises process of enterprise valuation and recognizes the scar- would create serious financial problems. Third, initial city of financial capital in private hands. public offerings are typically used, as has been the case There is good reason to expect that the problem posed in Britain, to secure a widespread ownership of shares by by asset valuation will be far more severe in the case of small investors (in order to create "peoples' capitalism"). Eastern European enterprises than has been the case in While this aim is desirable, it does not generally produce the West. The economic environment is shifting in fun- an effective structure of corporate governance and, from damental ways as the transformation to market econo- a logistical point of view, is surely inappropriate for all mies proceeds. Not only is the domestic environment but the largest enterprises. Fourth, reliance upon initial undergoing a profound change, but in several countries, public offerings would lead to the privatization of only international trade is being liberalized and currency the most profitable enterprises and leave the marginal or convertibility introduced. The Comecon system govern- unhealthy enterprises-which are viewed as less mar- ing trade and payments among the Eastern European ketable-in the hands of the government. countries will be revamped in January 1991. Further- Consider each of these four points in more detail. Ta- more, the legal and regulatory environment is also ble 20.4 shows the total worldwide number of privatiza- changing. Indeed, the level of interest rates needed to tions through public offerings between 1980 and 1987, discount future profits for the purpose of valuation could according to a World Bank study. The United Kingdom be greatly influenced by the strategy for stabilization it- accomplished 13 initial public offerings during this peri- self. As a result of all of these changes, many key relative od, and the country with the largest number, Italy, ac- prices have shifted and will continue to do so. Finally, in complished 15. Many authors have discussed the many enterprises there will be changes in the structure laborious preparations involved in each initial public of- of management or in management itself that will have fering. Since there is as yet no stock exchange in Poland, profound effects on the value of the enterprise. These no Polish investment banks and, indeed, no tradition of factors each pose fundamental conceptual problems for valuing enterprises for public sale, privatization through asset valuation and, taken together, cast grave doubts on initial public offerings would likely be quite slow in East- the viability of privatization schemes that require careful ern European countries. valuation of assets as a prelude to the sale of enterpris- The normal problems of initial public offerings are es. 33 greatly compounded by the shortage of public savings Another key point in the strategy is that most enter- with which to purchase shares. Table 20.5 presents a prises should be privatized in a common manner to rough illustration of this problem. Assume that the avoid debates between the government and individual firms are valued as a multiple of annual earnings, with enterprises. To the extent that a common procedure is three alternatives considered: price/earnings ratios of 6, followed, it will be harder for individual enterprises to at- 8 or 10. In each case, the market value of the 500 largest tempt to bargain for special advantages in the course of enterprises is compared with the financial savings of pri- the privatization. At the same time, however, it should vate households, which consists of currency and bank be possible for an enterprise to proceed outside of the deposits. Even for the lowest price/earnings ratio of 6, common procedure (for example, if it receives an attrac- the value of the largest enterprises is roughly 2.5 times tive bid) and yet remain subject to various due process the current stock of savings. standards and government oversight. Thus, the shares could not be sold quickly unless one Before detailing a method of privatization that uses a of three things happened. First, the price/earnings ratio direct distribution of shares, consider the four funda- could fall sharply, leading to the sale of enterprises for mental limitations of initial public offerings in the East- close to nothing to those in the public who hold cash. ern European context. First, public offerings require a Second, the government could lend money to the public careful valuation of each firm and a great deal of finan- for share purchase, in the form of grand leveraged buy- 243 Table 20-4. Completed Pnivatizations by Method, Various Countries, 1980-87 Completed public Completed private Completed workerl Country offerings sales management buy-out Africa Equatorial Guinea 0 6 0 Gabon 0 1 0 Gambia 2 1 1 Ghana 0 1 0 Guinea 0 25 0 Ivory Coast 1 18 5 Mali 0 2 0 Mauritania 0 1 0 Niger 0 12 0 Senegal 0 4 0 Sierra Leone 1 1 0 Uganda 0 1 0 Zaire 0 0 0 Asia and the Pacific Bangladesh 4 1 0 Japan 3 2 0 Korea 2 5 0 Malaysia 2 5 0 Singapore 12 0 0 Sri Lanka 1 4 0 Taiwan 2 3 0 Thailand 1 1 0 Papua New Guinea 1 0 0 New Zealand 1 3 0 Europe Austria 2 0 0 Denmark 1 0 0 France 14 10 1 Germany 6 0 0 Italy 15 17 1 Netherlands 1 0 0 Spain 5 25 0 Sweden 1 1 0 Turkey 2 0 0 United Kingdom 13 18 12 North and South America United States 1 2 0 Brazil 6 46 0 Chile 12 24 1 Jamaica 2 22 0 Mexico 3 7 0 Source: Vuylsteke (1988a), table 1, Annex E, pp. 169-72. 244 Table 20-5. Poland: Capital Value of the Largest State Enterprises, 1990 (billions of US dollars) Ratio to household Number of Net Estimated capital value financial savings? enterprises incomea P/E=6 PIE=8 PIE=10 PiE=6 P/E=8 P/E=10 Top 100 2.9 17.6 23.5 29.4 1.6 2.1 2.7 Top 200 3.6 21.6 28.8 36.0 2.0 2.6 3.3 Top 300 4.0 23.7 31.6 39.5 2.2 2.9 3.6 Top 400 4.4 26.3 35.1 43.8 2.4 3.2 4.0 Top 484 4.6 27.5 36.7 45.9 2.5 3.3 4.2 a. Data are annualized from January-June 1990. b. Ratio includes household holdings of cash and domestic and foreign currency deposits in the banking system, which totalled US $11 billion in mid-1990. Source: Authors' own calculations. Data were provided by the authorities and Informacja Statystyczna. out. Third, the enterprises could be sold mainly to for- can be used solely to purchase shares. The initial distri- eigners. None of these alternatives would be remotely bution of vouchers can be made on an equal basis for the acceptable on a political or economic basis. entire population or according to other criteria. Care Interestingly, there is one case in which a country did would have to be taken to ensure that vouchers do not try to sell a large number of enterprises very cheaply in become money substitutes, otherwise the issue of the context of an enormous credit squeeze-Chile, dur- vouchers could lead to a sudden expansion of the money ing 1975-78, just after the coup that toppled Salvador Al- supply that is not synchronized with the sale of shares.36 lende and brought Ernesto Pinochet to power. There, As will be mentioned later, the voucher schemes are still 232 firms were rapidly sold to the public through lever- likely to suffer from the other weaknesses of initial pub- aged buy-outs. The outcome in Chile confirms the worst lic offerings: their time-consuming nature and their in- fears of such a process.34 ability to create an effective ownership structure. Not only did the Chilean experiment produce weak In particular, the use of a voucher scheme in connec- firms and undue concentration of wealth, but the gov- tion with initial public offerings may involve prohibitive ernment was (rightly) attacked for selling the firms at administrative costs. If individuals swap vouchers for very low prices to the few financial groups that had some shares, the pattern of share ownership will involve a dis- access to cash. According to Rolf Luders, a former Chil- persion of holdings in small lots. A company worth ean finance minister (1979-81) and a strong advocate of US$10 million-medium-sized by Polish standards- privatization in general: might be in the position of distributing only $300,000 in dividends per year (with a 15 percent rate of return and The first round [of privatizations] was carried out in a 20 percent dividend pay-out ratio). If a voucher system the midst of a deep structural transformation process, involved the sale of 500 enterprises to 25 million adult during which there was little interest on the part of for- Poles, it could well lead to 100,000 or more shareholders eigners to invest in the country, and which was accom- per enterprise. As a result, the administrative costs panied by considerable political, social, and economic would leave few resources for profit distribution. uncertainty... These debt-led privatizations carried out The discussion in Eastern Europe of privatization in an economically unstable environment contributed to through initial public offerings has led to the notion that generate a considerable degree of financial asset con- only profitable enterprises would be privatized. Accord- centration, financial instability, and some important ingly, the marginal or unhealthy enterprises would be macroeconomic problems.35 left to the government to restructure or liquidate. This proposition arises naturally from an approach to privati- Ironically, because the newly privatized firms were un- zation based on the case-by-case sale of enterprises, be- dercapitalized and heavily indebted, many of them col- cause the apparent difficulties of valuing and marketing lapsed during the financial crisis of 1982-83 and had to enterprises may not be sustainable. It would be prefera- be re-nationalized. ble for all enterprises regardless of their financial posi- One novel approach might rescue initial public offer- tion to be corporatized and quickly put into private ings from the problem of limited purchasing power. The hands. There are several reasons not to leave the margin- government might transfer purchasing power to the al and unhealthy enterprises in the hands of the state. public in direct grants rather than loans. This is the ba- First, these enterprises could remain in state hands for a sic idea behind voucher schemes. The government is- long period as the government tries to determine wheth- sues purchasing power in a kind of voucher (scrip) that er restructuring or liquidation is appropriate. During 245 this time, the enterprises would be targets for further as- law provides for the sale to workers of up to 20 percent set stripping by an unrestrained management. Second, a of the enterprise for half price). The allocation of shares corporate board of directors will be more competent and among the work force would be determined by the man- less subject to political pressures than the government ager with approval by the new corporate board. On this in guiding the restructuring and, if necessary, liquida- particular issue, it may be wise to require that the work- tion. er representatives on the board of directors also approve the share distribution plan among the workers. An Illustrative Plan for Large-Scale Privatization in In addition to the 10 percent distributed to the work- Poland ers, around 5 percent of the shares would be reserved for The discussion now turns to how the considerations compensation for the managers and the corporate board. of the previous sections can be integrated in a single, Managers could receive stock options or outright share workable plan. The goal would be to complete the priva- ownership as part of an incentive compensation package. tization of the 500 largest state enterprises in Poland's It is expected that the distribution of shares to managers tlzatlon~ ~ ~ ~ ~ ~ ~ ~~~~a par ofe thei compesatio packag coulde providendan industrial sector within a period of four years. While the as part of their compensabon package could provide an privatization strategy of the Polish government is still important early spur to increased efficiency within the evolving and is likely to change further after the presi- firm." dential elections, it probably will share some-although The second tranche of shares, around 20 percent of certainly not all-of the features that outlined here. In- the total, would be used to capitalize a new private pen- deed, it should be stressed once again that the develop- sion system. The shares would be distributed to several ment of this proposal has benefitted from extensive new pension funds, which would in turn be distributed discussion with Polish officials in crystallizing the ideas to enterprises and individuals to back retirement pay- To achieve a rapid privatization, a common track ments. Each pension fund would receive a portfolio of would be followed for most of the large enterprises. This enterprises and would be responsible for the active over- approach has two advantages: the minimization of nego- sight of the corporations in its portfolio. It could also tiation between the state and the individual enterprises; trade its shares. and the routinization of the process. Most firms would During the following few years, enterprises would be be privatized in tranches, with each "slice" of shares "hooked up" with the new pension funds according to transferred or sold to a different kind of investor. A mi- the size of the enterprise and the age and wage distribu- nority of firms would be sold outright in a standard kind tions of the employees. The basic idea would be for the of privatization (either a public offering or a private state to scale back its own social security payments that sale). are now made directly from the budget, and to increase At the beginning of the process, each of the 500 larg- the payments being made by the capitalized pension est enterprises would be corporatized, that is, converted fund.3" The actual transition from budgetary expendi- into a joint-stock company with the shares initially is- tures to payments by private pension funds could be sued to the state as 100 percent owner. An initial board phased in over a period of 5 to 10 years. If the pension of directors would be appointed according to the privati- funds receive the income from 20 percent of the shares zation law: two-thirds of the seats would be appointed by of the 500 top enterprises, the annual earnings of the the government (most likely by a private investment pension funds would equal about $900 million, or about group that would be hired to advise the government), 20 percent of the annual pension payments now made while one-third of the seats would be elected by the from the central budget. workers. The initial board of directors would serve for The use of share distribution to capitalize the pension one year. system is not without complexity, and numerous logisti- A few enterprises would then proceed with a British- cal problems will have to be resolved to distribute claims style privatization through initial public offerings, pri- in a fair way and reduce social security payments from vate sales or auctions, although these enterprises are not the government in line with growing benefits from the discussed in this paper. (The privatization law should private plans. leave room for individual firms to pursue these routes, The third tranche would consist of 10 percent of the especially when private bidders come forward.) The bulk shares and would be used to capitalize the existing state- of the enterprises would begin the privatization in owned commercial banks and the insurance sector. tranches. The first tranche would be the transfer, or pos- Commercial banks would receive 6 percent of the shares sibly sale, of shares to an enterprise's workers. Most sim- (60 percent of this tranche) and would be expected to de- ply, the government could mandate that the workers velop into active investors along German lines. At the receive 10 percent of the enterprise shares for free (the same time, the commercial banks themselves would be 246 converted to joint-stock companies and prepared for with a fixed face value in the domestic currency. Shares privatization. would be tendered at a fixed price after a quick valuation. The capitalization and commercialization of com- Households could either buy shares with their vouchers mercial banks would have two main benefits. First, as ac- or buy claims on investment trust companies, which in tive investors the commercial banks would begin to play turn would use the vouchers to purchase the tendered an important role in scrutinizing the management of en- shares. The government could encourage the house- terprises. Second, the capitalization would also help the holds to deposit their shares with the investment trusts banks to improve their weak balance sheets, which need as a sound method of diversification. recapitalization in any case. It is estimated here that This second approach has won widespread support in with 6 percent of the shares in the 500 largest enterpris- Poland and is viewed as politically superior to the direct es, commercial banks would receive a transfer that is the distribution of investment trust certificates, since it of- equivalent of about 10 percent of commercial bank gross fers more choice to households. On the negative side, assets. however, it is vastly more complicated and could in fact Privatization schemes that rely in large part on the greatly slow the privatization. With the voucher plan, free distribution of shares are sometimes said to be dis- unlike the plan to distribute investment trust shares, advantaged in that the government foregoes a large, po- there must be a valuation of individual companies as tential revenue source. While this point may be true well as a time-consuming public offering. In addition, when shares are distributed freely to workers or house- the vouchers could be complex to issue and process. holds, the revenue loss may not be incurred when public The free distribution of investment trust shares could or quasi-public institutions are capitalized, as in the case be completed within about one year, while the system re- of the pension funds and the commercial banks. The cap- lying on vouchers would probably take a couple of years italization of the pension system will reduce the require- longer. In either case, after this phase is completed, the ment for budgetary funding of the social security system government will retain roughly 35 percent ownership in by an equal magnitude. The capitalization of the com- the partially privatized companies. A second board of di- mercial banks will likewise reduce a future claim on the rectors would have to be formed upon the expiration of budgetary resources of the banks by anticipating the the one-year term for the initial board appointed by the need for a commercial bank re-capitalization. government. The second board would be elected by the The fourth tranche will consist of 20 percent of the shareholders for a three-year period. The shareholder shares of the enterprises, which will be distributed gen- groups created by the pension funds, the banks and the erally to the adult population of Poland (roughly 25 mil- investment trusts would presumably dominate this lion). This part of the share distribution will cause a loss board of directors and, thereby, firmly establish control of wealth for the state sector, since the distribution to over management. The board of directors could be elect- households will not be recouped by budgetary savings ed by cumulative voting (essentially, proportional repre- elsewhere. There are two prevailing models for how to sentation) to make sure that each of the major holders of distribute these shares. In one model, the shares of the shares places representatives on the supervisory board. enterprises would be distributed to several private in- Following the free distribution of shares, any number vestment trusts (which are closed-end mutual funds), of methods might be used to dispose of the remaining whose shares in turn would be freely distributed to the government holdings, including public offerings, private adult citizens of Poland. After the initial distribution of placements of shares and further free distributions. Of the shares, the investment trusts would be free actively course, if share sales are the predominant means of dis- to manage their portfolios. posing of the last tranche, the government will have to Each individual would receive one share in one of the undertake a careful valuation of each enterprise and pre- investment trusts, so that if there were, say, 10 trusts, pare the sale. In addition, the government may wish to each would have around 2.5 million subscribers. The in- encourage enterprises and investor groups to come for- vestments trusts would pass through dividends and oth- ward with privatization proposals in a decentralized er income to the shareholding public after deducting the manner, provided that each deal receives an adequate fund's expenses and fees. Each trust would be managed degree of scrutiny. In certain cases, the government may by a Polish entity but would contract with a foreign ad- wish to retain a minority holding and can look to West- visory firm to assist in the establishment of the trust, in ern European experience for an appropriate pattern of the active management of assets and in the administra- government equity positions in the corporate sector. tion of the dividend distribution. The French concept of selling a block of shares to a An alternative model suggests free distribution "stable core" of investors is an attractive technique for through vouchers. Individuals would receive vouchers disposing of the last tranche. In the case of Poland, the 247 government would entertain bids from domestic, for- zation. The case of Poland was used to illustrate the po- eign or mixed investor groups. It should be possible to litical, legal and logistical problems that lie in the way of establish a stable core for most of the 500 largest enter- privatization. A concrete plan for privatization in Poland prises and complete the privatization of these enterpris- was presented, aimed at demonstrating how these con- es within a three-year period. The stable core would straints can be met and how the main economic objec- eventually become the primary investor group and, be- tives can be achieved. cause it would be entrepreneurial in nature, would take Poland, and other countries of Eastern Europe, must a dominant role in supervising corporate management. devise strategies for privatization that ensure that the In time, the supervisory role of pension funds, invest- transformation of the ownership structure goes forward ments trusts and commercial banks, although impor- uninterrupted. The standard, case-by-case approach tant, would no longer be the main force monitoring and based on initial public offerings is prone to get bogged controlling management behavior. down for political, economic and logistical reasons. An alternative approach is needed. Central Features of the Illustrative Plan The rapid conversion of state enterprises into corpo- rate form and the distribution of tranches of shares to In concluding this section, what are regarded to be various groups in the population, including workers, the central features of the plan, and what are regarded as commercial banks, pension funds and mutual funds, are illustrative but not fundamental are stressed. Several key advocated here. This strategy differs substantially from steps are identified here: corporatization that will estab- the standard methods of privatization that have been lish the legal basis of the new economic system; a partial used in the West: the sale of shares in an initial public of- distribution of shares to workers and managers for polit- fering and private placements to investor groups. The ical and incentive reasons; and the distribution of some free distribution of shares helps to sidestep the difficult, shares to financial intermediaries such as banks and mu- costly and time-consuming process of enterprise valua- tual funds, which will have some early responsibility for tion, as well as the scarcity of financial capital in private appointing corporate boards. Once this process goes for- hands in Eastern Europe. More importantly, corporati- ward for the bulk of the largest few hundred enterprises, zation combined with the free distribution of shares can it is not vital what fraction of shares the govemment occur quickly. Rapid privatization is needed to combat holds-it could range from 30 percent to 50 percent of the inevitable social, political and economic problems the enterprise. In the latter case, however, the govern- associated with the lack of corporate governance. ment would likely be a mostly silent partner in the daily It is vital to establish effective governance quickly. management of the enterprise. The first, and urgent, task is to introduce a provisional The proposition that long-term management of the system of corporate governance that can monitor the enterprises will be enhanced if the government can sell management and prevent the managers and workers a significant block of each enterprise to a core buyer is from squandering the capital income and capital assets accepted. This process will take time. The risks of wait- of the firms before full privatization takes place. The sec- ing, however, will be significantly reduced if a large part ond, and long-term, task is to foster a structure of own- of the enterprise is already in private hands and if the ership in which the new private owners will be in a preliminary struggle over the form of ownership, corpo- strong position to manage their newly acquired assets. rate versus worker management, for example, is deci- sively settled in favor of a corporate structure. Notes 1. This paper originally appeared in Brookings Papers on Eco- Conclusion nomic Activity 2 (1990): 293-341. 2. For a discussion of the authors' preferred strategy for making The most daunting challenge facing the countries of the transition to a market economy in Eastern Europe, see Lipton and Eastern Europe today is the transfer of state property to Sachs (1990). For two other discussions of a comprehensive strategy private hands in a manner that is rapid, equitable and fis- for transition to a market economy in Eastem Europe, see Blue Ribbon Commission: Project Hungary (1990) and Komai (1990). cally sound. To achieve this goal in the Eastern Europe- 3. There are, of course, exceptions. Trenchant recent analyses an context is particularly difficult, because it requires may be found in Kornai (1990) and Milanovic (1990). For an excellent the complete redefinition of property rights and wealth- overview of the privatization experience throughout the world, see holding in the society and the creation anew of the basic World Bank (1988). institutions of a market financial system. This paper has 4. Cited in Milanovic (1990, p. 45). For evidence on the rise of in- stitutional investors in the United Kingdom and Japan, see Cosh, reviewed the enormous scope of the privatization task Hughes and Singh (1989). They report that in the United Kingdom, for and suggested principles that should govern the privati- example, financial institutions held 58.9 percent of the value of listed 248 securities in 1985, up from 44.8 percent in 1976 (table 6, p. 16). 20. The possibility that the workers might appoint a minority 5. Small-scale privatization is moving forward in Poland. Approx- portion of the board, as in the company law in many Western European imately 17,000 retail outlets had been privatized through September economies, is not being ruled out here. However, Poland's company 1990. It is still not clear, however, if the government will be able to law, which dates from the 1930s, does not have any such provision for achieve its target of privatizing roughly two-thirds of the retail estab- worker representation. lishments during 1990. 21. For a sophisticated theoretical discussion of this point, see 6. The available evidence in Poland suggests a surge of private Dreze (1989). Dreze summarizes his findings as follows (p. 114): sector activity during 1990, with high rates of return and the rapid es- In economies operating with uncertainty and incomplete insur- tablishment of new enterprises. According to the official data, approx- ance markets, it is natural to find capital hiring labor [rather than la- imately 360,000 new private enterprises had been established during bor hiring capitall, because efficient labor contracts in capitalist firms January-November 1990. However, the data are subject to many biases. are easier to draw and monitor than efficient equity contracts for labor- Many establishments are not registered in order to avoid taxes; other managed firms. establishments are simply shell organizations created to reduce taxes The point is that it is generally more efficient for the owners of cap- for other related businesses. ital to hire labor than for laborers to rent capital. 7. The authors' own views evolved during the policy debate in 22. See Blasi (1988, table C-2, pp.264-66), which reports the pro- 1990 in favor of a free distribution of shares, and their views have been portion of firms owned by internal employee funds (ESOPs, retirement heavily influenced by several of the contributions mentioned in the funds, savings funds, stock-purchase plans, etc.) in the case of Fortune text. 500 companies. There is not a single case in which an ESOP plan con- 8. As is mentioned later, the idea of a "stable core" was central to tains as much as 15 percent of the shares of a firm and only 7 firms of the French privatization process in the mid-1980s. Notably, Kornai the Fortune 500 for which an internal stock fund of any kind contains (1990, p. 91) independently stressed the importance of such a stable more than 20 percent of the shares. core in his recent book. 23. The value of this discount is to be capped, however, at less 9. These data are out of date and overstate the extent of state own- than the value of one year's average compensation of the workers in the ership in many cases, such as in France and the United Kingdom, firm. which have reduced the share in the 1980s. 24. The seminal contribution linking dispersed ownership to in- 10. In Poland, the social democratic faction within Solidarity ar- effective corporate governance is Berle and Means (1932). An enor- gues for a more gradual, less free market strategy of reform. mous debate has arisen concerning ways to solve the governance 11. According to official statistics, there were 231,000 private es- problem and the extent to which it is a problem. Some economists, tablishments in the industrial sector in 1988. These firms tend to be such as Demsetz (1983), suggest the problem is largely overcome in very small, with an average employment of two workers. Polish Au- practice through a combination of managerial compensation based on thorities estimate that the private industrial sector accounts for only stock prices and through an adequate size of share ownership by mi- about 5 percent of sectoral output, although this figure is almost surely nority shareholders. Empirical evidence tending to support the Berle understated by the high level of activity in the underground (and and Means hypothesis has recently been provided by Morck, Shliefer therefore under-reported) economy. and Vishny (1988). In particular, these authors show that an enterpris- 12. For an extensive discussion of the difficulties and trade-offs e's market valuation (measured by Tobin's Q) tends to be lower when involved in privatizing natural monopolies and other firms in non- management holds a very small share of the enterprise capital than competitive environments, see Caves (1990) and Vickers and Yarrow when it holds a moderate amount of enterprise capital. (1988). 25. The main problem is that takeover bidders usually gain very 13. Sixteen firms have been dropped from the top 500 firms ow- little from a hostile takeover and therefore often do not undertake the ing to a lack of adequate data on performance during the first half of effort even when efficiency considerations would recommend it (see 1990. The remaining 484 firms are used here as a proxy for the top 500. Grossman and Hart 1980). On the other hand, some takeovers may go 14. This section draws heavily upon the research of Moffatt forward even when they are not justified by efficiency, if the takeover (1990) of the Washington law firm of Hogan and Hartson. The authors results in a gain in wealth for the bidder, not as a result of a rise in ef- thank Joseph Bell for sharing this research material. ficiency, but by a transfer of wealth from some stakeholders in the tar- 15. While the focus is on the phase of reform Communism in Po- get firm. See Shleifer and Summers (1988). land after 1980 in contrast to the period of strict central planning be- 26. For discussions of the French concept and development of fore that time, it should be noted that the extent of centralization and the "stable core" (noyau stable), see Friedmann (1989a and b). decentralization waxed and waned throughout the post-war period. In 27. In the recent debate about the shortsightedness of American the early years after the war, workers councils exercised some measure and British firms, considerable admiration has been expressed for the of management rights. These rights were largely lost with the Stalinist German and Japanese patterns of corporate governance. The argument crackdown in the late 1940s, when the instruments of central planning has been summarized as follows: were strengthened. There was a brief thaw in 1956 (at the time of Khrushchev's attack on Stalin at the 20th Party Congress), in which Much recent criticism of the City of London by British industrial- workers councils were again invigorated. By 1958, however, they were ists is rooted in the belief that institutional investors are too willing once again demoted in real influence. to sell out to an opportunistic bidder without having due regard to 16. See the discussion on Komai's concept of the soft budget con- the longer term strategy of the incumbent management. Institu- straint in Komai (1990), as well as other references therein. tional fund managers, it is said, operate on a different time horizon 17. For an excellent discussion of the different forms of gover- from industrialists and are prone to behave as speculators rather nance of state-owned enterprises, see Milanovic (1990). than owners. Unlike the bankers who have played such an impor- 18. As reported by Moffatt (1990, p. 18) regarding the Amending tant role in corporate governance in West Germany and Japan, the Law on State Enterprises, March 9, 1990. insurance companies and pension funds which dominate the more 19. The average real wage in industry increased by 15 percent and equity-oriented markets of the English-speaking economies are re- 12 percent in 1988 and 1989, respectively. See Lipton and Sachs mote from the boards of companies in which they invest. The re- (1990). sulting pressure on management for short term performance, it is 249 argued, is inimical to capital investment and research and develop- 38. In essence, the income for part of the retirement payments ment. (Plender 1990) would come out of enterprise profits. These profits are now financing the accumulation of physical and monetary capital by the enterprises. The argument seems to have found a recent practical response as It is likely that as the enterprises are pressed to pay out part of their well. French commercial banks, for example, are now moving to emu- earnings to the pension funds, the retained earnings of the enterprises late the German pattern of bank ownership of industrial capital. (Econ- would fall, with a consequent fall in domestic investment spending. omist 1990.) 28. Sheard (1989) contains an excellent discussion of the role of References the commercial banks in corporate governance in Japan. He explains how "the main bank system in Japan substitutes for the 'missing' take- over market in Japan," or "to put it somewhat differently, the main Berle, Adolf A., and Gardiner C. Means. 1932. The Modern Corporation bank serves to internalize the market for corporate control." (p. 407) and Private Property. New York: MacMillan. Sheard stresses that corporate governance by a main bank economizes on scarce information, a fact of enormous practical relevance in East- Bhagat, S., A. Shleifer and R. Vishny. 1990. "The Aftermath of Hostile ern Europe today. Takeovers." Discussion Paper 87. London: LSE Financial Markets 29. The large banks were created by the dissolution in 1989 of the Group, June. state mono-bank. These banks are therefore still in the state sector and have little actual experience in corporate oversight or loan analysis. Blasi, Joseph R. 1988. Employee Ownership: Revolution or RipofR?. However, they will have to play a vital role in reconstruction in any Cambridge, Mass.: Ballinger Publishing Company. event and therefore their administrative capacity will have to be built up. The banks should receive shares of state enterprises, however, only when the banks themselves have a clear timetable for privatization and Blue Ribbon Commission: Project Hungary. 1990. Hungary: In Trans- have a demonstrated program for enhancing their capacities to engage formation to Freedom and Prosperity. Indianapolis, Ind.: Hudson in corporate oversight. Initially, the shares could be transferred to each Institute, April. bank as a trustee of the government's shares, with the ownership actu- ally shifting to the bank only upon privatization of the bank. Caves, Richard. 1990. "Lessons from Privatization in Britain: State En- 30. Some people have called for an even more active role for fi- terprise Behavior, Public Choice, and Corporate Governance." nancial institutions by calling for the creation of holding companies Journal of Economic Behavior and Organization 13:145-69, thatwould immediately become majority owners of the state enterpris- es. Advocates of the holding company approach believe that it is crucial Cosh, A.D., A. Hughes and A. Singh. 1989. "Openness, Innovation and to create a dominant investor, with a majority stake, at once and later Share Ownership: The Changing Stgructure of Financial Markets." arrange for the sale of shares that would transfer corporate control. Working Paper 74. Helsinki: World Institute for Development Eco- Here the view is that it would be dangerous to entrust the ownership nomics Research, October. of a large number of state enterprises to a single, untested financial in- stitution. Such an arrangement would, in all likelihood, exacerbate problems of market power and would impede restructuring and liqui- Demsetz, Harold. 1983. "The Structure of Ownership and the Theory dation. Moreover, there would be no clear mechanism under which the of the Firm." Journal of Law and Economics 26:375-90. holding companies would ultimately divest themselves of the enter- prises. As long as the holding company maintained its control, there Dreze, Jacques H. 1989. Labour Management Contracts and Capital would be a period of limbo during which time it would be difficult to Markets: A General Equilibrium Approach, Oxford: Basil Black- create the financial institutions needed for a market economy. well. Economist. 1990, August 4. 31. See, for example, "Independent Directors with Bite on the Board" (1990, p. 19), which reviews the article by Gilson and Kraakman Franks, J., and C. Mayer. 1990. "Capital Markets and Corporate Con- (1990). trol: A Stuy of France, Germany, and the U.K." Economic Policy 32. See Franks and Mayer (1990) for evidence that hostile take- 10:189-231. overs in the stock market play a larger role in the United Kingdom than in Germany in correcting "managerial failure." In Germany, a change Friedmann, Jacques. 1989a. "Sur l'experience de privatisation: et sur in management is not commonly associated with a change of owner- les noyaux stables." Commentaire 45:11-18. ship. 33. If, on the other hand, valuation is done in a mechanical way, without close approximation to true economic value, privatization will . 1989b. Lecture presented at the Israeli International Insti- inevitably generate large windfall gains and losses that can become fo- tute in Tel Aviv, September 14. cal points for political protest against the entire privatization. 34. For three analyses of the debacle of Chilean privatization in Gilson, Ronald, and Reinier Kraakman. 1990. "Reinventing the Out- the 1970s, see Luders (1990), Hansson (1990) and Yotopoulos (1989). side Director: and Agenda for Institutional Investors." Working Pa- 35. Luders (1990, p.3). per no. 66. John M. Olin Program in Law and Economics, Stanford 36. It seems that the most effective way to ensure that the vouch- Law School, Stanford, California. August. ers do not become near-monies is to limit their liquidity by issuing them in registered and non-tradable form. Grossman, Sanford J., and Oliver D. Hart. 1980. "Takeover Bids, The 37. For a discussion of managerial compensation schemes and Freerider Problem, and the Theory of the Corporation," Bell Jour- their effects on firm efficiency, see Murphy (1985). nal of Economics 11:42-64. 250 Hansson, Ardo H. 1990. "Capital Market Development and Privatiza- Murphy, Kevin J. 1985. "Corporate Performance and Managerial Re- tion." Unpublished paper. World Institute for Development Eco- muneration: An Empirical Analysis." Joumal of Accounting and nomics Research, Helsinki, July. Economics 7:11-42. Hinds, Manuel. 1990. "Issues in the Introduction of Market Forces in National Center for Employee Ownership. 1989. "The Employee Own- Eastern European Socialist Economies," Mimeo. World Bank, ership 100." May. Washington, D.C. March. Plender, John. 1990. "Malaise in Need of Long-term Remedy." Finan- Informacja Statstyczna [Statistical Information Monthly] (1989, 1990 cial Times, July 20. and various issues). Warsaw: Central Statistical Office. Rudolf, Stanislaw. 1988. "The Objective Nature of the Democratization "Independent Directors with Bite on the Board." 1990. Financial Process in the Workplace." ComparativeLaborLawJournal9:399- 12mes, September 4, p. 19. 431. Jensen, Michael C., and Kevin J. Murphy. 1990. "CEO Incentives-It's Sajko, Kresimir. 1987. "Enterprise Organization of Eastern European Not How Much You Pay, But How." Harvard Business Review Socialist Countries-A Creative Approach." Tulane Law Review 90:138-53. 61:1365-82. Kornai, Janos. 1990. The Road to a Free Economy, Shifting from a So- Sheard, Paul. 1989. "The Main Bank System and Corporate Monitoring cialist System: The Example of Hungary. New York: W.W. Norton and Control in Japan." Joumal of Economic Behavior and Organi- and Company. zation 11:399-422. Lewandowski, J., and J. Szomberg. 1989. "Property Reform as a Basis Shleifer, Andre, and Lawrence Summers. 1988. "Breach of Trust in for Social and Economic Reform." Communist Economies 3:257- Hostile Takeovers." In Alan J. Auerbach, ed., Corporate Takeover: 68. Causes and Consequences. Chicago: University of Chicago Press. Lipton, David, and Jeffrey Sachs. 1990. "Creating a Market Economy in Tomlinson, Alexander C. 1990. "Proposal for a Hungarian National In- Eastern Europe: The Case of Poland." BPEA 1:75-147. vestment Trust." Mimeo. Washington, D.C. Luders, Rolf J. 1990. "Chile's Massive State Owned Enterprise Divesti- Vickers, John, and Vincent Wright. 1989. The Politics of Privatisation ture Program, 1975-1990: Failures and Successes." World Bank in Westem Europe. London: Cass. Conference on Privatization and Ownership Changes in East and Central Europe. (Forthcoming.) Vickers, John, and George Yarrow. 1988. Privatization: An Economic Analysis. Cambridge, Mass.: MIT Press. Milanovic, Branko. 1989. Liberalization and Entrepreneurship: Dy- namics ofReform in Socialism and Capitalism. Armonk, N.Y.: M.E. Vuylstecke, Charles. 1988a. Techniques of Privatization of State- Sharpe, Inc. Owned Enterprises. Volume 1, Methods and Implementation. World Bank: Washington, D.C. 1990. "Privatization in Post-Communist Societies." Mimeo. World Bank. Washington, D.C. . 1988b. Techniques of Privatization of State-Owned Enter- prises. Volume 2. World Bank: Washington, D.C. Moffatt, Gregory. 1990. "Memorandum: Employee Participation in State-controlled Enterprises in Poland, Hungary, Yugoslavia, . 1988c. Techniques of Privatization of State-Owned Enter- Czechoslovakia and the Soviet Union." Unpublished. Hogan and prises. Volume 3. World Bank: Washington, D.C. Hartson, Washington, D.C. Yotopoulos, Pan A. 1989. "The (Rip)Tide of Privatization: Lessons from Morck, R., A. Shleifer and R.W. Vishny. 1988, "Management Ownership Chile."World Development 17:683-702. and Market Valuation: An Empirical Analysis." Joumal of Financial Economics 20:293-315. 251 I 21 Markets and Institutions in Large-Scale Privatization: An Approach to Economic and Social Transformation in Eastern Europe' Roman Frydman andAndrzej Rapaczynski This paper provides a systematic approach to the sources become depleted at a rate that may not be social- problem of privatization in Central and Eastern Europe- ly desirable. an economies. The approach is systematic in the sense of There are two standard ways of dealing with the prob- providing a theory of what privatization is supposed to lem of the commons: regulation; and the creation of accomplish in Central and Eastern Europe and of sup- property rights. In the first case, a communal decision is plying a means for evaluating the relative advantages made concerning the use of the common resources, and and disadvantages of different privatization strategies. this decision is then coercively enforced against those While the case of Poland is the focal point of the discus- who attempt to free-ride on the efforts of others. In the sion, the analysis is intended to be applicable to the re- second case, resources are assigned to the exclusive use gion as a whole. of individual agents who, because they have to pay all the costs and derive all the benefits from the use they make -Why Privatize? of the resources, have appropriate incentives to choose those uses that yield the greatest net benefit. In the first The first thing to understand about privatization in case, the social use of resources is made on the basis of Eastern Europe is that, in contrast to other countries political decisions; in the second, it relies on the maxi- and given the environment in these transitional econo- mization of individual interests, in conjunction with the mies, it does not entail a simple transfer of ownership market as a mechanism for allocating resources. from the state to private individuals. Rather, it is a pro- While all societies use political decisions to regulate cess by which the very institution of property, in the certain aspects of the economy (in particular those in sense in which lawyers and economists employ the term, which market mechanisms are vitiated by persistent is re-introduced into Central and Eastern European so- free-riding and externalities), the socialist systems of cieties. Central and Eastern Europe made practically all deci- At the core of every socioeconomic order is the prob- sions related to production2 through the political sys- lem of the efficient use of socially available resources. tem, with factory personnel playing the role of state Whenever the use of these resources is not restricted, functionaries. The name "command economy" conveys the so-called "problem of the commons" arises. Consider precisely this eschewing of market mechanisms, as well the case of a primitive society in which no one has exclu- as the fact that all property-related arrangements in gen- sive rights to land and all members of the community eral were replaced by an administrative system in which are free to use it for their own purposes. In this society, the state preferred to control the behavior of each agent every time a person invests time and energy to cultivate directly, rather than relying on the agent's own pursuit the land, he or she bears all the costs of producing the of self-interest. In this sense, the socialist economies of crop but can only derive a small part of its benefits; con- Central and Eastern Europe did not have any property versely, whenever any person removes something from system (including state and not just private property) the commons, he or she derives all the benefits of what governing their productive activities. It is not surpris- is removed but bears only a fraction of the costs of pro- ing, therefore, that in all Central and Eastern European ducing it. In a system of this kind, there is a systematic countries it is nearly impossible to answer the simple incentive to underproduce and overconsume, and the re- question of who owns what in the state enterprises: legal 253 determination of ownership was irrelevant under the old by state-appointed bureaucrats-the so-called nomen- system, which instead directly prescribed the conduct of klatura-who used to respond to other bureaucrats factory officials. higher up in the mammoth hierarchy of the planned The need to re-introduce the very institution of prop- economy. This hierarchy has by and large been disman- erty into the productive resources of Central and Eastern tled, and the enterprises are supposed to go it alone. European countries means that the structural reform of However, in the absence of any new external control over their economies cannot proceed primarily at the macro- management, the old nomenklatura people, instead of economic level. This realization, given the recent reform maximizing the enterprises' returns, are scrambling to efforts in a number of Central and Eastern European find the best deal for themselves. Some are trying to con- countries, is of great importance. vert the state enterprises into their own fiefdoms that The case of Poland is instructive. The first stage of the they can then convert into joint ventures with foreign economic reform there, known as the "Balcerowicz participants, whereby they will get a hefty pay-off and Plan," consisted of a series of macroeconomic measures, the foreigners will get the enterprise for a song. Others such as credit restrictions, wage restraints and the re- are attempting a home-grown "privatization" by which, duction of subsidies, designed to arrest the inflationary without any capital input or another legitimation, they pressures in the economy. The effects of this series of might end up as owners of the previously state compa- measures was in part predictable: first prices shot up- nies. Most managers, however, absent their traditional ward and then inflation slowed quite dramatically and bureaucratic support, are trying to maintain themselves prices remained relatively stable (although not as stable by forging a new alliance with their workers, to whom as had been hoped). Among other expected effects was a the reform has given an inordinate amount of power at fall in production and a rise in unemployment, although the enterprise level. The managers are willing to decapi- to an extent different than expected and perhaps for rea- talize their firms and neglect all measures that might re- sons that had not been foreseen. quire sacrifices by the work force, while maintaining The authors of the Balcerowicz Plan also expected salaries and employment at the highest possible levels. that the macroeconomic measures undertaken since In this situation, the freeing of prices and the emer- January 1990 would result not only in the elimination of gence of markets for the products manufactured by state the strong inflationary pressures evident at the end of enterprises are by themselves not sufficient to discipline 1989, but also in the creation of the basic conditions for the managers' behavior. Without the pressure of share- a market economy. The lifting of subsidies, together with holders who can cashier the management that does not other monetary measures, were expected to result in a produce high enough rates of return on a firm's invest- re-adjustment of prices. The possibility of assessing the ments, the only sanction the product market provides is costs and revenues of each enterprise more realistically bankruptcy. However, before reaching that point, a state was in turn supposed to provide proper incentives for enterprise can exist for a long time with its traditional management and to put state enterprises on a sound inefficiency by using up its sources of credit, cutting footing. Privatization would merely complete the pro- back on reinvestment or at best coasting along on the cess begun by the macroeconomic reform: when the borders of profitability. Further, when a bankruptcy does market could be used to determine the real viability of occur, no mechanism is available for restructuring to individual enterprises, they could then be valued and put things right. gradually sold off through a variety of well-known tech- Thus, reliance on the product market as a disciplinary niques. mechanism, with its threat of bankruptcy, is potentially It is a relatively safe proposition that, without some very dangerous. To be sure, some enterprises should be fairly dramatic steps at the microeconomic level, the closed. However, insofar as inefficient management hope of achieving a structural adjustment of the Polish might cause a large number of potentially viable enter- economy through the macroeconomic stabilization pro- prises to go bankrupt, a lot of resources might be wasted gram could not have been fulfilled. The reasons are re- and the economy might plummet, without there being lated precisely to the absence of an appropriate legal and any clear way of moving out of the depression. organizational structure at the enterprise level. The only way to remedy the crippling inefficiency of The structure of the Polish enterprises, as with those post-socialist state enterprises is to move as fast as pos- of the other countries in the region, is still largely a sible toward a genuine property regime. An immediate function of the old regime, and the behavior of their move that, although of limited scope is of considerable managers is determined by the conditions under which practical importance, is to introduce a new legal system they operate. Polish enterprises at this time are not even of genuine state property, even before a transfer of own- structured as joint-stock companies. They are governed ership from the state to private hands takes place. This 254 corporatization, or, as it is sometimes called, commer- try at prices the public saw as very low, popular opin- cialization, would entail an immediate transformation of ion might turn against the privatization program as a all state enterprises into joint-stock companies (with the whole. treasury being the sole shareholder) and the appoint- The people of Central and Eastern Europe also have ment of outside directors. Establishment of a genuine a somewhat ambivalent attitude toward the privatiza- form of state property, especially in a society with no sig- tion program and the concept of a market economy as nificant private sector to shape the behavior of state-ap- a whole. On the one hand, nearly everyone under- pointed directors, cannot result in a far-reaching stands that a move in the direction of capitalism is improvement in the functioning of the enterprises. It necessary and can be expected to yield, in the long can, however, provide some remedy for the worst cases run, significant improvements in living standards. On of mismanagement and abuse now common in the Pol- the other hand, it is also clear that, in the short run, ish economy.3 the move toward a market economy means further Nothing will remove the need for speedy privatiza- sacrifices in the form of potentially high rates of un- tion. It should not be seen as the last stage in the transi- employment, something the people of Central and tion from a centrally planned economy to capitalism, a Eastern Europe are not familiar with. stage during which the final touches are applied to an al- Finally, while a certain amount of differentiation in ready functioning system. To the contrary, insofar as levels of wealth based on risk-taking and superior privatization consists of a transfer of control into the business acumen is usually socially acceptable, an ex- hands of private shareholders who, in a mutually com- tremely unequal distribution of wealth that creates a petitive environment, are trying to maximize the returns permanent division between the haves and have-nots on their investments, it is an indispensable condition for is inherently destabilizing. It might be especially dan- efficient control of the performance of management. gerous in Central and Eastern Europe, where people This control over management is, in turn, the essence of have become accustomed to a certain amount of genuine restructuring, that is, of the transition from a equality and where the current political climate is not command economy to a true market order in which not very stable. Great attention must therefore be paid to only the value of particular products, but also of the en- choosing a strategy of privatization that does not ex- terprises producing them, is determined by the relation acerbate the anxieties of the population, but rather of supply and demand. Unless this process is completed, gives it some tangible stakes in the success of the un- the reform efforts in Poland and the other Central and dertaking. Eastern European countries will probably fail, and the (3) effective control over the management of priva- economic situation is likely to deteriorate even further. tized enterprises. The move away from bureaucratic control over the economy cannot mean a simple re- The General Principles of Privatization moval of all control mechanisms with respect to the functioning of the enterprises. This important point To have a chance of working, a privatization plan does not appear to be widely understood and was ap- must satisfy four main requirements: parently responsible for the belief that the removal of (1) speed. The privatization must be accomplished price controls and the emergence of markets would quickly, as must be evident from what has been said by themselves usher in a significantly more efficient already. If privatization, as is argued here, is the core system of production at the enterprise level. In fact, of the process by which state enterprises are restruc- decision-making in a complex modern economy, in- tured, the economic reform in Central and Eastern volving as it does extensive information to make even Europe cannot proceed without a radical transforma- the most trivial decisions and a complicated system tion in ownership. for allocating responsibilities, requires a whole pano- (2) social acceptability. As antiquated and inefficient ply of institutions that properly structure the incen- as Central and Eastern European industry is, it has tives of the actors involved and reduce the complexity been built at the cost of enormous sacrifices by the of real-world situations to a manageable number of general population over the last 45 years: the indus- relatively simple rules. In other words, when govern- trialization program initiated in the 1950s has been ments eliminate the control mechanisms of the com- pursued through drastic cuts in consumption. Al- mand economy, they need to replace them with though governments relaxed the austerity somewhat mechanisms that play the same role as the control in- in later years, the policy of investment in heavy indus- stitutions in a Western market economy, which have trial infrastructure has retarded the rise in living evolved over time and which are often taken for standards. If the governments were to sell this indus- granted. 255 Prime among these institutions is a system that in the region, the entry of foreign capital gives rise to provides incentives for managers of enterprises to special political problems and raises additional ques- maximize the interests of shareholders. In a properly tions of legitimation. The people of Central and East- competitive environment, this condition requires ern Europe very much want to catch up with the that production be structured to serve the interests of Western world, and they expect to be helped in their consumers. In developed capitalist societies, this task efforts to do so. At the same time, they are afraid that is accomplished, with varying degrees of effective- foreign capital will come to dominate their econo- ness, through a variety of institutions, such as take- mies and jeopardize their economic and political in- over mechanisms (supported by the whole legal, fi- terests. nancial and organizational infrastructure of the stock To succeed, a privatization plan for Central and market) or an elaborate banking system that super- Eastern Europe must, on the one hand, provide a vises company management (the system in Germany clear avenue for the entry of foreign capital and exper- and Japan). Without something to play a similar role tise, but, on the other hand, must place this entry in in the Central and Eastern European economies, a setting that makes it acceptable from the point of privatization could result in the extreme fragmenta- view of Central and Eastern Europeans' perceptions tion of holdings and no effective system of external of their own interests. supervision of enterprise managers. This situation in turn would undermine the whole meaning of privati- The Problem of Valuation zation, which, as argued here, involves not just a sim- ple change of ownership but rather a radical Privatization requires that a value be established for restructuring that transforms the incentive system of enterprises to be privatized. The state could convey the the economic agents at the enterprise level. title to some private party without assessing the enter- (4) assured access to foreign capital and expertise. prise's value, but such a naked transfer would not ac- Nearly everyone understands that the capital-starved complish anything, quite apart from the legitimacy and heavily indebted Central and Eastern European problems it might raise. The purpose of privatization is economies badly need Western funds to modernize not to transfer a title but to initiate a restructuring of en- their aging industrial infrastructure, introduce new terprises and a rationalization of the Central and Eastern technologies and so forth. It is equally clear that the European economies. For these objectives to be met, region also needs Western know-how and manage- someone must evaluate the potential of each enterprise ment expertise, without which it will not be able to to be privatized, that is, assess its relative value as com- use properly whatever Western financial aid is made pared with other possible investment opportunities. available or to bring its production up to the stan- Only in this way is it possible to decide where best to in- dards of the developed world. What is less often real- vest the limited resources available for upgrading the ized, however, is that Western expertise is most economy. needed in the effort to construct the general infra- Given the absence of a developed market economy in structure of a modern market economy, particularly Central and Eastern Europe, there are seemingly insu- the control institutions that will supervise manage- perable obstacles to the use of the traditional forms of ment at the enterprise level. To be effective, the entry enterprise valuation. The traditional methods essentially of foreign capital and expertise into Central and East- aim at an "objective," that is, inter-subjectively recog- ern Europe cannot take place through advisory and nized, assessment of the value of an enterprise. In the consulting services. The only way in which Western case of a publicly traded company, the inter-subjective financial institutions can play a truly creative role in element is self-evident, since it is equal to the price ob- the region is to base their entry on sound business tainable on the market at any given time. At the same principles, that is, to afford them the opportunity to time, individual valuations may differ from the market gain or lose by their activities. This condition in turn price.4 In the case of a privately held company or a fully means that the privatization program must create the state-owned company for which a market price is not conditions that make entry attractive from a business available, different individuals may have different sub- point of view and that the entry take place in a prop- jective assessments of the company's value that cannot erly competitive environment. be separated from the valuation. In the general context While serious Western participation in the con- of a market economy, however, it is possible to come up struction of the infrastructure of market economies with some approximations of the price that would be ar- in Central and Eastern Europe is needed for a suc- rived at by the market itself, since the agent doing the cessful shift from the bureaucratic command systems valuation will in part base the estimate on a number of 256 analogies to the methods commonly used by investors in volved in shoe manufacturing will have a lot of the market under similar conditions, such as price/earn- information about conditions in the shoe industry, ing ratios, the firm's performance over the last few years which normally the price system would disseminate to and the average prices of similar enterprises on the mar- the world at large but which remain private in the ab- ket. sence of a market economy. Third, when assets are large- An "objective" valuation does not make much sense ly illiquid (because there is no established market), under the conditions in Central and Eastern Europe, certain assets, quite apart from any informational dis- since without markets it is impossible to establish any parities among agents, will have very different values to reliable benchmarks against which to measure the value different people. A beer producer, for example, might put of the enterprises. Data from the period when the enter- a very high price on a ton of yeast, but the same yeast will prises functioned within the regime of a command econ- be useless to a shoemaker, who, if he cannot readily re- omy tell nearly nothing about a firm's present value. The sell it to someone who needs it, will not offer any price peculiarly Central and Eastern European institution of for it. Fourth, by the very fact that an agent has certain inter-firm credit, that is, the mutual indebtedness plans with respect to some assets for which he is bidding, among companies along the production process, intro- he is in possession of information that other people, duces a further element of uncertainty into the already such as accountants or outside consultants who will not clouded company books: around 40 percent of the book be involved in the exploitation of the assets in question, value of some of the companies being prepared for priva- are lacking. Again, this informational disparity is com- tization in Poland consists of outstanding liabilities from pounded by the absence of historical knowledge con- other enterprises, many of very long standing. Without cerning the most predictable uses of even the most evaluating the soundness of all the enterprises owing standard resources. Finally, every valuation involves money to the firm, it is impossible to predict what por- some skill or "hunch" or "tacit knowledge" that is not tion of these funds will ever be recovered. Similarly, quite arbitrary (some people, using their "hunches," do there is no basis for arriving at a valuation by analogy consistently better than others) but that cannot be ex- with other enterprises of the same type because there is plained in objective terms understandable to a third par- no capital market. With no reliable track record, it is im- ty. For example, a good venture capitalist does not base possible to make an informed guess about how a given his assessment of a firm's prospects exclusively on the firm would do in a free market economy. This situation value of its assets, price/earnings ratios and the like. is exacerbated by the fact that Poland's economic condi- Rather, the capitalist attaches the greatest importance to tion changes all the time, and it is impossible to predict the "feel" of management's skills, an assessment that, al- the state of the whole economy or even its segments a though intangible and often impossible to explain, may few months ahead. The state still sets the long-term in- be decisive for the venture's success. terest rate, about which some assumptions are necessary The predominance of subjectivity in valuation poses to calculate the present value of future streams of in- significant problems for a privatization program, prima- come, and its future course-even for the next few rily because even to the extent that valuation is not arbi- months-is largely unknown, even though relatively trary, it is impossible to explain or legitimize in objective small variations in the interest rate may very radically af- terms. As a consequence, any help or advice that might fect the estimated value of an enterprise. be gained from even the most reputable consulting or All these factors combined mean that, given conditions accounting firms is likely to be worthless. These advisors in Central and Eastern Europe, the "objective" elements must explain their conclusions to their principals, a task of valuation are de minimis and that the subjective ele- that is, under the circumstances, impossible. In turn, for ments, although present in any valuation, will dominate a privatization model to have a chance of working, it the assessments of companies during the transition to must avoid making the valuation the responsibility of capitalism. First, because there are no reliable bench- the state (which, to legitimize its decisions, must use marks for an objective assessment, the agents who per- outside consultants and try to arrive at an "objective" form the valuation must make more or less arbitrary valuation). Instead, the burden of valuation must be guesses about such things as the real value of fixed assets, placed on those parties that, like an ordinary investor in the appropriate price/earnings ratio or the interest rate a market economy, will bear the consequences of their for the next few years. These guesses will naturally differ decisions. Only these parties can rely on their subjective from person to person. Second, when the market does not estimates without having to explain their reasons.fi convey certain types of information through the pricing Even if the burden of valuation is placed on the party system, the variations in the information available to indi- that will bear the consequences of its decisions (deriving viduals increase dramatically. For example, someone in- extra gains from having arrived at a more precise valua- 257 tion than other parties and experiencing losses by mak- prises in, say, Great Britain were having to operate in ing errors greater than the others), the remaining high competition with other private companies, and their degree of uncertainty may make agents very reluctant to managerial systems (even if often less efficient than risk their assets on the basis of very imprecise guess- those of their private sector analogues) were basically a work. If, for example, an investor must decide whether to product of the surrounding capitalist business culture. If purchase some shares of Nowa Huta (a somewhat anti- some of these state enterprises were in the red, it was rel- quated steel mill in Poland) or a piece of real estate in atively easy to provide a measure of subsidies and reform Switzerland, the uncertainty in assessing the value of so as to bring them, within a relatively short period, into Nowa Huta will make the investor choose the safer in- profitability and to put them up for sale. The sale itself vestment in Switzerland. The exception might be if the was also rather easy: in a full market economy, which is price of Nowa Huta were discounted to such an extent as characterized by most of industry being in private hands, to make it competitive. However, the public might per- a developed stock market and the use by all enterprises ceive that price to be too low, and serious problems of le- of modern accounting methods, the process does not dif- gitimation could arise. fer very much from that of a private, closely held corpo- To deal with this assessment problem, the risk to or the ration "going public" by issuing shares to investors at uncertainty of the assessment of the agents must be re- large. It is enough for the state to hire the services of an duced. A way to accomplish both these objectives is to re- investment firm (or a consortium of such firms), which duce the universe of competing opportunities relative to underwrites the issue and sells the shares to the public. which the agent must evaluate the enterprise in question. This simple description of privatization in the West In the example just given, if the agent has no choice but suffices to show that it cannot serve as a model for priva- to invest some of his assets in one of the privatized enter- tization in Central and Eastern Europe, where capital prises in Poland, his valuation problem is reduced to an markets do not exist and the aim of the privatization is assessment of the value of Nowa Huta relative to other to introduce the structure of the market economy. The Polish enterprises and no longer to all other possible in- very idea that most enterprises can continue to be owned vestment opportunities. Not only is the problem of the by the state until they are profitable, in order to be sold size of the discount required by the uncertainty lessened, off afterwards, cannot be taken seriously: the state has but the uncertainty itself may be reduced as well. The rea- been unable to run these enterprises efficiently for 45 son is that large areas of uncertainty are found with all en- years and is not likely to change now, even if it is no terprises and may therefore be ignored for purposes of longer Communist. Had the state been able to take care internal comparisons. of the enterprises it owned, there would be no need to Special privatization vouchers accomplish precisely privatize them. this reduction in the universe of competing investment Even when abstracting from this problem, there is opportunities. Making the agents bid for the privatized still the valuation problem, which makes an ordinary companies using a specially restricted form of currency sale through a public offering (with or without an under- (vouchers), which cannot be used for any other purpose, writer) impractical. Even if this problem is also ignored rather than having them use money, eliminates all the and it is instead assumed that a market economy exists other options that compound the already serious valua- that makes such an undertaking meaningful, valuation tion problems. is costly and time-consuming. The valuation of over 6,000 state enterprises in Poland (or even of the 500 or The Sale Model so biggest firms) would take decades and undermine the entire privatization exercise.7 Privatization in Central and Eastern Europe cannot Even if all these problems were assumed away (by follow the sale models elaborated in recent years in such proposing, for example, that all state enterprises be auc- countries as Great Britain or France because of the im- tioned without any preliminary valuation in the hopes perative of speed in the privatization of most Central and that buyers would be able to make some decisions on the Eastern European industry and the difficulty of valua- basis of scarce information), there is still one more cru- tion. In these highly developed Western European coun- cial argument against large-scale privatization through tries, privatization had an entirely different focus than is public sales, especially in those countries in which a sta- the case in Poland: in those countries, the task was not bilization program has eliminated the accumulated restructuring the national economy but merely the sale overhang of the local currency. With Poland as an exam- of a few state-owned enterprises that were functioning in ple, the authors have calculated that, under the very op- a fundamentally market environment dominated by pri- timistic assumption that people are prepared to spend vate property. Prior to their privatization, state enter- 20-30 percent of all their savings to buy shares in the 258 privatized enterprises, the amount of money available to tion (up to 20 percent) of the shares of state enterprises purchase the state companies would be between 2.4 per- will be sold to employees at discount prices. Manage- cent and 3.6 percent of their book value. While this last ment is also frequently interested in buying a large number says relatively little about the "true" value of the enough block of shares to allow it (perhaps in coalition state enterprises, the discrepancy is staggering enough with the workers) to maintain control. Both these to make clear that, if purchases by foreigners are left groups, which have access to key information and are of- aside for the moment, privatization through sales would ten in a position to keep it secret, are therefore interest- amount to a giveaway that would increase the inequali- ed in having the enterprises valued as low as possible so ties by several scores. Given that the public does not see as to strike the best deal for themselves. The Ministry for the differences in wealth in Poland as a legitimate re- Ownership Transformations also does not want to set the ward for thrift or industriousness but rather as spoils price too high, since the enterprise might not sell. After distributed by the old regime to its loyalists, the give- months of preparation and the expenditure of hundreds away would cause the new authorities tremendous polit- of thousands of dollars, failure to sell would be regarded ical problems.8 (If, on the other hand, the government as proof of the ministry's inefficiency. The end result tried to avoid the accusation of giving away the national might be that if a foreign buyer is not found for a signif- wealth to the old nomenklatura by setting prices too icant block of shares, a substantial portion of the owner- high, it would risk not finding enough buyers, a situa- ship of the privatized enterprises will end up in the tion that might have the very harmful effect of lowering hands an alliance of workers and the old nomenklatura, the general level of confidence in the Polish economy.) with the rest being held by small shareholders (if they In Poland at least, sale of a large proportion of Polish are found) who would be too insignificant to interfere. industry to foreign investors would pose similar practi- All in all, the situation might not be very much different cal and political problems with as serious (if not more from the present one, in which there is no significant so) consequences. In any event, contrary to widespread outside control over the enterprises. fears of foreign domination among the Poles, foreign Preservation of the status quo could be avoided if a capital is by no means eager to invest heavily in the Pol- foreign investor were found to purchase a significant ish economy. It might be very hard to find foreign buyers block of the shares of each privatized enterprise. For this unless the prices offered are very low. That eventuality is reason, bringing in foreign capital would need to be an very unlikely, since the threat of being accused of giving integral part of a program of selling selected state enter- away the national wealth to foreigners was sufficient to prises for cash. However, as noted, it is by no means ob- cause the authors of the Polish privatization law to re- vious that foreign capital is willing to enter Poland strict foreign entry and require special permission for under the conditions being offered. If the foreign inves- any foreign interest to acquire more than 10 percent of tor can buy no more than 10 percent of the enterprise a privatized enterprise. This limitation, in conjunction (the portion that can be held without special dispensa- with the proposed preferential sale of up to 20 percent of tion under the privatization act), 20 percent of which en- the shares in an enterprise to its workers, makes invest- terprise is controlled by the employees and a significant ment in Poland even less attractive to foreign investors, portion of the rest by the nomenklatura management, who might be seriously hampered in their efforts to re- the investment will not seem promising. To make the in- structure the company (as that measure might often re- vestment more attractive to foreign investors, they must quire significant lay-offs) and put it on a sound business be given an opportunity to buy a large enough block of footing. shares to ensure some influence, at a price that is suffi- For these and similar reasons, it is clear to most peo- ciently low to outweigh the uncertainty. ple, at least in Poland, that large-scale privatization can- While the entry of foreign capital is of crucial impor- not be accomplished through sales. Nevertheless, the tance to the restructuring in Central and Eastern Eu- state is still trying to sell at least some state enterprises rope, it is not clear that investment is the most through one or another form of public offering, with the advantageous form. First, this type of foreign capital hope that a significant portion of the shares might be does not constitute a productive investment since it is sold to a foreign investor who would become an active goes into the state treasury. Not only may the money not participant in the restructuring efforts. be put to best use in this way, but also, given the very Beyond the technical difficulties, a number of politi- high external debt of most Central and Eastern Europe- cal issues that are involved in valuing enterprises to be an states, large inflows of the proceeds from privatiza- sold through a public offering of the British kind have to tion might create intensified pressure to increase the be considered. Note the following coincidence. Accord- debt repayment, in which case the money would flow ing to the Polish privatization law, a considerable por- right back out of the country. Second, the fundamental 259 rule of trading is not to sell before prices go up. In view workers who are employed in the most valuable factories of the general political and economic situation in the re- will receive an undeserved windfall, while many other cit- gion, the prices of the privatized enterprises are now izens, including those employed in state administration likely to be very low. If, however, after several years the or the private sector (potentially the most dynamic and general situation improves and the enterprises are re- entrepreneurial segment of the population) will be left structured, the same companies might be worth several with nothing. Quite apart from considerations of equity times more. If the condition of the Central and Eastern and the distribution of wealth, this proposal is very wrong European economies cannot be improved without sell- from an economic point of view. The principal interests of ing a very substantial portion of existing assets to for- the workers-employment and remuneration-do not eigners, perhaps the price is worth paying. However, if parallel at all the interests of the public (which wants the there are other ways to utilize foreign expertise that do best products at the lowest possible price) or the long- not involve mortgaging the country, the restructuring of term requirements of the economy as a whole (which re- the Central and Eastern European economies should quires long-term investment and growth in productivity). bear fruit for the people of the region. Moreover, if the individual workers were in some ways re- Finally, assume that, despite all the difficulties, it is stricted in their ability to sell their shares, the plan might possible to privatize a portion of state-owned enterprises impede the transfer of control to an outsider whose input through traditional forms of sale. Since only the most could discipline management behavior. Effective supervi- profitable enterprises are likely to find buyers, the com- sion would have to come from the workers themselves, panies selected for British-style privatizations will be the and there is very little evidence that such supervision, es- few most attractive plums of the Central and Eastern Eu- pecially in an economy in which worker-owned firms do ropean economies. What will happen to the rest-and, not really compete with firms organized along more cap- with them, to the economy as a whole? italist lines, can produce the desired results. In other words, as with some of the flawed sale plans, a program of Free Distribution Models free distribution to workers threatens to leave things much as they are now in countries such as Poland and to Given that the sale model of privatization is not via- impede economic restructuring.9 ble, the restructuring of the Central and Eastern Euro- Despite the dangers of worker giveaways, many Central pean economies must involve more unconventional and Eastern European countries face very considerable means. An important element of an unconventional political pressure to move in this direction. In Poland, this strategy is rapid privatization through a program of free pressure has resulted, despite initial resistance by the gov- distribution of the shares of the state-owned companies. emient, in a series of provisions in the privatization law There are basically three variables around which to ana- that allow workers to buy, at seriously discounted prices, lyze free distribution plans. The first concerns the bene- up to 20 percent of the shares of the companies in which ficiaries of the free distribution: to whom is the they are employed. Any privatization program in Poland ownership being given away? The second is the mode of will probably have to be reconciled with a significant ele- distribution: are the shares to be distributed directly or ment of worker participation.10 through some intermediaries, or will the beneficiaries If possible, the potential damage resulting from this receive some form of currency (vouchers) with which giveaway should be contained in some way. One method they can choose which shares to acquire? (The mode of is to give the workers a choice between their right to ac- distribution is decisive with respect to whether the ben- quire shares in their companies via this avenue or to re- eficiaries will have some say over which shares they re- ceive the benefits of other forms of free distribution to ceive.) The third variable is the role of the beneficiaries which they may be entitled (qua citizens, for example), in the governance of the privatized companies: are they rather than allowing them to have both opportunities. to become active or passive owners, and if they are to be Another is to restrict the shares acquired in this way to passive, who will supervise management? beneficial ownership, without the right to appoint direc- tors or otherwise actively participate in the governance Labor Ownership of the company. The most deeply flawed free distribution proposals en- Free Distribution to the General Public visage a giveaway or heavily subsidized sale of the shares of state-owned enterprises to their workers. From the While a giveaway of shares to workers poses both eq- point of view of social justice, free or subsidized distribu- uity and efficiency problems, a program of free distribu- tion to workers involves fundamental inequities, since the tion to the citizenry at large offers the promise of an 260 equitable and potentially efficient solution to the need gram distributes the state-owned assets in some for speedy privatization. The main advantages of a free demonstrably equal manner, the giveaway needs no spe- distribution program are twofold: it reduces the problem cial justification. This point is especially true for Central of valuation; and it eliminates the issues related to the and Eastern Europe, where it makes eminent sense to shortage of domestic capital or the reluctance of foreign say that all of society has paid a very heavy price for the investors to enter. construction of the national industry in the last 45 years. A giveaway program cannot eliminate the valuation The main problem with free distribution to the public problem entirely, however. It might be possible to exe- at large is assuring that the new owners, either directly cute a free transfer of ownership to some party without or through representatives, exercise sufficient control worrying about valuation. At the same time, as stressed, over management of the privatized enterprises. Other- if privatization is to result in a restructuring of the com- wise no change in the status quo can be accomplished, pany and not just a transfer of ownership, someone will and industry will not be restructured. Some free distri- have to value the enterprises to decide the best way to re- bution schemes can be immediately eliminated as unac- structure each company. Similarly, the choice of who ceptable because they do not meet this criterion. should be entitled to exercise the control associated with Suppose, for example, that the state wants to distribute ownership (and not just to enjoy its benefits) requires the shares of the privatized enterprises directly to the some method of discovering the party best able to super- population at large. Clearly, without determining the vise the restructuring. The best (indeed, probably the relative value of each company with respect to every oth- only) way to make this determination is to find the party er company to be privatized, any attempt to give differ- who puts the greatest value on the enterprises. ent portfolios of shares to different people must raise A free distribution program can, nevertheless, reduce serious questions about equity, since there is no way to the valuation problem because at least some of the most assure that one portfolio is worth as much as another. troublesome aspects can be left out of the accounting. An implicit principle underlying direct distribution of Prime among the aspects that can be ignored is the need shares to the population at large is that every person to express the value of the privatized companies in mon- should receive exactly the same portfolio. As such, each etary terms (a task that requires establishing their value person would have to receive the same number of shares relative to all other investment opportunities, such as (say, one) of each company to be privatized. In a country real estate in Switzerland or paper mills in Sweden). In- such as Poland, each company would have to issue at stead, the allocation problem may be dealt with even if least 35 million shares and would end up with 35 million the valuation is done in some form of restricted curren- shareholders. The coordination problems the sharehold- cy, such as vouchers that can be used only to purchase ers would face in their efforts to supervise the manage- shares in privatized companies (a limitation that nar- ment are so staggering that no supervision would be rows the universe of opportunities to the set of the com- possible, and management would be subject to no exter- panies to be privatized). nal control. Even more obviously, free distribution programs elim- Problems of this kind are endemic to many other free inate the problem of the shortage of capital: regardless of distribution schemes. Suppose that, given the difficul- whether the shares in the privatized companies are dis- ties just described, the state decides not to distribute the tributed directly to the beneficiaries or whether the bene- shares directly to the population but rather to issue spe- ficiaries are given vouchers with which to acquire them cial vouchers that are distributed in equal numbers to all (either at a pre-set price or at an auction), there is no dan- citizens who can then use them to "purchase" (either at ger of the state being unable to "sell." a pre-set price or at an auction of some kind) shares in Free distribution to the public at large also solves the privatized companies of their choice. Through such most of the legitimacy problems associated with selec- an indirect distribution, it is possible to avoid the ex- tive giveaways and the sale model. Selective giveaways treme outcome of companies with 35 million shares. are, by their very nature, suspect: there can be no satis- The problem is that the voucher "capital" of any individ- factory (sufficiently objective) answer to the question of ual purchaser is insufficient to allow him to acquire why one person is more deserving than another. more than a small fraction of any one enterprise." As a Similarly, given the valuation problem, inherent in result, the ownership of the privatized companies would sales in the Central and Eastern European context is the be extremely fragmented, and no effective mechanisms question of whether a given asset is being sold at a price for shareholder control would be possible (at least for that corresponds to its "real" value (whatever that some time). This problem needs to be solved if free dis- means). Inevitably accusations of covert selective give- tribution is to be a viable basis for a privatization pro- aways arise. If, on the other hand, the privatization pro- gram. 261 The Problem of Control prises, individuals in turn would hold the shares of the intermediary institutions themselves, so that they would The Core Investor be indirect beneficial owners of the assets held by the in- termediaries. The intermediaries may then perform var- One way to deal with the problem of control-one ious services on behalf of the small investors, from that has been proposed by some in connection with the pooling their resources for purposes of diversification to privatization discussion in Poland-is to combine free making all kinds of investment decisions on their behalf distribution to the population with the sale of a signifi- to, most importantly, exercising supervisory functions cant block of shares (around 10 percent or more) to a with respect to the management of the enterprises in "core investor" who would assume an active role in the which they are invested. restructuring and subsequent supervision of manage- A number of privatization proposals involve financial ment. intermediaries of one kind or another. Their respective Some of the difficulties related to the core investor advantages and disadvantages are analyzed here system- idea were raised in the analysis of the sale model. For a atically. Again, there are several variables around which number of reasons, the core investor must be a foreign- a taxonomy of intermediary institutions can be devised. er: very few Central and Eastern Europeans could afford Five are listed below before turning to the discussion of to buy a significant block of shares in a large company, concrete proposals: nor are there many people in the region with sufficient expertise to facilitate and supervise the introduction of (1) the relation between the intermediaries and the modern production and management techniques, and state, including the role of the state in their forma- only a foreign investor can facilitate contacts with poten- tion and later functioning, the conditions of entry tial foreign joint-venture partners or entry into foreign that determine the existence of the intermediaries markets. One possible problem is that the core investor and the nature of their regulation. would be the only significant investor in the company (2) the relation between the intermediaries and the despite having a relatively small block of shares. It might small investors for whom the intermediaries, perform be very hard to dislodge this investor given that corpo- a variety of services. Of particular importance here is rate raiders are not likely to appear for some time. In the way in which small investors acquire shares in the turn, unless other shareholders have large enough intermediaries and the degree of choice they have con- stakes in the same enterprise to exercise a restraining cerning such issues as entry and exit. force, the core investor would basically be subject to no (3) the relation between the intermediaries and the control. If the investor does not provide adequate super- companies in which they are invested, and particu- vision or exploits the company in favor of other (foreign) larly the way in which the intermediaries acquire the entities in which he or she has a higher share of owner- shares of the companies in which they invest and ship, no one would be in a position to do anything about their level of involvement in the supervision and con- it. The most troublesome aspect of the core investor trol of the management of these companies. idea, however, is that, as noted, for both political and (4) the relation between the intermediaries and other economic reasons it might be very difficult to get foreign kinds of financial institutions, especially banks and investors to enter the Polish market. A privatization plan investment banks. that relies on finding a core investor for every company (5) the relation between the intermediaries and for- to be privatized does not have a realistic chance of mov- eign and international financial institutions, in par- ing rapidly enough to restructure the Polish economy." ticular the role of these institutions in organizing and managing the intermediaries, as well assisting Cen- Financial Intermediaries tral and Eastern European participation. The other, more promising, way of resolving the con- The intermediaries and the state. Several privatiza- trol problem associated with free distribution is to sepa- tion proposals for Poland envisage the state setting up fi- rate ownership from control and to assign the latter to nancial intermediaries. According to some proposals, special intermediary institutions. These intermediaries the state would appoint the directors of the funds (in one are usually envisaged as holding companies or mutual of the plans, they would be nominated by the Ministry of funds, which would be the legal owners of the shares of Ownership Transformations and confirmed by the Par- the privatized enterprises, although they may also hold liament), and the intermediaries would receive the these shares in some looser form of trust accounts on be- shares of the privatized companies directly from the half of individual small investors. If the intermediaries state according to a set formula (one of the plans, for ex- are the legal owners of the shares of the privatized enter- 262 ample, proposes that each of five funds receive 4 percent choose the intermediaries in which to invest or would of the shares of each privatized enterprise). In many of automatically receive a certain number of shares in these plans, the government would strictly control the them. A corollary to this question is whether individuals number of intermediaries and would use them as an ex- could invest in the shares of the privatized enterprises clusive medium to distribute the free shares (so that, at directly or would have to acquire the shares of the inter- least initially, individuals could only own shares in the mediaries. intermediaries and not directly in the privatized enter- There is a certain appeal to restricting individual ben- prises). eficiaries of a free distribution to shares in the interme- Plans of this kind run very serious risks of making the diaries, rather than devising ways of allowing them to intermediaries new bureaucratic institutions that would acquire shares in the privatized companies themselves. be closely associated with the activities of the state and Similarly, there are some advantages to not giving the dependent on the state for their existence and function- beneficiaries, at least initially, the right to choose the in- ing. A foreseeable effect of such an arrangement would termediaries whose shares they are going to hold. The be a dramatic reduction in the funds' readiness to make reasons for these restrictions are always the same: ad- decisions on the basis of ordinary business principles, ministrative simplicity (which eliminates a lot of the their reluctance to take risks of any kind and, above all, transaction costs involved in other solutions) and the in- their security in the assurance that the state, closely formational barriers facing small investors, which would identified with the intermediaries in the minds of the limit their ability to avail themselves of the benefits from public, would have to come to their aid if either they or the choice were it available. Thus, for example, if the the companies in which they were heavily invested were shares of the privatized enterprises were somehow dis- in danger of going under. Another obvious risk is that tributed among the intermediaries (the constraints on the small number of intermediaries might encourage this possibility are discussed later), it might be adminis- their collusion and empire-building tendencies. tratively much simpler to give every citizen one share in The only way to make the intermediaries adhere to each intermediary, rather than worry about devising a genuinely business-oriented control functions with re- scheme that allowed individual beneficiaries to choose spect to the management of the companies under their among the intermediaries or to acquire shares in the supervision and of making their interest closer to that of privatized enterprises directly. This point is particularly their shareholders rather than the state bureaucracy is relevant since any such scheme would involve a costly to make them private, profit-driven institutions func- distribution of vouchers to all the individuals involved, tioning in an environment that forces them to compete who would then choose between using them to purchase for the favor of their shareholders. This approach does shares in the privatized enterprises or to acquire an in- not mean that the intermediaries should be unregulated; terest in the intermediaries. It might also be argued that, rather, they must be independent of the state. Among if given this choice, most individuals would not know the most important areas to be regulated is how the how to use it, and finding out might require more effort funds' management is compensated, so as to tie its inter- than the choice itself would be worth, given the small in- ests as much as possible to its performance on behalf of vestments involved. the funds' shareholders and the long-term interests of A decision to restrict consumer choice in these mat- the economy."3 Similarly, while the state may limit the ters also has its costs. Consumer choice might be quite number of intermediaries, its authorization to operate uninformed in a situation in which, as is the case in all should not be an administrative, bureaucratic decision; Central and Eastern European countries, there is no re- instead, it should auction a certain number of licenses to liable information concerning the relative value of the operate an intermediary to private parties satisfying cer- alternatives among which the consumer is supposed to tain basic conditions. choose. Nevertheless, consumer choice itself means the Clearly, state regulation of the intermediaries is a disappearance of an important factor that could poten- very broad subject related to the totality of their opera- tially provide a significant element of external control tion. Some aspects of this problem are discussed in this over the intermediaries' performance. If the intermedi- paper. However, the issue really requires special treat- aries had to compete for the vouchers to be received ment that touches on nearly the entire field of securities from the public at large, or if they had to "sell" their regulation. shares in some other way, their success would in part de- pend at least on the satisfaction of the people they are The relation to the small investor. The main question supposed to serve. If, on the other hand, the shares of the regarding the relation between the intermediaries and intermediaries are automatically distributed to the ben- small investors is whether the latter would be able to eficiaries, the dependence of the intermediaries on the 263 consumers must diminish and that on the state must in- are a fixture on the economic landscape of Central and crease, if for no other reason than because the decision Eastern Europe.'4 concerning their creation would have to be made not by the consumers of their services but by the state. (If the The relation to the privatized companies. There are consumer has no choice about which intermediaries to two main issues under this heading: how active a role "invest" in, their entry must be a state decision.) Some the intermediaries play in exercising the control func- forms of consumer approval could still play a role in the tion on behalf of the small investors; and how the inter- intermediaries' behavior; it would be possible, for exam- mediaries will acquire shares in the companies to be ple, to tie the compensation of managers to the price of privatized. the shares of the intermediaries. However, the fact that The two questions are related to some extent, since at least the initial position of the intermediaries would the mode of acquisition may determine the suitability of be independent of consumer choice might allow some the intermediaries to exercise the control function. In funds to entrench themselves, especially if their origin fact, the most important question concerning the alloca- in a state decision were to lead to their having leverage tion of shares in the privatized enterprises to the inter- over state assistance in adverse times. mediaries is how to assure that they take an early Another, and perhaps more significant, cost of disal- interest in examining the potential of the companies to lowing the beneficiaries of the free distribution to own be privatized and compete among themselves to spot the shares in the privatized companies directly is a long- most effective ways of restructuring them. They will do term concentration of all shareholding in very few so only if they can determine, at least in part, which hands. While some concentration of holdings is desir- companies they will acquire and if their blocks of shares able (since it allows for effective control by sharehold- will be sufficiently large to permit them to influence ers), the exclusion of the small investor makes a genuine management and prod the restructuring in the most securities market less likely to arise. If there were 10, 20 promising direction. or even 100 shareholders in the country, most of whose For this reason, unlike in the case of the allocation of holdings consisted of large blocks of shares that gave free shares to individual beneficiaries, a simple solution them privileged access to inside information, the volume involving the allocation of shares to the intermediaries of trading would certainly be very small; indeed, it would in some random or mechanical fashion has very little to be so small that all trading would likely be private, and recommend it. Although it could be argued that the ini- most holdings would likely remain illiquid. In the long tial mode of distribution does not matter since the inter- run, this situation might not only seriously impede the mediaries will later trade among themselves to reach the creation of stock markets, but it might also make the in- allocation they want, this approach is not appropriate. It termediaries very difficult to value and leave them forev- is desirable that the intermediaries begin researching er "closed," i.e., incapable of moving to a system in the companies to be privatized as soon as possible. The which they would have to redeem their shareholders' best way to insure their doing so is to force them to make shares on demand. important allocative decisions by a fixed date (as is the The "closed" nature of the funds is related to another case when the allocation takes place through an auction, possible means of consumer control over the manage- for example). Second, once an initial allocation is made, ment of the intermediaries: the possibility of exit. As some intermediaries immediately find themselves in a long as the fund remains closed, exit is possible only better position to evaluate some companies than others through the sale of the intermediary's shares to a third because they have access to inside information. At the party. The illiquid nature of the intermediary's assets same time, the informational asymmetries among the might make the market price of its shares an unreliable potential traders put some parties at a distinct disadvan- indicator of how it is doing. On the other hand, given tage. As a result, they may be reluctant to trade.'5 conditions in the region, illiquidity of the fund's assets is Third, trading presupposes a number of institutions entirely unavoidable, at least for the time being: it will be to facilitate it, such as stock markets with well-estab- a while before the privatized companies in which the in- lished prices for the shares to be traded and specialists termediaries are invested will have a reliable market who help effectuate certain transactions by underwrit- price. Still, it is worth considering "opening" the funds ing them. In their absence, trading may be very difficult. (i.e., obliging them to redeem their shares through the In Central and Eastern Europe, the funds may not have sale of a portion of their assets) in the future, since this at their disposal any large amounts of cash with which to measure would not only make their valuation more reli- trade, especially in conjunction with a privatization plan able but might also allow for the dissolution of some and that relies on free distribution. If most transactions have a move away from a system in which the intermediaries to take the form of barter arrangements, trading will be 264 very slow, and the needed reallocation of resources may being quite arbitrary and drawn out. If the second is cho- take a very long time (during which the Central and sen, it might be preferable to move directly to a different Eastern European economies will languish and the re- allocation system in which the shares of the privatized form movement might collapse under populist political companies are themselves distributed through an auc- pressures). Finally, back-door transactions among a tion mechanism."7 small number of agents are an open invitation to collu- In an auction, which is another way to allocate, the sion among the intermediaries. shares of the privatized companies to the intermediaries, Underwriting, the allocation mechanism by which the shares could be "sold" at a special auction during shares of privatized companies are initially assigned to which the intermediaries (as well as private individuals, the intermediaries, is the most complex and difficult if permitted) bid for them with the vouchers they receive part of any privatization plan involving free distribution. (depending on the plan) or from their own shareholders. To understand the difficulties, remember that free distri- The main difficulty with a privatization plan involv- bution is necessary because the traditional sales mecha- ing intermediaries is designing an auction that can ac- nism will not work under Central and Eastern European complish, without the use of real money, the following conditions. In the case of large issues of shares (as op- objectives: posed to isolated transactions among individuals), this mechanism normally involves an underwriter who as- (1) allocation of the shares of the privatized compa- sumes the risk of selling the whole issue for a small per- nies in such a way as to reflect the relative valuations centage of the sale price. Whatever the considerable by the intermediaries. advantages of this solution, it cannot be used in Central (2) acquisition by the agents of adequately large and Eastern Europe without some very significant blocks of shares in the privatized companies to permit changes. Even if the shares of the privatized companies them to implement their restructuring plans and en- were sold for money, the uncertainties of valuation and sure that every privatized company has at least one the absence of any track record of past transactions or of (and optimally more than one) large shareholder (so specialists who know the local conditions would make that none are left with no effective outside control). the underwriting extremely risky. Even assuming some- (3) revelation of some information about the relative one foolhardy enough could be found to engage in it, the valuation of the privatized enterprises by the bidders underwriting fees would be so staggering as to make the both to the other bidders during the auction (so that exercise impractical from the point of view of the state. they can adjust their bids to what they learn about the Nevertheless, there have been some interesting pro- emerging market prices) and to the general public. posals to use a modified underwriting mechanism in the (4) clearing of the share-for-vouchers market (at least context of a free (or near free) distribution of shares.'" approximately) so that bidders are not left with The problem is to find some way to underwrite with- worthless paper at the end. out forcing the underwriter to risk enormous amounts of money that could be used elsewhere. In other words, Devising an auction that satisfies the above condi- the idea is to change the incentives for the underwriter tions is a very complex task. However, although the au- in such a way that, instead of risking a large amount of thors are still working on the details of the design, the ordinary currency, he or she would be risking future task does not appear impractical. The main features are proceeds from the activity of underwriting itself. The un- described below (a more detailed discussion of the auc- derwriter might, for example, obtain free credit from the tion design is presented in appendix 2). state (in the form of vouchers or some other arrange- The enterprises to be privatized should be divided ment) and use it to pay for the shares of the issue he or into several groups, each comprising no more than 150- she would underwrite. If the underwriter makes a mis- 200 companies. Each of these groups should be auc- take and is unable to "sell" (again for vouchers) all the tioned off separately. After the first group is sold, there shares at the price he or she expected, he or she will be would be an intervening period during which the new punished by having the profits reduced to zero and per- shareholders would elect the boards of directors of the haps forfeiting some initial deposit. enterprises and the policy-makers and the public would The issue that should be considered in connection be able to assess the initial consequences of the chosen with the underwriting idea is that some other mecha- strategy of privatization. To avoid inter-temporal prob- nism would be needed to help choose the underwriters lems related to bidders' trying to apportion their vouch- themselves. This mechanism could be a multi-criteria ers among all the enterprises scheduled to be privatized, selection run by the state or could be an auction. If the the vouchers for each privatization phase would be is- first alternative is chosen, the process is likely to end up sued separately, and the validity of the vouchers would 265 expire at the end of a given phase. In effect, the vouchers such an exploration needs to be emphasized here. It is, would be a form of self-liquidating credit extended by the for example, likely that deeper analysis will yield a state to the public for the duration of each privatization number of restrictions on the relations of the phase. intermediaries and the companies in which they have The preliminary design for the auction of each group invested. Prime along these lines is putting some cap on of companies privatized at a given phase involves several the percentage of the shares an intermediary may own in stages in which the agents are asked to rebid several any one company (35 percent seems a good candidate in times. The actual sale would take place according to a light of the research on American companies) and rule that facilitates convergence and limits the agents' resolving a host of restrictions on self-dealing and strategic manipulation. A key feature of the proposed conflicts of interest. The analysis may also suggest auction is that the enterprises would not be auctioned mixing the intermediaries with other significant off seriatim; all would be bid for simultaneously at every investors (such as the core investor), whose incentives stage. may be structured in a different way, so as minimize the There would be two basic components of every auc- agency problems involving the intermediaries. tion, which would be handled separately. In the first, the The relation to other financial institutions. A corol- agents would be able to bid for, say, three large blocks of lary of the matter just discussed is the relation of the in- shares in each company-one of 20 percent and two of termediaries to other financial institutions, especially 15 percent. (This system is designed to ensure that each banks and investment banks. Again, the American model company will have some large shareholders who will ex- could be followed, in which fiduciary institutions such as ercise effective control.) In the second one, the remain- mutual funds are forbidden by law from engaging in the der of the shares would be sold through a different provision of other financial services, on the theory that procedure: the agents would apportion their remaining conflicts of interest may lead them to abuse their fidu- vouchers among the companies of their choice and ciary duties to their shareholders. At the other extreme, would receive a number of shares determined by the lev- once again, is the German model, in which banks super- el at which the prices would clear. vise companies on their own behalf as well as on behalf As to the level of the intermediaries' involvement, of their trust account clients, in the process also lending they would fulfill a number of roles, from relatively pas- money to the companies under their control (thus fur- sive to very active. At one end of the spectrum would be ther increasing their leverage). Somewhere in the mid- institutions modelled after the American ones, which, as dle is the standard investment house, which may a rule, must diversify very broadly and cannot hold a arrange for the financing of companies short of capital, large stake in any single company. As such, they cannot place directors on the boards of companies to guard the take an active part in supervising the management of the lenders' interests and provide expertise in the manage- companies in their portfolios. Intermediaries of this kind ment of companies in which the investment firm's cli- serve mostly to pool the assets of their shareholders for ents have a stake. the purposes of diversification and access to expert ad- An important part of the regulation of financial inter- vice. Institutions of this kind would not be suitable for mediaries in Central and Eastern Europe will be determi- the restructuring of the Central and Eastern European nation of the degree of their separation from the other economies. types of financial institutions. It is clear that the closer At the other end of the spectrum are the financial their relations are, the greater is the potential for conflict institutions prevalent in Germany and Japan, which of interest and the greater the need for other kinds of reg- have very significant stakes in the companies in which ulation (such as restrictions on self-dealing). On the other they have invested and play a very active role in hand, one of the greatest lacunae in the Central and East- supervising management. While the authors believe that em European economies is a modern banking system and the Central and Eastern European economies need other types of modem financial institutions. Moreover, precisely these kinds of institutions to supervise the setting up the infrastructure for an intermediary, which restructuring, there are clear agency problems that involves establishing local branches, opening up accounts must be guarded against when one shareholder controls for individual beneficiaries and similar tasks, is not unlike the whole corporation and can exploit the other owners. setting up the infrastructure for a standard bank. The ab- While it is beyond the scope of this paper to explore in sence of restrictions that prohibit intermediaries from detail the legal framework and incentive mechanisms providing banking services may greatly facilitate the es- that might be the most appropriate for reconciling the tablishment of a modern banking system and help priva- requirements of management supervision with the tized enterprises obtain financing for their operations. protection of minority shareholders, the importance of Similarly, the fact that there are no stringent restric- 266 tions on combining the fiduciary services of the interme- ther legitimize the role of foreigners in the privatization. diaries with merchant and investment banking services To facilitate the formation of joint ventures, internation- may allow the intermediaries to become a very flexible al banking institutions could aid the Polish partners source of funds for privatized enterprises: they could through grants and subsidies, while, should the Polish make the excess funds of some companies available to government require the intermediaries to make some others in their portfolios, in this way channelling do- payments (in the form of security deposits or license mestic savings into investments. They could also ar- fees), it could also offer special reductions in those pay- range for financing from outsiders. These services may ments in proportion to the degree of Polish participa- be particularly important if the intermediaries are linked tion. to Western financial institutions, since they could then serve as a bridge for foreign banking expertise and a win- Conclusion dow to Western sources of debt financing.18 The relation to foreign financial institutions. The im- Privatization in Central and Eastern Europe does not portance of the entry of foreign capital and expertise into entail a mere change in ownership. Rather, it involves a Central and Eastern Europe is reiterated here. Also em- complex social and economic transformation aimed at phasized is the reluctance of foreign capital to enter changing the way every company is run and every busi- some Central and Eastern European countries such as ness decision is made. Without a rapid transition to pri- Poland, and the reciprocal fear in those countries of for- vate ownership and a dismantling of the rigid eign domination. Thus, serious political problems Communist structure of industry, these economies will hinder the entry of foreign investors, at the same time continue to decline. that their entry as advisors is of little use. Moreover, privatization is not just a goal of the mar- The entry of foreign banking and investment banking ket-oriented economic reform. It should also apply to institutions is very appropriate to the setting up and the process whereby the companies are privatized so that running of the financial intermediaries under the priva- competition and market mechanisms can be used from tization program. First, foreign expertise may be badly the very beginning to decide who should be in charge of needed to help establish the infrastructure of a modern the restructuring. Otherwise, the state will decide what capitalist economy, with which the region has had prac- to do with each enterprise, and the economy will remain tically no experience. This infrastructure is particularly bogged down in the same bureaucratic quagmire that important, since it permits the emergence of market has paralyzed it for 45 years. forces and makes possible a chain reaction of growth and While private institutions should have exclusive re- development. Second, the foreigners would be entering sponsibility for the restructuring, the state can influence not primarily as buyers of Central and Eastern European the process by providing the general legal and regulatory industry (although part of their compensation might, framework. Among other things, the law would regulate and should, include stock options) but rather as manag- the incentive structure of the intermediary institutions, ers of the funds working on behalf of the local owners of prescribe anti-trust rules and tariffs and provide a safety the underlying assets. Since their success would directly net for the work force in transition. contribute to raising the value of the equity in local What distinguishes the approach espoused here from hands, their presence might be more easily accepted the others is its unique combination of several seeming- than it would be in other contexts. Moreover, if relations ly incompatible features. It combines widespread own- between the funds and their shareholders are structured ership and a measure of social justice with concentrated in such a way that the capital being managed by the in- control and economic efficiency. It avoids the initial termediaries is directly proportional to the number of lo- monetary valuation of enterprises, while immediately al- cal citizens who chose that fund over others, the degree locating the productive resources to the private agents of foreign influence over the running of the local econo- who value them most. It confers ownership on the citi- my could be seen as exactly proportional to the welcome zens of Central and Eastern European countries while of the local population. providing a link to foreign financial institutions capable While foreigners have a very important role to play in of assuring access to the world financial markets and ex- the intermediary institutions, for many reasons it might pertise in management supervision. Above all, it offers a be appropriate to implement their entry into joint ven- chance to move very fast toward a novel system of eco- tures with local partners. This approach would speed up nomic governance that could result in effective private the transfer of expertise to the local population and fur- control of Central and Eastern European industry. 267 Appendix 1. Valuation of State Enterprises in Table 21-1-1. Pro Forma Balance Sheet ofFirm X the Polish Economy: A Brief Case Study as ofDecember 31, 1989 (US$ millions) The valuation of a state enterprise, called X, is dis- Fixed assets 3.7 cussed here as a case example. The Polish government Inventories 1.8 plans to offer X for sale to the public in the fall of 1990. Cash and bank deposits 4.8 The government asked a prestigious British accounting Debtors 7.9 firm, called A here, to appraise the firm. The present dis- 18.2 cussion is based on that firm's preliminary appraisal. While the final appraisal may differ from the one pre- Bank loan (0.8) sented here, even this brief discussion of the preliminary Creditors and provisions (9.7) appraisal illustrates the fundamental problems inherent Welfare fund (0.6) in "objective" valuations performed by independent ex- (11.1) perts on behalf of Central and Eastern European govern- Net assets 7.9 ments. It will be shown that, of necessity, valuations of Central and Eastern European enterprises have to in- volve predominantly subjective judgments. Thus, to be ($9.7 million due to X's creditors). The prevalence of this really legitimate, they should be performed by agents credit and substantial uncertainty associated with its re- having a genuine stake in the future of the privatized en- payment are additional complications in the valuation. terprises. As mentioned in the third section, establishing the value Firm A used three standard methods of valuation: the of this credit requires evaluating the financial position of book value of the enterprise's assets; the price/earnings all the firm's debtors and creditors. In turn, however, ratio; and the discounted value of future profits or oper- those enterprises are likely to be burdened with their ating cash flows. own inter-firm credit arrangements, and so on. In sum, Firm X has assets in the form of buildings (partly un- the calculation does not provide any reliable idea of the der construction), machinery, tools, equipment and mo- company's real worth. tor vehicles, outstanding receivables and inventories. According to A, the second method used to value X- The pro forma balance sheet as of December 31, 1989 using the price/earnings ratio-is more applicable be- (see table 21-1-1) shows that the government valued the cause the main business of X is the export of labor ser- fixed assets at US$3.7 million, which represented 52 per- vices. As such, the firm's value depends primarily on the cent of the appraised net value of the assets. However, in quality of its management. However, as is typical for the absence of a real estate market and of secondary mar- Central and Eastern European economies, X derives a kets for machinery and equipment, the government had considerable portion of its earnings from exports to to value the assets in an apparently arbitrary manner. Comecon countries. The artificial nature of the histori- The accounting firm A reports that the government re- cal prices and exchange rates in these countries makes cently revalued the fixed assets of X by $2.2 million, that computation of meaningful figures for earnings highly is, by 146 percent. This unexplained revaluation ac- problematic. The computation is further complicated by counts for 31 percent of the value of the company's en- the prevalence of inter-firm credit. tire net assets. The valuation of the fixed assets is further The absence of securities markets in Poland forced A clouded by the inclusion of the unfinished Center for In- arbitrarily to fix the price/earnings ratio at 3. The result- ternational Cooperation, a "white elephant" X is building ing value of X was (at least) $17 million, with a potential in the provincial Polish town of Kielce with a view to error of $8 million. A simple check of the robustness of hosting international conferences. In the absence of this calculation shows that had the uncertain price/earn- markets, A reports having understandable difficulty val- ings ratio been set at 4 instead of 3, the potential valua- uing the center. Together, the government revaluation tion error would have been $16 million, or 94 percent. of the fixed assets of X and the value that A has calculated The third method, which is based on discounted fu- for the center amounted to $3 million, or 81 percent of ture cash flows, requires projecting cash flows and the the appraised value of X's fixed assets and 42 percent of proper discount rate. Here, A had to make several "key" its net asset value. assumptions. First, it assumed that the future rate of The balance sheet also shows that inter-firm credit profit of X would be the same as in 1989, the year before (the result of the notorious practice of inter-company the Balcerowicz Plan. Again, typically for the Central and lending) represented 43 percent of X's assets ($7.9 mil- Eastern European economies, it is very unlikely that the lion due from X's debtors) and 87 percent of X's liabilities pre-reform data were relevant for forecasting the future. 268 Second, A projected that sales to East Germany in 1990- transferred to new owners with the help of privatization 91 and 1992-94 would be between 30 percent and 120 vouchers. percent of the 1989 figure, whereas sales to the Soviet The 500 enterprises will be disposed of in several Union, Czechoslovakia and Hungary would amount to phases, with 150-200 to be privatized in the first one. An 30-75 percent of the 1989 sales. Third, it was assumed appropriate number of vouchers will be issued for each that the rate of inflation in all countries in which X op- phase, with one-half going to the public at large (each erated would be the same in future years. Because of the citizen receiving one voucher), one-third to the Social inherently unknowable nature of the environment in Security office (to capitalize the state pension fund) and which X would be operating in the future, A also had to one-sixth to a number of state banks. Thus, citizens will make a number of other equally arbitrary assumptions receive vouchers equivalent to 30 percent of the value of to "compute" future earnings. Finally, faced with the im- the privatized enterprises, the Social Security office 20 possible task of determining the discount rate, A used a percent and the banks 10 percent. figure of 25 percent supplied by the Polish Finance Min- The state will then invite the creation of a number of istry. intermediary institutions that will offer shares in them- The peculiar outcome of this procedure is that despite selves in exchange for vouchers. (Entry will be free, so the arbitrariness of all the numbers A used, the value of that any person or institution, foreign, domestic or X that the discounted future cash flows yielded was vir- mixed, will be able to create an intermediary as long as it tually identical to the one obtained with the price/earn- satisfies the minimum conditions specified in a special ings ratio method. The only way to explain this outcome law.) The vouchers the citizens receive will have to be is that the seemingly arbitrary assumptions were care- used to "purchase" shares in the intermediaries, with fully tailored to yield the reported coincidence of results. each person having a choice as to which intermediary to invest in. The banks will be free to use the intermediaries Appendix 2. The Polish Government's Large- or to trade on their own account. The Social Security of- Scale Privatization Plan: A Preliminary fice will have a choice of depositing all or some of its Analysis vouchers in the intermediaries or of creating one or more special pension funds of its own. On November 21, 1990, the government of Poland Once the vouchers are transferred to the intermediar- announced a comprehensive privatization plan, Rzecz- ies, the 150-200 companies privatized in the first phase pospolita, which it is proposing to implement in the near will be sold at a specially arranged auction, such as the future. One part of the plan, which deals with the priva- one described in this paper. The state will also deposit tization of some 500 largest enterprises in Poland (re- the 30 percent of the shares it owns with the intermedi- sponsible for about 70 percent of the country's industrial aries according to some predetermined formula.20 production), utilizes the framework presented in this pa- However, the state will not become an ordinary share- per, although it also contains some elements proposed holder in the intermediary institutions; instead, it will by others. While the plan may undergo a number of apportion its shares to the intermediaries and limit its modifications as a result of changes in government, it is role to the appointment of one director to the board of expected that its main features, as described below, will each intermediary. The intermediaries will be responsi- be implemented. ble for selling the state's shares in each privatized com- pany to other investors, either by private placement or in The Plan the open market. Once a certain percentage of the total state holdings administered by a given intermediary is The enterprises presently owned by the state will be sold, the state director will resign from the board of the divided into several categories. A basic dividing line will intermediary. separate the 500 largest enterprises from the "small and The transfer of ownership and control having been medium size" ones (some 2,000 industrial companies completed, the intermediaries and other shareholders and 3,500 others). The latter categories will be privatized for each company will appoint new directors, and the re- through sales or liquidation, while the former, with a structuring will begin. The new owners (with, it is few exceptions, are to be given away. hoped, the intermediaries in the dominant position) will Each of the 500 large enterprises will first be convert- be free to change the management of the privatized ed into a joint-stock company. Following this corporati- companies, to split them up (or perhaps combine them, zation, 10 percent of the shares will be given (free of as long as no anti-trust violations result), to sell a part of charge) to the workers,'9while the state will retain an- their holdings, to approve various joint-venture arrange- other 30 percent and the remaining 60 percent will be 269 ments between the privatized companies and other enti- efficient intermediaries. If budgetary considerations are ties (foreign or domestic), and so forth. The sale by the the sole reason the state is proposing the retention of intermediaries of the 30 percent of the shares they hold shares, there is very little to recommend the idea as op- on behalf of the state will, it is hoped, allow "core inves- posed to a simple taxation scheme. tors" to become involved in a number of the privatized However, the state's shares are designed to accom- companies, as well as (together with the transactions in- plish another objective: to provide for the possibility that volving the 10 percent of the shares held by individual a part of each company's assets might be sold to the pub- workers) to create a market for a sizable proportion of lic or to a "core investor." Both these possibilities have the shares of the privatized companies. This situation something to recommend them. While it has been ar- would increase the liquidity of the assets held by all the gued here that a core investor was unlikely to be found shareholders and allow for a market valuation of the in the initial stages of the privatization, it is conceivable privatized companies and intermediaries and for the that he or she could be brought into a number of the possibility that some of the funds may become "open." privatized enterprises within a few years. Moreover, while the core investor, if he or she were to be the only Analysis of the Plan big shareholder, could engage in exploitative behavior with respect to the rest of the shareholders and be very This plan is clearly a big step in the right direction. difficult to dislodge, that danger does not apply when the The choice of a free distribution model of privatization core investor has to share his or her authority over the for a large part of the industrial sector raises the hope company with a number of other large players, such as that the plan may be executed in a reasonable amount of the intermediaries or pension funds. To the contrary, his time without the worst kinds of political problems.21In- or her presence on the board of directors might have sig- sofar as the plan envisages the sale of a substantial por- nificant benefits. To begin with, the core investor may tion of the assets of the 500 largest companies, it have enterprise-specific expertise that the other large postpones that sale until the intermediaries are in shareholders (which are basically financial institutions) charge and the restructuring is underway. As was may lack. Even more important, the core investor, hav- stressed, the general level of confidence in the Polish ing a different type of interest from the other large economy might be much higher at that point, while the shareholders, may counterbalance their influence when presence of the intermediaries on the boards of directors it is not in the best interest of the company and their of the privatized companies might reduce the costs of own shareholders. This possibility could arise because monitoring for outside investors and thus reduce the in- the structure of the institutional investors, such as the vestors' ex ante risk and raise the prices of the shares. proposed intermediaries, always raises the possibility Moreover, the fact that the intermediaries, rather than that the incentives of their managers may not be not ful- the state, will be looking for buyers also insures (if the ly compatible with the interests of their shareholders. intermediaries' incentives are set correctly) that the tim- These "agency problems" may be limited, but they will ing and price will be better. probably never be completely eliminated. In fact, the ab- The plan's proposal to distribute vouchers to individ- sence of robust stock markets in Central and Eastern uals allows for the free entry of the intermediary institu- Europe will limit the range of devices to be used to eval- tions and removes the state from a significant role in uate the performance of fund managers and determine their creation. This process, together with the auction their compensation and may make the situation worse in scheme, which makes the intermediaries take an early this respect than it is in many other countries. The fact interest in particular enterprises, puts the intermediar- that the representatives of the institutional investors ies in a clearly competitive posture with respect to one might have to negotiate their strategies for the priva- another and raises the hope they will not degenerate into tized companies with a player whose incentives signifi- inert bureaucracies. cantly differ from theirs may protect the company (and In addition to the elements discussed in this paper, indirectly the shareholders of the intermediaries) from the government plan contains a number of other ele- possible over-reaching. ments that raise very interesting possibilities but also A core investor will not be found for all companies. In some significant dangers. The most controversial idea is these cases, the government's plan foresees that the in- that the state keep 30 percent of the shares of the priva- termediaries managing the state's shares would sell tized companies. State shares are undesirable in and of them, within some reasonable time, to the public. This themselves: as argued, this heavy involvement of the act might also have the beneficial long-term effect of state poses the risk that the intermediaries will be polit- strengthening the market for the shares of the privatized icized, as well as the specter of an eventual bail-out of in- companies so as to increase their liquidity and allow a 270 more reliable evaluation of the intermediaries' perfor- ers (the citizens) parallel the already mentioned mance. problems in devising an appropriate distribution scheme At the same timej it must be stressed that the exist- for the state's shares.22 ence of a significant state portion in the privatized com- Overall, while the requirement that the intermediar- panies raises serious difficulties. While the government's ies divest themselves over time of a certain proportion of plan proposes eliminating the state completely from the the assets under their management may be a good idea, management and boards of the privatized companies it is not clear that these assets ought to belong to the and to reduce its role in the intermediaries to one direc- state until they are sold. Moreover, even if the state were tor on their boards, this solution might not work. It is to retain those assets until they were sold, 30 percent is easy to underestimate the political pressure that can be much too large a portion for this purpose. Something on brought to bear on the state when it is seen as able to re- the order of 10-20 percent would be much less danger- solve a problem that a powerful constituency considers ous. very urgent. For example, if the state is known to have a The idea of capitalizing the state pension fund by giv- 30 percent ownership in enterprises that are laying off ing it 20 percent of the shares of the large enterprises has massive numbers of workers (as might happen in the ini- even more attractive aspects, although the dangers are tial stages of restructuring), the state could face irresist- also significant. While many budgetary outlays may be ible pressure to exercise a more active role. This better covered through taxation than through privatiza- eventuality is all the more likely if some other big share- tion, the "pay as you go" pension system is not only very holders, such as state banks and state pension funds, are burdensome on the budget, but it also constitutes a perceived (not without reason) as belonging to the state much larger burden for each generation that must pay and as easy subjects of governmental pressure. the full cost of the pensions for the retired population, State shares pose a host of technical problems that rather than allowing their own savings to grow with the might be very hard to resolve. First, it is not clear how economy before they are drawn on to pay for the old age its shares are to be distributed among the intermediaries pensions. (without which the role of the state would be even hard- At the same time, the objective of capitalizing the er to limit). On the one hand, it would perhaps be easiest pension fund is not fully in harmony with the purpose of to distribute the state's share in proportion to the rest of restructuring the Polish economy, which appears to be the assets of each intermediary, so that the state would the main objective of the privatization. The key issue, end up owning 33-1/3 percent of everything each inter- again, is the very size of the pension sector and its close mediary owned. However, the state's shares in any given association with the state. To understand the dangers, it company would then be divided among a large number is enough to note that if the Social Security office were of the intermediaries. This situation would make sale to to distribute its vouchers evenly among all the interme- a core investor that wanted to buy all the state's shares diaries, it would hold 33-1/3 percent to 40 percent of the or at least a large part of them very difficult, since the shares of each intermediary (depending on what the transaction costs of negotiating with many sellers might banks did with their vouchers). It would be by far the be very high. On the other hand, for reasons too complex biggest single shareholder and the dominant player in to enter into here, any other arrangement might create this sector of the economy. Alternatively, if the Social serious distortions in the behavior of the intermediaries. Security office did not deposit its vouchers with the in- The second technical problem relates to the sale of termediaries but were to invest them directly, and if, say, these assets by the intermediaries. Given the enormous 20 intermediaries were created, the state pension fund scope of the state's holdings (30 percent of the total), set- would be between 10 and 13 times larger than the aver- ting an arbitrary date (say, three to five years from the age intermediary. This situation, combined with the fact time of the initial auction) as the deadline for selling the that the pension fund is a state institution, means that state's assets (especially if the remaining 5,500 state the state, even without counting the 30 percent of the companies were also to be put up for sale during the shares owned by the treasury, would continue to domi- same period) would wreck absolute havoc with the secu- nate the economy. rities market. On the other hand, leaving the decision on It is not easy to design a satisfactory solution to this timing to the intermediaries could avoid a "fire sale" ef- problem. The obvious answer is to split the state pension fect, although some incentives would have to be created fund up into at least 20 separate funds so that its power for the intermediaries to sell the assets at all. The diffi- would be somewhat dispersed. However, a formal break- culty of designing an incentive scheme and its potential up may not be sufficient, for the state would still control for creating serious distortions in the behavior of the in- each pension fund separately. Under sufficiently strong termediaries as fiduciaries of their non-state sharehold- 271 political pressure it might simply coordinate the actions role as prime overseers of the restructuring. Viewed of all the funds to produce a desired outcome. Perhaps from this perspective, giving 30 percent of the privatized the only way to avoid this possibility is to "privatize" the enterprises to the state and 10 percent to the state banks social security fund, in the sense of giving some non- entails a potentially substantial cost in terms of the fu- state parties vested, legally protected property interests ture restructuring effort. in each pension fund. The best candidates for such ben- This argument is reinforced by considerations of le- eficiaries would be the insured themselves. Thus, for ex- gitimacy. The authors' back-of-the-envelope calcula- ample, individual accounts could be set up for each tions show that the value of the whole giveaway (for the insured person and the property rights over the account 500 largest companies) under the present government transferred from the state to the beneficiaries.23 plan, even when measured according to quite optimistic The next element of the government's plan-the dis- criteria,24 amounts to about $180 per person. While this tribution of 10 percent of the shares of the privatized en- amount is not insignificant given Poland's condition (es- terprises to the nine state banks-appears to be pecially considering that several members of each family unnecessary. In their present form the banks are simply will receive vouchers), it is still not very large and is like- state institutions, and there is no reason to believe they ly to be spread over several years. Given this fact, serious would be either independent of the state's political goals consideration should be given to eliminating the give- or contribute meaningfully to the restructuring. The ap- away to the banks and reducing the state's share to no parent purpose of their inclusion was a desire to capital- more than 20 percent. As an additional measure, work- ize them and make them stronger. However, they are ers receiving free shares of their enterprises should not themselves candidates for a privatization program and be allowed to receive the vouchers as well. A simple ar- thus a part of the problem rather than the solution. As gument that this approach is more than fair is that the was pointed out, the future of the Polish banking system average giveaway to a worker, under the government's lies in the further development of the intermediaries plan, will be more than eight times greater than the val- rather than in the existing state banking system. Howev- ue of the vouchers received by the average citi- er, the banks may have an important role to play in this zen25[Thus, giving the workers a choice between the two process, since they are the natural domestic partners of forms of giveaway should not lower the program's legit- the foreign organizers of the intermediaries. The state imacy among them ( in many cases they get a very good banks possess an extensive infrastructure, including lo- deal26), while it might make more assets available to the cal branches, contacts within the world of the state en- citizenry at large. terprises, and local expertise. As such, they may have Finally, the government's plan is not clear about the quite enough to offer to the new financial institutions. way in which the remaining 5,500 companies will be Giving them additional capital in the form of the vouch- privatized. This paper discussed in some detail why the ers (the value of which may be as high as $2.5 billion) ap- prospect of selling a large number of companies quickly pears unnecessary. is not realistic. Quite apart from this problem, however, More generally speaking, it must be realized that the it is important to note that, for many of the reasons men- assets to be distributed are limited. The main goals of the tioned above, as well as because some of the smaller privatization effort must be kept firmly in mind in companies are the most dynamic and potentially the eas- choosing the beneficiaries. Among the top priorities is to iest to turn around quickly, a substantial portion of these make sure that the corporate governance structure that enterprises should be included in the giveaway scheme. arises out of the privatization program be efficient and that a sufficient level of popular support be secured for a Appendix 3. Money and the Question of program that, while it is clearly in the long-run interest Voucher Denomination of the country, may inflict substantial costs on society in the short run. One issue that often arises in conjunction with pro- With respect to the first objective, care must be taken posals to use vouchers as part of a free or subsidized dis- that the intermediaries receive enough vouchers that tribution program is whether to combine the use of their formation is truly attractive and that they can vouchers with the use of money in the privatization and reach sufficient critical mass to bid successfully for the whether vouchers distributed to the population or the fi- large blocks of shares offered during the first stage of the nancial intermediaries should have some monetary de- auction described in the paper and envisaged in the gov- nomination. This appendix explains why those emnment's plan. Only then will they be able to play their approaches are inappropriate. 272 Combining Vouchers with the Use of Money the long run) offers a steady flow of revenue instead of a one-time injection to cover the budgetary expansion This approach means that the price of the shares to be that in later years would have to financed from other sold is expressed in regular currency. As a result, some of sources. the valuation problems involved in the ordinary sale of Second, the people who have money would not be the privatized companies would hamper the privatiza- able to use it to acquire greater than average shares in tion program. A further problem is that the vouchers the privatized enterprises. This argument would have would have to have money denominations, a require- some force if it could be assumed that the people in Cen- ment that raises additional issues. tral and Eastern Europe who have savings are the more The amount of the vouchers to be issued is a serious entrepreneurial element who should not be excluded problem. Insofar as the state is concerned, the vouchers from acquiring a greater share. This reasoning does not, are like money: the state must accept them as payment however, appear sound. To begin with, the distribution for equity in lieu of local currency at face value. The total of savings in Central and Eastern Europe is more likely value of the vouchers issued must therefore have some to reflect past political connections than a genuine en- relation to the total value of the enterprises to be priva- trepreneurial spirit. Further, the sale of state enterprises tized. The task of valuation is then even more daunting: would soak up savings that might be needed for other in- not only must each enterprise be valued separately, but vestments (such as the opening of small businesses). In also the value of all of them combined must be estimated addition, if trade in vouchers without any denomination at once. is allowed, the people who want to invest more than Even though the vouchers would be denominated in their allotment would be able to buy more vouchers on a certain amount, their market value (either explicit if the open market. Finally, a good privatization proposal trading is allowed or illicit if it is not) is certain to be would envisage that the shares of the privatized enter- much less than that amount. Moreover, the discount is prises (or at least those of the intermediaries) would be likely to be very significant, since not only is the use of traded on the stock exchange soon after the initial priva- vouchers restricted to long-term investment but also the tization auction. Those who want to invest their savings "real" (i.e., market) value of the shares bought for them in long-term investments would have the opportunity to is very uncertain. The psychological effect might be that do so, and the process (more orderly and extended in people would feel cheated: the government would be time) would contribute to the establishment of genuine telling them they are getting, say, a 1,000 zloty subsidy, market prices for the privatized enterprises. (It should while they see the voucher as worth, say, 300. The result- also be noted that the price of the vouchers or the shares ing political liability might exceed the gain from the free acquired for them would be higher if the vouchers were distribution. the only means of payment at the initial auction. In this Given that the (real or hypothetical) market value of case, people at large would have a greater sense of having the vouchers would be much less than the nominal val- obtained something valuable, a perception that would ue, the use of the vouchers in an auction would only con- raise confidence in the privatization as well as in the fuse the budding price system. Nominally, shares would economy as a whole.)27 be bought for, say, 1,000 zloty, while their market value outside the auction would be much less. Notes In the face these issues, what would be gained by combining the use of money with the vouchers? Two rea- 1. The firstversion of the authors' privatization proposal, suggest- sons can be put forward. First, the budget would be hurt ing the institutional setting analyzed here, appeared under the title if the state could not sell at least a part of the enterprises "Privatization in Poland: A New Proposal" in June 1990. A revision of for cash. This argument is a serious one. However, the that paper was published in Polish in Res publica in September 1990 under the title "Sprywatyzowac Prywatyzacje: Nowa Propozycja state has alternative means of raising funds, and even if Przemian Wlasnosciowych w Polsce" [On Privatizing Privatization: A these means were not otherwise more appropriate, the New Proposal of Ownership Transformation in Poland]. effect of the sale of enterprises on the privatization The authors are grateful to Ned Phelps for his comments and inter- would be enough to make them so. In any case, the alter- est in the ideas presented here from the very beginning of this project. native ways of raising revenue are more appropriate. A They also thank Professor Bronislaw Geremek, Minister Jacek Kuron, Drs. Marcin Krol and Aleksander Smolar, and Messrs. Lejb Fogelman, consumption tax (such as a value-added tax), for exam- Damian Kalbarczyk and Henryk Wujec for their early encouraging re- ple, would increase the savings rate rather than absorb- actions. In addition, they thank the following people for discussions ing the savings available for investment. A corporate tax, and comments: Professors Bruce Ackerman, William Baumol, Bernard while perhaps not practical in the short run (and per- S. Black, John C. Coffee, Marek Dabrowski, Owen Fiss, Harvey J. Gold- haps, since it involves double taxation, not a good idea in schmid, Jeffrey Gordon, Irena Grosfeld, Stanislaw Gomulka, Henry Hansman, Cezary Jozefiak, Barbara Katz, Alan Klevorick, Grzegorz 273 Kolodko, Michael Montias, Joel Owen, Mark J. Roe, Susan Rose-Acker- 17. This road has been chosen by the Polish government, as de- man, Roberta Romano, Jacek Rostowski, Jeffrey Sachs, Alan Schwartz, scribed in appendix 2. Ferdinando Targetti, William Vickrey, Stanislaw Wellisz and Charles 18. The latter may be especially useful, given that the limitations Wilson and Messrs. Andrew Berg, Ian Hume, Grzegorz ledrzejczak, on the entry of foreign capital may restrict the availability of foreign Stefan Kawalec and Jacek Kwasniewski. equity financing. In many situations, especially applicable to the Cen- Roman Frydman also gratefully acknowledges the grant for this tral and Eastern European context, in which share prices may be de- project from the C.V. Starr Center for Applied Economics at New York pressed for some time, debt financing has many advantages over University. equity. 2. Agricultural production in Poland was the most significant ex- 19. The privatization law that the workers receive up to 20 percent ception. of the shares of the privatized enterprises at half price, not to exceed 3. Thanks are owed to Professor Jeffrey Sachs for alerting the au- the value equal to their last year's wages. The present plan abandons thors to the importance of corporatization. the idea of selling the shares at half price and the limitation relative to 4. One of the reasons for this difference may be that an investor last year's income because of valuation difficulties. may be prepared to pay a premium for a certain block of shares that 20. Beyond specifying that the state will not be able to hold shares would yield control of the company, Another may be that an individual in the privatized companies directly and will have to use the interme- assessment of the company's future differs from that implied in the diaries, the government's plan does not specify the formula to be used market price. to distribute the state's holdings. Presumably, vouchers may be used in 5. To show the difficulties and the arbitrariness of the valuation of this connection as well. Central and Eastern European enterprises, appendix 1 presents the 21. The first phase of the program is to be completed in 1991, with case of an enterprise valued by a prestigious British accounting firm over half the assets of nearly all large companies being in private hands for the Polish government in connection with the privatization pro- by the end of 1993. gram there. 22. The problems alluded to here arise from the fact that the inter- 6. Katz and Owen (1990) make a similar point. mediaries' interest in the state's shares is significantly different from 7. A look at the number of enterprises re-privatized in Britain dur- their interest in the shares of the non-state parties. (The difference in ing the 10 years of the Thatcher administration gives an idea of how the interests of the state and the other parties is a similar situation: the time-consuming valuation and sales are. state wants to get the best price for its assets, while the other parties 8. Many of the problems listed here would cease to be serious if so want to maximize their long-term returns.) With respect to the non- many vouchers were issued that most transactions took place in state parties, the intermediaries stand to make money from the man- vouchers anyway. If so, there would seem to be no reason to use mon- agement fee, while, with respect to the state's shares, they can expect ey-denominated vouchers at all, especially since they appear to offer no some percentage from their sale. Since the returns themselves and the advantages. strategies for their maximization are likely to vary in several respects, 9. The interests of the workers are important and do need to be the managers of the intermediaries may be tempted to exploit some of protected by institutional arrangements. The appropriate mecha- their fiduciaries (by, for example, using some of the resources under nisms, however, are trade unions and governmental regulation of em- their management to beef up the prices of others). ployment conditions, rather than worker ownership. 23. The mechanism for accomplishing this system requires sepa- 10. Appendix 2 discusses the Polish government's recent privatiza- rate treatment that so far has been absent from the government's pro- tion proposal. posals. 11. Appendix 3 discusses the problems involved in attempts to 24. A price/earnings ratio of 5 is assumed here, a figure that is combine the use of vouchers with ordinary sales and the question of quite high for the Polish economy. There is also some question wheth- whether privatization vouchers should have monetary denominations. er the earnings (based on the First six months of 1990, when profits 12. Core investors may still be helpful for many companies. How- may have been depressed but when wages were frozen) are reliable. ever, it would probably be more appropriate to bring them in at a later 25. The factor of 8 does not depend on the value of the assets in- stage, when the other important players are already present on the volved; it simply follows from the relative percentages in the giveaway company boards and the price of entry is likely to be higher. and the proportion of beneficiaries in each class. 13. The design of the compensation structure and of the control 26. It should be kept in mind that the average wage at the time structure of the intermediaries themselves is one of the most complex when the data used here were collected was about $80 per month. and important tasks of any privatization proposal that envisages a sig- 27. Another important question is whether it is good policy to con- nificant role for the intermediaries. sume all the capital reserves people in Poland are holding in order to 14. Formation of the stock markets may also be furthered by either feed the state treasury, instead of utilizing them for other badly needed forcing the intermediaries, or giving them an incentive, to divest investments. themselves of a part of their holdings through public offerings. Appen- dix 2 discusses this possibility in the context of the Polish govern- ment's privatization proposal. 15. The authors are indebted to Bulent Gutelkin and Gavin Wil- References son's Memorandum of August 10, 1990, to Messrs. Krzysztof Lis and Stefan Kawalec of the Polish Ministry of Finance for this point. Katz, Barbara, and Joel Owen. 1990. "A 'Big Mac' Approach to 7Dena- 16. Two advisors to the Polish Ministry of Finance, Professors tionalization."Comparative Economic Studies XXXII (3):82-92. Stanislaw Gomulka and Stanislaw Wellisz, provided some interesting ideas on this matter. 274 22 A Note on the Privatization of Socialized Enterprises in Poland' Manuel Hinds The government of Poland is designing a strategy for The Government's Proposed Approach the massive privatization of socialized enterprises. The proposed strategy would transfer ownership of these en- The Nature of the Privatization Scheme terprises to the private sector through both sales and transfers without payment. Preparation of this plan sug- The government's proposed approach to privatization gests the government has concluded that large-scale involves, as the initial step, its reassertion of ownership privatization is needed. The fact that the proposed plan rights over the capital of the socialized enterprises. Once includes transfers without payment suggests further it does so, it would transfer 20 percent of the shares of that the government realizes this approach is the only each enterprise to five newly created holding companies. one to rapid privatization. Subsequently, these holding companies would transfer This paper assumes that the government recognizes their shares without payment to the population. Because these points. It also assumes that the enterprise reform the shares would be immediately tradable, they would would include not only privatization but other actions encourage the emergence of a capital market. Moreover, such as the breaking up of monopolies, enforcement of as a result of the transfer of the shares to the population, financial contracts and resolution of the issues of com- the enterprises would become private institutions. pensation of the previous owners of identifiable property. Simultaneously, the government would sell another This paper focuses on mechanisms for privatization 10 percent of the shares of each company to the banking once these general issues have been resolved.2 The next system and another 10 percent to the workers of each section briefly discusses the privatization scheme the enterprise. The latter would most likely be sold on pref- government is now considering, including its long-term erential terms. The govemment would retain the re- objectives. The following two sections present an analy- maining 60 percent for subsequent sale, to take place sis of the scheme, the first section focusing on the requi- after the prices of the shares have been firmly estab- sites of privatization, the second on the extent to which lished in the capital markets. the proposed program meets those requisites. The final According to the plan, the shares would have unequal section suggests some modifications that could improve voting rights. Those transferred to the holding compa- the effectiveness of the proposed program. nies, banks and workers, although representing only 40 The paper does not present a complete analysis of the percent of the claims on the capital of the enterprises, issues of privatization. Its purpose is to contribute to the would carry majority voting power. Thus, although the ongoing discussion by suggesting ways to carry out a government would retain most of the shares, it would ef- rapid privatization that results in the development of fectively be transferring control of the enterprises in part capital markets and improvement in enterprise manage- to the private sector (the privatized holdings and the ment. 275 workers) and in part to the banks (which belong to the porations when their ownership is too dispersed. In this government).3 case, each shareholder has such a small sum at stake that The proposed approach has several advantages. First, it is not attractive for him or her to spend time and mon- the central government would be able to transfer control ey trying to control the managers. As a result, the man- over the enterprises to the other sectors relatively quick- agers' power goes unchecked, to be used to appropriate ly. Second, although the government would be surren- the rents of capital through huge wages and benefits ex- dering enterprise control rapidly, it would eventually actly as happens with workers in labor-managed enter- cash in on the benefits of the expected improvement in prises. Enterprise efficiency suffers, and the prospects management because it would at some point be selling for growth are diminished. its 60 percent of the shares. Third, the plan gives the In market economies, the problem of mismanage- workers of the enterprises an incentive to accept the de- ment is solved by take-overs. Entrepreneurs who believe mise of labor management. they can manage an enterprise more efficiently than the At the same time, the proposed plan embodies certain current managers can buy enough shares to control the features that would jeopardize achievement of the objec- enterprise. They then fire the managers and appoint tives of the privatization. Those objectives are discussed their own. Take-overs are also used to remove inefficient below, while the problematic features of the plan are an- owners and allow quick reallocation of the capital. In alyzed in the fourth section. many cases, the managers buy the enterprise and, as owners, impose discipline on the firm. Because of these The Long-term Objectives mechanisms, the unchecked power of managers does not long survive in market economies. Poland's privatization has two main, related purpos- es: to improve the management of enterprises; and to The Requisites of a Privatization Scheme disperse economic power among a large number of agents. The relationship between the two objectives is To be successful, a privatization scheme must meet complex, however. The dispersion of economic power is certain requisites, as discussed below: facilitation of the desirable because it directly benefits income distribution role of entrepreneurs; speed of privatization; equity; and and political diversity. In addition, up to a point, it rein- other oft-cited requisites such paying the government forces the objective of improving management, given for the shares it transfers. that excessive concentration of economic power leads to inefficiency. On the other hand, excessive dispersion of Facilitation of the Role of Entrepreneurs ownership can also lead to inefficiency. Without a con- trolling shareholder, management is free to do whatever It is clear that to achieve the objective of efficilency in it wants with the enterprise, and the result can be chaos. an economy with dispersed ownership of capital, it is necessary to allow entrepreneurs to emerge and to bid Therefore, a delicate balance between these two objec- for control of the enterprises when they see a possibility tives is needed. of improving their management. Thus, a privatization As the experience in Poland painfully demonstrates, plan should make it relatively easy for potential entre- efficiency requires that enterprises be managed by some- preneurs to put together the resources to take over con- one who will defend the interests of capital, that is, trol of enterprises to improve their management both at someone who will strive to maintain and increase the the time of the privatization and later. As will be dis- enterprises' capital through economic management of cussed, the govemment's proposed plan makes this pro- its resources and judicious investment. In the absence of cess very difficult, if not impossible, at least in the initial an owner, it cannot be expected that somebody else will years. play that role since they can benefit more by taking away from the owner's income than by defending the owner's Speed of Privatization interests. This point is recognized in economic theory as the principal-agent problem, and history has proven its In economies where the private sector is already pre- validity again and again. Managers not subject to the dis- dominant, as in the United Kingdom, the speed of priva- cipline of owners tend to mismanage the enterprises. tization is important but not crucial. The reason is that They are not natural advocates for capital. They have the benefits of privatization, although important, are their own interests, which, if left alone, they will pursue marginal relative to the functioning of the economy as a rather than those of capital. whole. In Poland, in contrast, speed is crucial because The principal-agent problem does not pertain only to inefficient enterprise management is causing rigidities public sector enterprises. It also occurs with private cor- that are hampering stabilization and economic growth. 276 In addition, the socialized enterprises are perceived to be Shares should be sold. There are two main arguments a government responsibility, and their failure to spur for selling enterprises rather than transferring their economic growth is blamed on the government. ownership without payment. The first is that, on equity There is also an informal perception, even if denied grounds, the government should be compensated for the formally, that the government is financially responsible assets it is transferring. The other is that people will not for the enterprises, which play on that perception by ask- appreciate the assets if they do not pay for them. ing for subsidies and privileges. This behavior is evident Neither argument is compelling. As to the assertion nowadays. Enterprises have refused to carry out the re- that the government should be compensated for assets forms needed to improve their efficiency. (In a market being taken away from it, the government is just an in- economy, a liquidity squeeze comparable to that im- termediary that owns the socialized enterprises on be- posed by the government under the stabilization pro- half of the citizenry. Compensating the government for gram would have forced them to do so.) Instead, they the value of the assets makes sense only when they are have been pursuing tactics that permit them to avoid changing in thesuing hopescs the t g ernmentwill them ttransferred to a sub-set of the population. In this case, changing In the hopes the government will bail them however, the payment would not be compensation to the out if they fail for lack of liquidity. For example, at the government, an abstract entity that is only a representa- time of this writing the food industry was threatening a tive of the citizenry. Rather, those acquiring the assets shortage of supplies if the government did not bail it out. would be compensating the other citizens by paying the The government needs to resolve this situation as of be assets to the overnment by ayin the soon as possible, whence the premium on speedy priva- price of the assets to the government. If, as in Poland, tization-the faster it takes place, the better. (Later it the assets are being transferred to all the citizens, who will be shown that the proposed government plan will have been the ultimate owners all along, then this com- not achieve the necessary speed.) pensation is unnecessary. The second argument-that people do not appreciate Equity the value of the assets if they do not pay for them-pre- supposes that the people benefitting from the rents of Another requisite of privatization is that it be fair. the capital have paid for it. Such is not the case. The im- That is, if shares are transferred without payment, they portant point is that associating the benefits of owner- should be given equally to all citizens. If shares are sold, ship with the burden of management creates a strong the buyers should pay the market price. The proposed incentive for the owners to devote energy to manage- government plan achieves equity through transfers ment. This relationship does not exist in the current sys- without payment to the population but may face serious tem. equity problems in the sale of 20 percent of the shares to Furthermore, the statement that people do not appre- the banks and workers. ciate what they have not paid for is only a half-truth. It is only true where the giving away is a ongoing feature of Other Oft-cited Requisites society. It is not true in the case of a once-and-for-all transfer. Receiving shares without payment does not de- Other features are frequently mentioned as necessary tractfr. their apea itheuecpient koey wil be for privatization. Foremost among them is that the state tract from their appeal if the recipients know they will be be compensated for the transfer of the assets. Another is quite valuable afterwards because no more shares will be that to appreciate the assets, people need to pay for given away once the process is finished.5 A clear example them. Still another is that enterprises should be in mint of this point is that most people appreciate what they in- condition to be privatized. That is, the enterprises herit. should be made more efficient prior to transferring their It is likely that some people will not appreciate the ownership. shares. However, many others will recognize immediate- These requisites are derived mostly from the model of ly that getting the shares is a once-in-a-lifetime opportu- privatization developed in the United Kingdom in the nity. They will want to buy whatever the first group last decade. This paper maintains that what worked in wants to sell. From an economic point of view, it is good the United Kingdom may not do so in Poland because that the second group acquire the shares, since they will the circumstances there are completely different. In fact, devote their efforts to making their shares profitable. not only are these requisites unnecessary for Poland, but Some people may not appreciate the value of the meeting them could be damaging because they conflict shares at the outset because they do not understand with more fundamental requisites, such as speed in im- what the shares are. For social reasons, it is desirable proving management practices. The following para- that the process be designed in such a way that people graphs explain the rationale for this assertion. have time to come to appreciate the value of the shares. 277 (The fourth section, on transition, gives some ideas on is the major owner in the economy, it will easily bureau- how to achieve this objective.) cratize the whole process. The presence of foreign advisors will not change these Enterprises should be improved before transferring facts. Poland should not lose sight of the fact that it will their ownership. As noted, another commonly men- face competition from highly efficient Western corpora- tioned requisite of privatization schemes is that the en- tions that are not controlled by foreign advisors but by terprises be put in mint condition before privatizing down-to-earth entrepreneurs. No public sector corpora- them. This requisite embodies two assumptions. One is tion in the world is a competitive threat in the interna- that the government intends to sell the shares, an as- tional markets. No country dominated by public sector sumption that is not always true.6 The other is that the enterprises is a major factor in the international mar- government may get a higher price after it invests in im- kets. proving the enterprise than it would otherwise. This as- Even in predominantly market-oriented economies, sumption is highly debatable. such as the countries in the Organisation for Economic Nobody would be interested in acquiring, even at zero Co-operation and Development (OECD), public sector or a negative price, an enterprise that is functioning un- corporations have shown an inability to operate without der a system that ensures continuous losses and pre- huge injections of capital from the state, which, in the cludes sound management. Thus, nobody would be eyes of the European Common Market (EEC) in the re- interested in buying shares in enterprises that are con- cent controversy over Renault, are more accurately de- trolled by their labor force or unruly trade unions or that scribed as subsidies.7 Moreover, the Bank's experience in are subject to arbitrary government regulations. The restructuring public sector enterprises all over the world government does need to eliminate these problems prior is quite discouraging. There is no reason to believe the to privatization. However, once this condition is met, it experience in Poland will be different. is very likely the value of the enterprises would decline rather than increase should the government attempt to How Well Does the Government's Plan Meet Its improve them. Objectives and the Requisites of Privatization? The experience of the 1970s is highly relevant in this respect. The Polish government invested heavily in the This section looks at how well Poland's proposed modernization of the socialized enterprises, borrowing privatization scheme would meet both the long-term ob- abroad to finance the process. Because the investments jectives of the effort and the requisites for successful did not generate the cash flow needed to service the re- privatization (speed and equity). lated debt, the international financial position of the country deteriorated. Huge losses have accumulated in The Long-term Objectives the banking system (both valuation losses related to the stock of the external debt and collection losses). The gov- It is doubtful the proposed plan would improve man- ernment should not repeat this experience. agement and decentralize economic power as much as is It can be argued that the stabilization program has needed. The main reason is that the initial distribution corrected the disincentives, ensuring that investments of voting power in the enterprises would tend to concen- will be carried out efficiently. However, the incentives trate management power in the five holding companies prevailing in the labor-managed enterprises still do not and a limited number of banks. Assuming that the shares encourage efficient investment. As is becoming obvious the government keeps have no voting power and that the in the current stabilization effort, socialized enterprises voting power for all the shares involved in the first 40 do not react to market signals in the same way that cap- percent the government transfers is equal, the holding italist firms do because they do not have an advocate for companies would control 50 percent of the votes and the capital in their ranks. Despite the current liquidity banks and workers 25 percent each. Moreover, as dis- squeeze, socialized enterprises have, as noted, failed to cussed below, the resulting ownership structure would take steps to improve their efficiency. Furthermore, even not be conducive to improving efficiency for several rea- if the government reasserts its ownership of the enter- sons: the potential for bureaucratization of enterprise prises' capital and passes control over to holding compa- management; the excessive power the managers of the nies, the investments are not likely to be efficient unless holding companies would have; the potential for collu- the private sector carries most of them out. The reason sion among the holding companies; the creation of con- is that the public sector lacks the flexibility and motiva- flicts of interest within the banks; the creation of tion of the private entrepreneur. In addition, if the state obstacles to the emergence of entrepreneurs who would 278 try to improve management; deterrence of foreign in- them. The same would apply to bank managers. This vestment from the EEC; and return of investment deci- practice would negate the objective of creating a flexible sion-making to the government. capital market. Potential for bureaucratization of enterprise man- Collusion of the holding companies. There would be agement. The operation of enterprises is an extraneous a natural incentive for the managers of the holding com- activity for banks, while the power of shareholding work- panies to collude with each other to exercise monopoly ers or people buying shares from them would be too dis- power in the different markets of their enterprises. It is perse to influence their management. Therefore, most very easy to conspire against competition when five peo- likely the five holding companies would end up running ple control all the markets. the enterprises. Given the number of enterprises to be privatized, Creation of conflicts of interests within the banks. each holding company would find itself having to man- There would also be strong incentives for the managers age thousands of them.8 The logistics of this task are of the holding companies to collude with those of the such that the holding companies would become gigantic banks and for banks to use their financial power to main- institutions that most assuredly would become bureau- tain and expand their empires. This problem is extreme- cratized. It would soon become difficult to distinguish ly grave because, on top of all the negative effects of the holding companies from the former branch minis- having a small group of people controlling the economy, tries. The only difference would be (presumably) that the this situation would create a serious conflict of interest holding companies would be responsible for enterprises for the banks. engaged in a range of activities, unlike the situation with In the proposed scheme, the banks would control the branch ministries.9 some 25 percent of the voting power and would have a substantial amount of money at stake in the enterprises. Excessive power to the managers ofthe holding com- In this situation they would tend to give preference in panies. The power of the managers of the holding com- their lending and pricing decisions to their own compa- panies would go unchecked because the ownership of nies and to be lenient regarding the creditworthiness of these companies would be quite dispersed. Gaining con- those enterprises. Such practices are a clear recipe for fi- trol over an important share of any of the holding com- nancial crisis. panies would require an extremely large amount of The example of the Federal Republic of Germany is money.1° frequently cited as proof that linkage between enterpris- As a result, it would be extremely difficult for poten- es and banks does not necessarily lead to bad credit deci- tial entrepreneurs in the private sector to acquire suffi- sions. This example ignores the fact that banking cient voting power in these companies to have a say in supervision there has been strong and has emphasized their management. The population would most likely re- an arm's length relationship between the banks and en- act to the situation with apathy, which would ensure terprises. Furthermore, the banking regulators have dis- that nobody would challenge the power of the holding couraged the banks from increasing their current stakes companies' managers. in the enterprises and from participating in new enter- In addition, the accumulation of economic power prises (the banks acquired their current positions in the would become the first priority of the managers of the enterprises as a result of debt-to-equity swaps in the holding companies. Even if their normal objective were 1930s). to maximize their profits, it is very probable that the On the other hand, there are numerous examples of concentration of economic power would become the real countries where an ownership linkage between banks objective. Moreover, as noted, the objectives of the man- and enterprises has resulted in financial crises. The list agers are different from those of the owners. Command- includes Spain, Chile, Yugoslavia, Mexico and even Po- ing one-fifth of the industrial power in the country is land itself (where public sector banks have traditionally more attractive than any compensation the dispersed financed public sector enterprises to comply with im- owners could give the managers as a reward for efficien- plicit or explicit government wishes)."2 cy. The managers' power would be so great that they Even where banks do not lend money to the enter- could appropriate such compensation. In addition, they prises they own, they will also not be likely to invest in would enjoy enormous political power. Under such cir- shares, because shares are more risky than the assets cumstances, to keep and enhance their power the man- banks normally carry, their value is volatile and both the agers would tend to hoard shares rather than trade cash flow and income they produce are unpredictable. 279 Obstacles to the emergence of new entrepreneurs companies they own. The enormous power of the man- able to improve the management of privatized firms. agers of the holding companies would endanger eco- Under the proposed plan, it would be even more difficult nomic freedom, especially if they collude, as they are for potential entrepreneurs to acquire control of individ- likely to do. In this suffocating environment, new en- ual enterprises than to take over the holding companies. trepeneurs would fail to emerge. Even if they were to Even if they were able to buy all the shares in the hands emerge, however, they would not have command over of the workers, they would still not be able to exercise the savings necessary for investment. In short, if Poland control over the enterprises because the banks and hold- adopts this privatization scheme, it will simply exchange ing companies together would have 75 percent of the one variety of monopolistic control over the economy for votes. another, equally damaging, one. A vivid example of the problems that the scheme Problems for foreign investment from the EEC. The could cause is provided by Chile in the late 1970s and unequal voting rights on the shares pose other prob- early 1980s. In those years, the government sold many lems. For one, they detract from the transparency of the industrial enterprises and banks. Since only a few people capital markets. For another, the inequality would likely were willing and able to buy the enterprises, five groups become an obstacle to economic integration with the that also owned the banks ended up owning most of EEC. According to The Economist (1990), " ...the EEC them enterprises. These groups used the banks to fi- will almost certainly restrict, or even outlaw, multi-tier nance their own enterprises. The result was the largest share structure .... ''3 The inequality would also hinder financial crisis Chile had ever suffered and one of the investments by EEC investors in Polish enterprises. For worst, relative to the size of the country, in any place in these same reasons, Sweden will probably eliminate its the world in the recent past. Chile is not the example to system of unequal voting powers. It makes no sense for follow." Poland to adopt a system that other countries are drop- ping for good reasons. Speed Giving the power to allocate investment back to the Given the problems analyzed above, the proposed government. Most importantly, the proposed system privatization scheme would not create the flexibility the would run counter to one of the primary objectivs of r economy needs rapidly enough. In fact, it would intro- wol run coura toan of theprimaro obete duce no flexibility to speak of. Even if the aforemen- the reform program-that of transferring from the state tioned problems were ignored and it were assumed that to the private sector the power to decide the volume and flexibility would emerge eventually, the privatization allocation of investment. Whereas transferring the as- wouldiproceed to elol becausyo the ermouszoh- sets from the public to the private sector does not re- soule preeurs because of the enormous ob- quire the use of savings, if the assets were sold, savings eteses. would have to flow from the private to the public sector. enterprises That is, to pay for the 60 percent of the shares the state Equit would eventually sell, the private sector would have to Y use savings that it could be investing. It would then be As noted, transferring shares without payment to the up to the government to allocate the investment of those population meets the equity requirement. However, the savings. The purchase of 60 percent of the capital of sales of shares to the banks and workers pose a serious most enterprises would also use up the better part of pri- equity problem. The reason is that there is no way to val- vate savings for a long time. ue the enterprises without a capital market. In fact, the current state of the Polish economy makes the valuation Summary. In many respects, the proposed scheme of public sector enterprises even more difficult than is would not meet the government's long-term objectives. the case in market economies. There the public sector It would create substantial rigidity in the system. It enterprises are operating in an environment in which would discourage the development of capital markets the price of the factors of production and other basic because it would be practically impossible for entrepre- prices are set by competition in the private sector."6 The neurs to take over the holding companies and enterpris- market-based system also allows reasonable projections es. The equity markets would therefore not afford of the cash flows generated by investments. This situa- mobility for domestic capital. The scheme would also tion would not pertain at the start of the privatization in hinder foreign investment from the EEC. It would re- Poland, where the government would still own the ma- duce flexibility in the allocation of banking credit be- jority of the enterprises and factor markets and compe- cause the banks would tend to give preference to the tition would not yet exist. 280 Trying to resolve this problem by hiring engineers to dustrial enterprises but not for the small ones or for value the enterprises entirely misses the point. All the those engaged in services such as transportation, storage engineers can do is estimate what the value of the assets and retailing. State farms could also be treated different- would be in a foreign market economy, which is what ly. they know. However, it is clear that the same equipment * The role and number of the holding companies has a different value in, say, Sweden than in Poland. The should be revised to ensure that they behave in a way potential for commercializing products, the wage level, consistent with the desired flexibility in the capital mar- the cost and availability of complementary services, the kets. quality of inputs and all the other factors that determine * The amount of shares to be sold to the population comparative advantage are different. Particularly impor- should be drastically reduced so that the amount of sav- tant, the risks are different. Nor are these differences ings transferred from the private to the public sector is negligible: they can mean the difference between becom- minimized. ing rich or going bankrupt. * A mechanism to ensure that entrepreneurs are The value of the shares to be sold at the outset of the able rapidly to acquire control over the enterprises dur- privatization would have to be set arbitrarily. Most likely, ing the transition should be designed. the market price of the shares would be extremely low. * The measures need to be properly sequenced, es- The riskiness of the investment because of the uncertain pecially with respect to the privatization and the restruc- future of each enterprise would reduce the price, while turing of enterprises and banks. the proposed scheme would further depress the price of the shares sold to workers and banks. Since the sales Equal Voting Rights would be made to specified buyers, there would be no other bidders, so that it would be a buyers' market. As noted, unequal voting rights for different shares The government may feel comfortable with underval- undermines transparency, opens the door for pressure uation of the shares sold to enterprise workers. Under- groups to enhance their powers, may cause problems for valuation would smooth the process, facilitating the foreign investment and serves no purpose that cannot be elimination of the controlling powers of the Workers' obtained through other, less problematic means. These Councils. However, underpricing the shares sold to the points should not preclude firms from issuing common- banks is not easily justified. ly preferred shares, which are subordinated debt." The sale of preferred shares, however, should be something Modifications to the Proposed that enterprises decide on in the future, and savers Approach should decide if they want to buy them. Approachhbtono BnPucaeso Sae As stated, the proposed scheme is basically sound. Its Prohibition on Bank Purchases of Shares problems could be solved easily with some modifica- Selling shares to the banks presents serious prob- tions, as proposed in this section. It should be noted, lems. Most probably, the sales would lead to inequity in however, that these ideas are preliminary. The intent is the process, a misallocation of resources and financial to indicate the direction of changes rather than spell instability. The government should not only refrain from them out in detail. These and other ideas should be dis- selling shares to the banks, it should also prohibit them cussed exhaustively and compared with alternative from buying any in the open market. At the very least, it schemes during the design of the program. should limit the amount of shares the banks can buy to The discussion in this section focuses on seven pro- a very small portion of their equity capital. posed reforms to the scheme the government is consid- Following the international conventions recently ering. The first two recommendations relate to general adopted on the capitalization of banks (Cooke's Commit- rules regarding the development of capital markets, tee), their capital requirements should be estimated on while the rest apply to the privatization. The seven re- an individual basis, based on the riskiness of their port- forms are: folios. That is, banks with higher risk portfolios should be required to maintain a higher ratio of capital to assets. * All shares should be given equal voting power. The higher the level is that banks may invest in shares, * The banks should not be allowed to buy shares. the higher the capital requirement should be. * The program should discriminate among differ- One exception to this rule should be that banks can ent kinds of enterprises. A scheme for transferring accept shares as collateral and foreclose on them if nec- shares without payment to the population as a whole is essary to collect. However, regulators should ensure that an adequate solution for the large and medium-size in- the value of the shares is heavily discounted when ac- 281 cepted as collateral (to cover the risk of the volatility of Sales of these assets should proceed as quickly as pos- equity stock) and that the banks sell the shares thus ac- sible to facilitate trade. The sales can be financed by the quired to the public in general within a prescribed peri- banking system without any inflationary effect as long as od (not to exceed one year). the government does not spend the proceeds of the sales. Instead, the proceeds should be used to repay the gov- Need to Discriminate among Different Groups of ernment's debt with the National Bank of Poland. In Enterprises turn, the National Bank should sterilize the money. That Currently, big conglomerates control the provision of is, it should not grant credit or in any other way create many services that specialized, smaller enterprises could money with these resources. provide much more efficiently. Examples of areas where The Poland/World Bank/EEC Task Force on Agricul- this situation is particularly true are transportation, re- tural Sector Reform has produced several suggestions tailing, storage, distribution and all kinds of services that on the privatization of state farms and cooperatives spe- give mobility to products. Typically, the big conglomer- cifically (see Schumacher 1990). The state farms could ates provide these services only for their own purposes. be treated either as large business complexes and priva- That is, they transport, store and distribute only their tized using the same scheme as for large industrial en- own goods. This practice adds to their monopolistic terprises. Alternatively, the government could establish power in their main line of business. Additionally, since a system for leasing the land. Under this scheme, enter- nobody else provides these services for other purposes, prises would rent the land from the government but the development of other businesses that also need the would manage them to maximize their own profits. services but cannot get them-either because they do Small-scale industrial companies could also be sold at not have the capital or because their demand for the ser- public auction, following the same procedures as those vices would not justify investing in them-is con- for the service enterprises. The following paragraphs re- strained. fer mainly to the privatization of medium- and large- Although it makes sense for some big enterprises to scale industrial companies. have their own distribution network, at a time of rapid decentralization of these services the functioning of the Achieving Flexibility: The Role and Number of Holding economy must be lubricated. These services should be Companies available to anyone and should be provided competitive- ly. Since most of these services can be operated profit- Regarding medium- and large-scale industrial enter- ably on a small scale, it is probable that new businesses prises, the government needs to solve two problems: will be created to provide them. However, for the sake of eow to privatize efficiently; and how to improve the speed, the process should be accelerated by separating management of those enterprises that remain in the so- those services from the big conglomerates and privatiz- cialized sector. In both regards, holding companies can ing them quickly. Furthermore, privatization of these play a useful role,malthough the identityand organization services can be carried out efficiently by dismantling the of the holding companies are different in the two cases. units providing them and then transferring the individ- ual assets. An example would be trucking, with the Holding companies as managers of public sector en- trucks being transferred individually. Other examples terprises. Management of the assets that would remain are retail and storage facilities. in the socialized sector may be improved by organizing These assets should be transferred through sales. them under one or several holding companies that are They are small enough to sell easily, and the problem of responsible for maximizing the return on their capital.'9 valuation does not exist or is minimized. Because the as- Holding companies may be useful because the mobil- sets can be used for multiple purposes, estimation of the ity of the factors of production requires that decisions on profits that they could generate is easier. In the case of investment and disinvestment be taken by representa- trucks, for example, international prices can be used. tives of the owner rather than those who manage the en- Unlike in the case of complicated production facilities, terprises. The mobility of capital comes precisely from trucks will always be needed, and their opportunity cost decisions by owners whether to: invest the profits of is clearly their price in the international markets. The their enterprises in the same enterprise or in other activ- prices should be low enough to attract many bidders. In ities; liquidate or restructure loss-making firms; or sell this case, the best course is to auction the trucks of so- firms. These decisions can only be taken by an agent ex- cialized enterprises. A similar procedure can be followed temal to the enterprise whose fate is being decided, that with the other assets involved in the distribution and is, by the owner, who should decide how much to invest commercialization of goods.18 and in what activities and enterprises. 282 Thus, there are two levels of management of social- this end, they should promote the continuous emergence ized enterprises, both concerned with obtaining the of entrepreneurship through transparent competition for highest yield from the capital being managed. The kind the use of savings, as well as competition for the control of manager needed at the enterprise level would be re- of enterprises. The system should be very flexible, allow- sponsible for making the enterprise profitable and for ing for voluntary and involuntary take-overs. convincing investors (both public and private) to put re- Institutional investors can play a very useful role in sources into it. The manager needed at the second level, meeting both requirements of a market. However, to be which could be organized as a holding company, would effective, they should specialize more on the side of mo- provide for the mobility of resources across the social- bilizing and allocating financial resources than on be- ized sector and between it and the private sector. At this coming enterprise managers. Rather than being level the manager would work within a capital budget instruments to control enterprises, the holding compa- constraint, which could be lifted only by the enterprise's nies should be mechanisms to offer diversified risks to being profitable and by convincing the holding company savers. That is, they should become mutual funds, main- that future operations would be more profitable than the ly concerned with providing a solid return on the savings alternatives open to the holding company. The two levels of their participants. Their comparative advantage is that of management should be integrated under one set of they can pool the resources of many people to invest in a rules and perhaps only one institution. bundle of instruments offering different degrees of risk Organizing public sector enterprises into holding and expected profitability, including a mix that offers companies would not solve the overall economic prob- lower profitability but also lower risks than is true of in- lem of managing Polish enterprises. Privatization is nec- vestments in the shares of individual enterprises. This essary to create a critical mass of economic agents approach places their products between bank deposits reacting to market signals in an efficient way. The factor and equity investments in individual enterprises. Many markets will not emerge without large-scale privatiza- people would be attracted to them. tion, and in turn the holding companies managing the To achieve the objective of maximizing the return on public sector enterprises would not have to use the pric- the savings of their participants at a given level of risk, es of the factors of production as given quantities estab- the mutual funds should remain independent of the en- lished by the market. There would no point of reference terprises they invest in. That is, they should be able to in- with which to judge their behavior. As a result, they vest and disinvest quickly, basing their decisions only on could easily become yet another layer of bureaucracy, as risk and profitability, something they would not be able has happened in many other countries, such as Algeria to do if they were concerned with control issues. Their and Egypt.20 interest in management should be limited to making sure the companies they invest in are properly managed. The long-term role of privatized holding companies. They should vote on management issues through their Holding companies present many opportunities for investment strategies. If they do not like the manage- privatization, as well as many dangers. On the positive ment of a company, they would not invest in it." side, they can bring about all the benefits that institu- Institutional investors are a powerful mechanism for tional investors provide in market economies. However, encouraging good management, even though they do as was pointed out, they can also become an obstacle to not choose the managers directly. Their power is trans- the development of capital markets and the emergence mitted through the price of the shares, which fall if sav- of entrepreneurs. Thus, the potential benefits of having ers do not buy them. If the mutual funds disinvest in one institutional investors should be balanced against the company or refrain from investing in it, the groups con- danger of creating institutions that would conspire trolling the enterprises get a strong signal that they have against the emergence of markets. This balance can be to improve their management. Thus, specialization does achieved through specialization. not preclude institutional investors from playing a use- There are at least two requirements of capital markets. ful role in improving the management of enterprises. A First, they should be able to mobilize resources from the regulation prohibiting them from using the voting pow- population and make them available for investment. This er of their shares would prevent their perversion into condition requires their offering a varied menu of instru- holders of monopolistic power. ments through which savers can invest in accordance Having strong institutional investors that are focused with their attitudes toward risk, liquidity and other pref- on the profitability of their portfolios would be helpful erences. Second, the capital markets should promote ju- for other reasons. By investing most of their resources dicious investment of those savings. That is, they should for the long term, they would be a stabilizing force in the encourage the efficient management of enterprises. To market. They would also help make the market transpar- 283 ent. Their relatively large size would allow them to in- The number and size of institutional investors. Re- vest time and money to gather information and analyze gardless of their shape as mutual funds or pension funds, it in order to invest their resources better. the number of institutional investors should be increased In summary, the role of the holding companies considerably from the five suggested in the current pro- should be changed from that of controlling units to that posal. A balance should be struck between making them of institutional investors. The rules under which they numerous enough to encourage competitive behavior and operate should be drafted accordingly. For example, use making them big enough to have a diversified portfolio. of their voting powers would be prohibited. If the gov- One hundred might be a good number to start with. Of ernment opts for this role for the holding companies, these, 10 could be pension funds and the rest straightfor- they should be called institutional investors or mutual ward mutual funds. The sizes of the institutions could funds instead of holding companies. The remainder of differ, with the pension funds being bigger than the oth- this paper refers to them as institutional investors. ers in order to accommodate the pension liabilities.23 Even if they number 100, the size of the individual in- Linking the creation of mutual funds with pension stitutions being privatized will be too big if they keep reform. One possibility the government may wish to most of their initial holdings. The bulk of the shares consider is linking the creation of the mutual funds with should be owned individually, at least at the start of the reform of the pension system. Wide diversification of in- process. This approach would give more flexibility to the vestments by these institutions from the outset would process and would make it easier for entrepreneurs to make them safe enough to meet their long-term respon- emerge and take control of individual enterprises. It sibilities. If the government gives away the ownership of would also provide flexibility to individuals who prefer to these institutions, it might as well transfer to them the have their shares pooled so that they can organize their pension obligations being covered directly by the bud- holdings as they want. Some may want to establish true get. holding companies; others may prefer to have mutual Presently, Polish pensions are based on a pay-as-you- funds that specialize in certain types of business; others go system. That is, current contributions are used to pay may prefer to have only friends as fellow shareholders. If current obligations. No capital has been accumulated to the institutional investors keep most of the shares, the cover the obligations. The problem is that reductions in individuals will have to accept the government's design, the growth rate of the population will have a detrimental at least in the short and medium terms. effect on the viability of the system, as the pension liabil- In short, the institutional investors should begin op- ities will increase while the contributions decline, with erations holding only a minority of the shares of the the government having to cover the difference. The US privatized enterprises. The bulk of the shares would go social security system, for example, is in difficulty be- to individual investors. However, institutional investors cause of the decline in the birth rate following the "baby can also be used to transfer shares in enterprises to indi- boom." 22 viduals, in addition to transferring their own shares to In Poland, the government could capitalize the pen- them. Toward this end, the government could give the sion system with some of the shares to be given away. institutional investors two packages of shares, one of However, to make the transfer without payment equita- which they would keep and the other of which they ble, the resulting pension system should cover the entire would transfer. In addition, the institutional investors population. To make it more efficient, its management would transfer their own shares to the population. should be privatized and made competitive, even if the Minimizing the Transfer ofP.ivate Savings to the Public system is public in nature. To this end, the government Sector would pass laws requiring all workers and their employ- ees to make monthly contributions to a pension fund, The government should retain a portion of the shares but with the freedom to choose which fund to contribute for subsequent sale. One reason is that management to. should be controlled during the transition period. If the Some of the institutional investors created during the privatized holding companies are precluded from using privatization could be established as competing pension the voting power of their shares, there would be a period funds, while the rest would be set up as straightforward when the government would have to manage the enter- mutual funds. Although a detailed discussion of how to prises because of a lag between the time the shares are carry out this plan is beyond the scope of this paper, the given away and the emergence of groups able to control section on the role of the holding companies in the the enterprises through the acquisition of shares in the privatization provides some suggestions. free market. During that period, the government should keep enough shares to ensure control. 284 Beyond the reasons already cited, the government This system should not preclude take-overs by poten- needs to sell part of the assets to raise the resources tial entrepreneurs buying shares in the free market. For needed to resolve the problems created by the previous this reason, the shares the government keeps and the regime. Foremost among these is the need to restruc- amounts promised to the contract managers should be ture and recapitalize the banking system. The govern- only the minimum necessary to ensure control when ment will have to take on this responsibility as nobody initial trading in the shares takes place. If the contract else will want to: recapitalization of the banks essentially managers want to maintain control in the face of a po- involves the absorption of past losses, and no private in- tential take-over, they would have to buy more shares in vestors are going to use their savings to cover the losses the market. of others.24 The government may use the proceeds of the sale of shares in the enterprises to recapitalize the bank- Distributing bundles of shares. As was suggested, the ing system. government should distribute two kinds of shares with- To achieve these objectives while minimizing the sav- out payment to the population. One would be shares in ings syphoned off from the private sector, the govern- the mutual funds, the other shares in the individual ment should retain only the minimum required to companies. To simplify the process, initially the compo- ensure control in a dispersed market. The amount sition of the bundles of shares from individual compa- should be no more than 10-20 percent of the shares. nies could be the same as those from the mutual funds. Further to minimize the negative effect on private sav- If, as is proposed later, the composition is determined so ings, the government should use the proceeds from the as to cover a wide range of activities and geographic lo- sale of those shares to recapitalize the banks.25 cations of enterprises, the expected return on these The precise amount should be based on estimations shares would be close to the average that all the priva- of the amounts needed to recapitalize the banks. For its tized enterprises would generate. This approach would part, the government would get revenues from the sale be helpful for people who do not have a clear idea what of small industrial and service enterprises, which it they want to do with their shares, since, if they just kept would transfer 100 percent through sales. the bundle of shares, they would obtain the average rate of return and the same return on their shares in the mu- Facilitating Control by Entrepreneurs over the tual funds. Other people might want to change the com- Enterprises position of their shareholdings, a positive step because it would lead to trading in the shares and ultimately to the It is very doubtful that the government's manage- establishment of prices for the assets and transfer of con- ment will be very efficient during the transition. Howev- trol over the enterprises to emerging entrepreneurs. er, its control should at least help avoid the chaotic situation that would develop if nobody were to exert con- Protecting people who do not understand the pro- trol over the enterprises. Nevertheless, the government cess. Protecting people who do not understand the pro- should minimize the duration of its control during the cess is important initially for equity reasons. However, transitional period. The process should be designed so the protection should be provided in a way that does not that private entrepreneurs take control of the enterpris- endanger the success of the privatization. One way is to es as soon as possible. There are two complementary phase the transfers without payment so that the benefits ways to accomplish this task. One is to use management and losses that people can experience in managing their contracts during the transition. The other is to design shares become clear before the second round of transfers the distribution of the shares in such a way that facili- takes place. In the long run, these people could exchange tates take-overs. all their holdings for shares in the mutual funds, which would carry low-risk, balanced portfolios. They would be Management contracts. To expedite the process, the able to shift from one mutual fund to another by selling government may sign management contracts with qual- and buying shares. ified foreign or local entrepreneurs, paying for the man- agement services but also offering an option to buy a Sequencing and Transitional Issues part or all of the portion of shares retained by the Trea- sury within a specified period (say, after three years) at a The difficulty of the transition should not be underes- predetermined price. This system would give the manag- timated. The deep structural reforms needed to intro- ers a powerful incentive to manage the enterprise effi- duce market forces will cause considerable turmoil ciently so as to increase the market value of the shares. during the adjustment period. Many people may suggest In the process, they would benefit all the shareholders. gradual implementation of the process so that the econ- 285 omy has time to absorb some of the traumatic shocks be- hibit investment of public sector and banking system fore having to cope with others. funds in enterprises in which the government has a ma- This paper espouses the opposite point of view. The jority stake. Any investment in socialized enterprises market is an all-encompassing system whose benefits should be financed out of the enterprises' own resources. can be experienced only when all, or most, of the econo- The government should also stay away from coordi- my is functioning under its rules. Partial application of nating the restructuring, limiting itself to facilitating market forces is likely to misdirect entrepreneurship the process through fast privatization of the socialized away from productive activities into speculative ones. sector, cleaning up and privatization of the financial sys- This eventuality would hamper completion of the re- tem and creation of the infrastucture needed to foster form. It is better to address the turmoil quickly than to the mobility of resources. Creating inter-ministerial go through a protracted process that does not afford the committees to control or coordinate the restructuring benefits of the market. Although a full discussion of the and modernization of enterprises would most likely slow difficulties of the transition is well beyond the scope of the process and discourage private sector efforts. this paper, the following paragraphs discuss a few of them.2" The fiscal effects of privatization. The transfer of ownership is a zero sum game: both the increase in the Accounting. Any privatization scheme requires a public sector's wealth and in the income derived from it change in the accounting system. The systems developed would be counterbalanced by matching declines in the in the Western economies should be adequate, and es- wealth and income accruing to the government. If the tablishing a new, basic set of books for the enterprises government does not reduce its expenditures in line should therefore be a relatively quick task if the govern- with its reduced wealth and income, or reduces it in ment establishes a basic set of accounts to be kept for tax amounts smaller than the increased expenditure in the purposes and sets a deadline for the change-over. To ac- private sector, the effect on the economy would be infla- celerate the process, the government should allow com- tionary. This effect, however, is likely to take place inde- panies to hire foreign consultants to help them make the pendently of the method used to privatize. The transition. government will lose the income from the enterprises No reconstruction of the records of past operations is even if it sells them. needed. New operations, however, would have to be re- corded consistent with the operation of the market and Notes the tax system. For bookkeeping purposes, the initial valuation of the assets could be arbitrary, to be adjusted 1. This paper talks about a government proposal that preceded the after, say, two years, when the capital markets are more one discussed in chapter 21, "Markets and Institutions in Large-Scale Privatization: An Approach to Economic and Social Transformationin developed. Eastern Europe," by Roman Frydman and Andrzej Rapaczynski in this volume. Although the proposal was never implemented, the points Enterprise restructuring. The government should made in Hinds' paper are still relevant, and it was therefore included in avoid investing in the restructuring and modernization the conference papers. of enterprises, leaving this task for the private sector. If 2. For a discussion of the relationship of ownership issues and sta- bilization and economic recovery, see Hinds (1990). it were to manage the restructuring of the socialized sec- 3. Selling shares to the banks cannot be considered privatization, tor, the most likely result would be inefficient invest- at least as long as the banks remain part of the public sector. ment and outright waste. 4. As a result of the imprudent financing of leveraged buy-outs, As long as the banks remain in the public sector, the fashionable in the 1980s, take-overs have acquired a bad name. These socialized sector will enjoy considerable political power particular buy-outs constituted, however, only a very small minority of the changes in ownership that took place in that decade. In a market over the banking system as a result of the pressures ex- economy, transactions related to the control of enterprises occur con- erted on both the central and local governments. In re- tinuously. Nor is control sold just because the enterprise is doing badly. sponse, the tendency of the banking system will be to In many cases the owners sell because they think they have a better use lend to the socialized sector, to the detriment of the pri- for their capital (they want to concentrate their efforts in other areas vate sector. Experience shows that socialized enterprises or they have financial obligations that need rapid cash mobilization). vate sector. Experience shows that socialized enterprises 5. This argum ent is the m irror image of the recognized non-dis- tend to invest very inefficiently. Most likely, then, the tortionary effect of lump-sum taxes on the subsequent behavior of the flow of resources toward the socialized enterprises taxed agents. would not only starve the private sector of resources but 6. For a more detailed discussion of the relative benefits of selling would also result in substantial waste. and giving away shares, see the third section. To encourage the growth of the private sector and 7. Getting rid of the fiscal burden imposed by the dependence of avoid wasteful investment, the government should pro- public sector corporations on the government's budget is one of the avoid wasteful Investment, the government should pro- 286 main reasons behind the current trend to privatize in the OECD coun- 18. It might make sense for some big enterprises to retain some of tries. their distribution infrastructure, including storage facilities. However, 8. Even if management is shared with the banks, each bank and the process should be biased toward complete decentralization to cre- holding company would still be responsible for a very large number of ate a clean slate so that all enterprises have access to those services and companies. to stimulate competition. If it is profitable for some enterprises to ac- 9. Some mutual funds in large market economies have invest- quire distribution networks, they should do so in the future, as part of ments in thousands of enterprises. These funds do not, however, inter- their natural expansion. The government, however, should avoid re- vene in their management. They are passive investors that hold shares building the current monopolistic structure. just to get capital rents. In contrast, holding companies that manage 19. This point is critical. Efficiency will not improve if maximiza- enterprises typically handle a much smaller number. In the scheme tion of profits is not the sole aim of the managers of public sector as- proposed for Poland, the holding companies would be the control type. sets. 10. Buying 20 percent of the voting stock of one holding company, 20. For a more detailed discussion of ways to improve the manage- for example, would be equivalent to buying 2 percent of the total en- ment of public sector enterprises and the need to create a critical mass terprise voting stock in the country. of private enterprises to help improve the management of the public 11. Another frequently cited example is Japan. However, the eco- ones, see Hinds (1990). nomic and social organization of Japan is quite unique and can hardly 21. Specialization is desirable in part to avoid conflicts of interest. be taken as an example of what would happen in a Western culture It is also needed because the expertise required for managing an effi- such as that of Poland. cient institutional investor is different from that needed to run a hold- 12. When reforms are being planned, it is better to rely on the av- ing company. The former is financial in nature, while the latter is erage experience of many countries than on that of exceptional ones. mainly management of real resources. One clear example is the concentration of economic power in Sweden, 22. This problem affects only the pension liabilities of the system. where a small number of conglomerates control a sizable portion of Private pension funds are based on the accumulation of resources. the country's gross domestic product (GDP). This situation has not re- 23. It is desirable that the number of pension funds be kept small sulted in a lack of competition or efficiency. On the contrary, many because they need to be supervised closely as a result of their provision Swedish enterprises are among the most competitive in the world. of a public good. A Superintendency of Pension Funds should be creat- However, not many people would use the Swedish example to argue for ed, responsible for supervising prudent management of the pension re- concentrating wealth in a limited number of families as a way to sources. achieve competitiveness. 24. See the section on the recapitalization of banks. 13. The term "multi-tier share structures" refers to a structure of 25. The timing of this recapitalization most probably will not co- shares with unequal voting powers. incide with the collection of the revenues from the sales. This financial 14. The price of the shares kept by the government would be lower problem is, however, soluble (see the section on the recapitalization of than those in the hands of the other owners because the former would banks). not have voting power. Nevertheless, buying them would transfer sub- 26. This section does not touch on many important transitional is- stantial savings from the private to the public sector. sues, such as the sequencing of bank reforms and enterprise restruc- 15. To resolve the crisis, Chile dismantled the large conglomerates turing and needed fiscal and monetary policies. For a discussion of 15t Tontreolved the crisis,ndChilepriismantld t large cngmerates these and other transitional and sequencing issues, see Hinds (1990). that controlled the banks and enterprises, sent a large number of com- panies into bankruptcy and took control of the largest commercial banks, recapitalized them and sold them to new owners. After that ex- References perience, the government changed its strategy for privatization com- pletely, the new aim being to spread the ownership. Since then, the The Economist. 1990. "The Wallemberg Empire." June 23, p. 76. enterprises have been quite successful. 16. Competition for the factors of production and other inputs ex- Hinds, Manuel. 1990. "Issues in the Introduction of Market Forces in ists even if the publicly owned enterprises are not in the same line of East European Socialist Economies." World Bank. Washington, business as the private ones. D.C. 17. Subordinated debt means that, in case of a liquidation, the claims of the holders of these liabilities defer to any other liability hold- Schumacher, Augustus. 1990. "Structural Reform of State Farms in ers except those represented by common shares. That is, if the enter- Poland: Short, Medium and Longer Term Options." prise is liquidated, all other liabilities are paid first, then those of the preferred shares are paid and finally those of the common shares. 287 23 Comments on David Lipton and Jeffrey Sachs, "Privatization in Eastern Europe: The Case of Poland," Roman Frydman and Andrzej Rapaczynski, "Markets and Institutions in Large-Scale Privatization: An Approach to Economic and Social Transformation in Eastern Europe" and Manuel Hinds, "A Note on Privatization of Socialized Enterprises in Poland" E. Borensztein There can be little disagreement that privatization is not a coincidence that almost all Central and Eastern both the most important and the most difficult of the European countries have instituted taxes on excess wag- economic reforms the countries of Central and Eastern es that penalize increases above certain limits, with the Europe face. The papers by Manuel Hinds, Roman Fryd- taxes at rates that reach several hundred percentage man and Andrzej Rapaczynski, and David Lipton and Jef- points of the wage increases. An extreme consequence of frey Sachs all provide trenchant analyses of the this poor incentive system was the emergence of "spon- challenges of privatization. taneous privatizations," as described by Frydman and There are some fundamental points on which the Rapaczynski and by Lipton and Sachs, which took place three papers agree: privatization must be speedy, for especially in Hungary and Poland. there are large costs in delaying it; it must be compre- Thus, there is agreement that privatization must pro- hensive, as opposed to pursuing a case-by-case approach; ceed at once, even at the risk that the inefficiency of the and it must be distributive rather than based on the sale system of public enterprise management and control of assets. There are, however, important points on which blunts progress on all reforms. The question, then, is the authors disagree. The two main ones are: the diffi- how to proceed with the privatization? cult problem of how corporate control should be struc- The second important point of agreement in the three tured; and the desirable role of the state and other papers is that privatization cannot be accomplished by financial institutions in the new market structures. Al- conventional sale methods for a number of reasons, in- though the three papers refer specifically to Poland, ba- cluding: the near impossibility of obtaining a meaning- sically everything said is applicable to other previously ful valuation of the enterprises; equity and political centrally planned economies attempting to shift to a consideration; the lack of domestic savings; a lack of market economy. burning interest on the part of foreign investors to in- As to the urgency of undertaking privatization, a vest heavily in purchasing enterprises; and the political point on which the three papers agree, Poland, and in unacceptability of heavy foreign investment, even if it fact most Central and Eastern European countries and were likely. Therefore the idea of a free distribution of the Soviet Union, find themselves in a no-man's land in equity to the public in general, the "voucher scheme," which, although central planning no longer operates as appears to be necessary in terms of speed and compre- a system, the lack of private property and a clear profit hensiveness, and also on the grounds of equity and polit- motive impedes the emergence of a full-fledged market ical acceptability. economy system. In this situation, there is no clear au- The government cannot, however, simply issue the thority to monitor the behavior of enterprise managers. vouchers and sit back. Because of the thin dispersion of Moreover, the perception by managers whose tenure is ownership, managers effectively would be subject to no highly uncertain creates perverse incentives for exces- supervision because it is not worthwhile for a small sive wage and bonus payments, low levels of investment stockholder to incur the considerable cost of monitoring and a squandering of the assets of the enterprises. It is management while receiving only a minimal part of the 289 benefits. This issue is the problem of corporate gover- the country. Given that corporate governance is an area nance. It is the most difficult challenge of any proposal in which economic theory has not produced conclusive for distributive privatization. or universal results, it might be prudent to leave more On this point the three papers have important differ- room for spontaneous adjustment of the system. ences. Hinds propounds, as a final objective, a system of Frydman and Rapaczynski's proposal has the interest- enterprise control that is essentially the same as that in ing element of private initiative from the very beginning. the United States and the United Kingdom. In this sys- By the same token, however, choosing an appropriate tem, publicly held equity is an important component of mutual fund may require too great an informed decision capital, and take-overs are the main mechanism of man- on the part of the general public, and the establishment agement discipline: if an enterprise is perceived to be of a large enough number of mutual funds may be ham- poorly managed, some investor group will attempt to ac- pered by the lack of domestic expertise and reluctance of quire it and introduce changes in management and ad- foreign investors. Moreover, the auction process could justment measures. be excessively complicated, and it is possible that not all By contrast, Lipton and Sachs consider the take-over enterprises could be sold to qualified investors. The state mechanism as essentially flawed and propose instead a might end up saddled with inefficient enterprises and detailed plan of corporate control by several financial in- face politically costly liquidation decisions. termediaries (mutual funds, pension plans, banks, etc.), It seems clear that no strategy is dominant and that to which a "stable core" investor would eventually be any plan will involve unavoidable trade-offs between de- added. Although these financial intermediaries would be sirable objectives. One evident area where a trade-off is privately owned, they would be created by the govern- unavoidable is between speed and comprehensiveness ment, which would endow them with certain equity on the one hand and the extent of private initiative on holdings. Their initial directors would be nominated by the other: if privatization is to be accomplished at once, the government. in the sense of both ownership and control of enterprises Frydman and Rapaczynski propose a different plan. passing to private hands, there must be extensive state They are not averse to take-overs but believe that the de- involvement and little room for private initiative in de- velopment of stock markets and sophisticated financial signing the corporate governance structure. securities will not be possible for a long time. Therefore At one end of the spectrum is the Hinds proposal, in they also propose that mutual funds act as the control- which the final structure of who is going to own what ling shareholders of the enterprises. However, in con- and how is it going to be run are left completely to pri- trast to Lipton and Sachs, Frydman and Rapaczynski vate initiative, although the state would continue to op- propose free entry into the mutual fund market, compe- erate public enterprises until the markets and tition among the mutual funds to obtain vouchers from entrepreneurs developed. Effectively, privatization the public, and auctioning of enterprises to the different would not take place until markets and entrepreneurs mutual funds in such a way as to ensure one or more had been established. At the other end of the spectrum is large shareholders (at least initially). the Lipton and Sachs proposal, in which the state de- Starting a complex system of private ownership with- cides at the outset everything from shareholdings by out any previous experience, institutions or entrepre- each party to seats on the board of directors. Frydman neurs is Herculean work, and all of the proposed and Rapaczynski's proposal lies somewhere in between alternatives have some weak points. The corporate struc- this trade-off, given that establishment of intermediaries ture envisioned by Hinds requires the emergence of both and auctioning of enterprises would take several years. entrepreneurial investors with sufficient capital and Hinds counters that large holding conglomerates as well-developed financial markets that will provide the in the Frydman and Rapaczynski and Lipton and Sachs complicated financing that take-overs many times in- proposals would become powerful empires not subject to volve. Their emergence might take a very long time, and the threat of take-overs or proxy fights. That situation in the meantime the government would remain in con- would not be consistent with adjusting the size of firms trol of the enterprises. Further, there are serious doubts or the scope of their activities, both of which might have about the efficiency with which they would operate. become overextended. An alternative worth considering Lipton and Sachs's framework would achieve privati- is that of using private holding companies as the privati- zation of the majority of industrial enterprise capital in zation agencies, with a predetermined termination date, the shortest time. However, the structure of corporate as proposed in Blanchard, et al. (1991). This alternative governance appears to be too rigid; it does not leave any is not very different from Hinds' proposal except that the margin for spontaneous developments that may be ap- holding companies that control the enterprises would be propriate to the technical and informational resources of private and profit-motivated instead of state agencies. 290 Decisions on breaking up the enterprises, selecting buy- failures and in financial crisis. The state had to renation- ers, etc. would all be made with a profit objective. Al- alize a number of enterprises and banks to avoid their though this proposal cannot avoid the above-mentioned closing. It is interesting that when Lipton and Sachs trade-off in that the state must play a prominent role in note the Chilean experience, they attribute the failure to establishing and launching the holding companies, the excessive leverage in the privatized enterprises. On bal- holding companies would be only a temporary device, so ance, given the necessary fragility and absence of well- that different forms of corporate governance could de- developed financial markets (at least initially) in Central velop over time. and Eastern Europe, the Chilean experience appears Regarding the role of the state, the three papers also more relevant than the German or Japanese ones and contain different conceptions. In addition to assigning suggest caution in establishing close relationships be- the state a role in setting up and nominating directors tween banks and their clients. for the different financial intermediaries, Lipton and There is one point on which the three papers appear Sachs reserve a relatively large share for the state in the to agree but this commentator does not: the problem of privatized enterprises (illustratively some 30 percent), the monopolistic structure of markets, including both although most of it probably would eventually be sold to horizontal and vertical integration. Lipton and Sachs 4"stable core" investors. Although in a lesser proportion than that of its holding shares, the state would have seats cientder of competitioniaured by t stin on th boads ofdiretors cient degree of competition IS assured by the existing onydthe bardso dRepctrs.i would prefer to keepthe market structures and foreign competition. This conclu- F'rydman and Rapaczynski would prefer to keep the sion appears to contradict the views that attribute the state completely out of the picture if possible with re- lincrease inpric te pie lbaliation iho gards to both the establishment of financial intermediar- land t"nopolstic pricig In any levent,in th more ies ad paricipaion n entrpris contol.Iand to "monopolistic pricing." In any event, in the more ies and participation in enterprise control, centralized economies of Central and Eastern Europe, Hinds questions the merits of having a large public such as Czechoslovakia, the problem of monopolistic stk netrrs aia, bae on a.ifrntprpc market structure should certainly be dealt with. Given tive. As the state sells this stake, it will absorb a large tat b re umonli certainly easier an porton f prvat saingsandcrow ou prvateinvst- that breaking up monopolies is certainly easier and less portion of private savings and crowd out private invest- liiiu beor thnatrpiaiato,ato.nmr ment. While this scenario is true, some increase in gov- l ernment revenue would be needed in any event unless ket structure appears to be the necessary starting point the privatization strategy alters significantly the level of of privatization.! A related issue is the likely existence of government spending, and the only issue is to choose be- an excessive size and/or vertical integration for some tween a tax increase or maintaining the rights to in- firms, which has resulted, for example, from uncertain come-producing assets. access to inputs. This issue is different, because enter- Another point of disagreement is the role of the banks prises themselves would benefit (to the extent that they in the new corporate structure. Lipton and Sachs pro- are not compensated by gains in market power) by di- pound that the banks receive shares in the enterprises vesting divisions or activities. Thus this type of break-up (which would also solve their need for capitalization) would not require government intervention. However, and that they become active in enterprise control, in- this type of inefficiency increases the risks that corporate spired by the German or Japanese models. (In the United structures exist that could support a tendency to empire- States the Glass-Steagall Act prohibits banks from own- building. ing equity, and they cannot participate on corporate The three papers contain elements that are extremely boards as lenders). valuable not just for Poland but for any of the previously Hinds argues strongly against a role for banks in the centrally planned economies attempting a transforma- ownership and control of enterprises because of a poten- tion into a market economy. Although it is impossible to tial conflict of interest. Frydman and Rapaczynski, while design a privatization proposal to fit all the reforming supporting universal banking (allowing banks to own Central and Eastern European economies, some coun- equity), do not predetermine a specific role for banks. To tries (such as Czechoslovakia and Romania) are attempt- back up the case about problems with conflicts of inter- ing large-scale privatization in ways not unlike those est, Hinds cites the Chilean experience after the first followed by Poland. Moreover, the analysis in these pa- wave of privatization. A few powerful conglomerates pers may be even more important for those economies were generated that owned both banks and their largest that may follow the case-by-case traditional approach to customers. After a few years they ended up as massive privatization (as seems to be the case with Hungary). 291 Notes References 1. This point is effectively argued by Tirole (1991), who also puts Blanchard, Olivier, Rudiger Dornbusch, Paul Krugman, Richard La- forward a proposal for joint work by authorities and international in- yard and Lawrence Summers. 1991. Reform in Eastern Europe. stitutions to ensure competitive market. Cambridge, Mass.: MIT Press. Tirole, Jean. 1991. "Privatization in Eastern Europe: Incentives and the Economics of Transition." Mimeo. Massachusetts Institute of Tech- nology. Cambridge, Mass. 292 Part VIII Summing Up and Overview 2A The Symposium in Review and a Glance at the Future Stanislaw Gomulka, Johannes E Linn and Jan Svejnar Comments by Stanislaw Gomulka The discussion at the conference suggests that the transition typically involves four phases: The intent of this conference was for participants to * Phase 1 centers on macrostabilization, if needed, leave better informed and wiser about the best way for and on the liberalization of prices, the latter to include Central and Eastern European socialist countries to pro- (internal) convertibility of the local currency and the ceed with the transition to a market economy based on dollarization of trade with former Council for Mutual private ownership. In this context, the papers cover an Economic Assistance (CMEA) partners. This phase is unusually wide range of topics. This transition may be preceded by a short preparatory period in which the gov- seen as the latest, and by far the most important, stage ernment takes measures, such as the elimination of the in a long history of reforms. In several socialist coun- budget deficit and the monetary overhang, to establish tries, the original centrally planned economy of the So- the right "initial conditions." viet type first became a modified centrally planned * Phase 2 is about structural adjustment. It in- economy of the Hungarian type and then a market-ori- volves the commercialization and privatization of state ented socialist economy of the Polish-1990 type. Within enterprises and the development of labor and capital the next five years or so it is to be changed into a fully markets. It is also a time of big changes in the composi- fledged market economy of the Western type. In discuss- tion of the output of products and the reorientation of ing these developments, the authors have identified the links in international trade. It begins with phase 1 but reasons for abandoning a particular system and embrac- continues some three to five years longer. ing specific reforms. * Phase 3 is intended to induce the recovery of At this time of deep economic crisis in the USSR and growth from the deep recession brought about by the Central and Eastern Europe, it is easy to overlook that supply shocks and contractionary demand policies of the traditional centrally planned economies once were phases 1 and 2. The recession produces high unemploy- capable of respectable, on occasion even impressive, ment, typically of more than 10 percent of the labor force growth. However, the post-1975 slowdown came sooner and hopefully less than 20 percent. The recovery should and was deeper than in the capitalist West.' This dispar- begin in phase 2 and continue for several years after the ity provided the empirical support for much of the West- end of phase 2. It will be driven by the export sector and ern understanding of these economies. The social private initiatives, fueled by the underutilized resources tensions that developed during the slowdown, indeed of labor, land and capital. The new laws on property own- stagnation, forced the governments in two key coun- ership by foreigners and full remittance of profits abroad tries, the USSR and Poland, to abandon prudent macro- will attract foreign investors. In the case of Poland, there economic policies. This action very quickly led to an should also be an agreement on the reduction of foreign open crisis: it appeared that their economies were disin- debt and/or debt service payments. The catching-up with tegrating and the political authority collapsing. The cri- Western Europe should begin in this phase. sis may be seen as the necessary precondition for * Phase 4 is a period of sustained, balanced growth, abandoning the socialist experiment and persuading the accompanied by a macroeconomic policy intended to population at large to accept the high cost of transition produce a low-inflation environment. to a market economy, as well as the inequalities and un- certainties to be expected under such a system. 295 Much theoretical analysis and policy debate in Cen- Comments by Johannes F. Linn tral and Eastern Europe have focused on phases 1 and 2. Four models of transition may be distinguished in these As someone who is not a specialist on reform in the phases: (1) the German supershock strategy; (2) the Pol- socialist economies in Central and Eastern Europe, I lis- ish (classical) model; (3) the proposed programs in Hun- tened with great interest to the discussion of the previ- gary and Czechoslovakia; and (4) the presidential ous sessions and found there was a lot to learn. The (compromise) model for the USSR. The major distin- participants addressed five principal questions. They are guishing factor among them is the speed with which the reviewed briefly, followed by a summary of the emerging domestic prices for tradables are brought to internation- answers. al levels, the subsidies are removed and the domestic The first question is whether the socialist economies markets are opened to international competition. The should undertake reform. There was no disagreement essence of the transition is the shift in the job of allocat- around the table-an affirmative answer was virtually ing resources from planners to prices. Speed refers to taken for granted. This unanimity is, however, one of the the time taken to put a proper price system in place. great miracles of the present time, given that a little over Speed, however, also has implications for the depth of a year ago the answer would have been a lot less clear, at the recession. least as far as official representatives from the Central This latter point leads again to the helpful role of cri- and Eastern European countries were concerned. sis in creating the necessary commitment to deep and Second, there is the question of where the reforms ul- rapid reforms despite the high cost. The early literature timately should end up. That is, what is the economic on reform tended to imply the presence of substantial system toward which the Central and Eastern European and immediate benefits from market-oriented changes. countries are and should be striving? This gathering did However, the current reforms in Central and Eastern not discuss this question to any great extent, and it is ad- Europe entail a major discontinuity in systemic change, dressed below. with large shifts in demand and large redistributions in Third, what is the appropriate timing and sequenc- the incomes of households and enterprises. In this situ- ing of reforms? Much of the debate around the table fo- ation, resources used inefficiently will quickly become cused on this question. In the end, it appeared that the unemployed, and their restricted mobility will ensure debate came out not far from the conclusions drawn by that they remain unemployed for a prolonged period. Stanley Fischer and Alan Gelb in their papers. The core The reformers and nations of Central and Eastern Eu- of their answer to this question is that the timing and se- rope must be prepared for a "long march," which they quencing of reformpackages that include some progress must accept as unavoidable. and action on virtually all fronts throughout the reform While price liberalization may produce large short- need to be considered. The challenge for the policy-mak- term costs, privatization is likely to bring about large er then becomes one of keeping a large number of balls and long-term changes in the distribution of wealth and in the air at the same time and carefully timing the income, with profound social implications. The defen- progress in each area so that it has a maximum favorable sive behavior of state-owned enterprises in Poland this impact on the overall progress of reform. year highlights the need to accelerate the privatization The fourth and important question is how the re- rapidly, and to do so despite the virtual absence of inter- forms are to be designed and implemented in practical mediary capital market institutions, such as mutual terms. Some of the papers considered issues of detailed funds or private pension funds, and the lack of internal design and implementation, but overall the discussion financial capital. The issues to be solved include the was perhaps too far removed from the practical reality method(s) of pricing assets, the extent and method of policy-makers in the Central and Eastern European distributing the free shares, the potential use of existing countries face. Much more needs to be done in this area. institutions such as banks and the state pension fund, Finally, some participants asked how the impact and and the role of foreign capital. The two conference pa- progress of the reforms can be measured and monitored pers on these issues contrasted the virtues of state-driv- with available national statistical indicators. It is likely en and market-driven marketization. Both call on that the reforms will push activity out of the measured intermediate capital market institutions, which would into the non-measured sectors of the economy. If so, the need to be created, to play a key role. It is therefore costs of adjustment (as reflected in the measured decline worth noting that neither the Hungarian program nor in economic activity in the measured state sector) may the Polish plan rely on these institutions. They see a role be overstated relative to the benefits of reform (which for them, but only in the later stages of privatization. would be reflected primarily in the private, parallel and 296 non-measured sector). This situation may affect both the arrangements in the industrial market economies. The professional assessment of the impact of the reforms and United States is characterized by high labor mobility and their political sustainability, as people and politicians low employment combined with low job security and a understandably focus on the official national statistics. relatively meager social safety net. Western Europe, in Improvements in the economic and social indicators are contrast, has traded labor mobility and low unemploy- therefore a high priority, both through refurbishing the ment for high job security among those employed and a conventional national statistical systems, as well as relatively generous social safety net. It would be useful if through ad hoc surveys at the firm and household levels the economists could help clarify the trade-offs among to provide a quick snapshot of the impact and progress these models and perhaps other altematives and point of the reforms. out to policy-makers the importance of setting clear To return to the second question with more detailed goals in this area. comments-what is the economic system toward which With respect to monetary control, there was little dis- the socialist economies in transition should ultimately agreement around the table that economic reform must aim-it is an important area where research and policy begin with the establishment of firm controls over ag- analysis should provide more guidance to policy-makers. gregate monetary and credit flows. However, it appears The comments here focus on each of five areas: owner- that in some Central and Eastern European countries, in ship rights; labor markets; monetary institutions; trade; particular Czechoslovakia and the Soviet Union, contin- and the role of government. uation of central monetary management and currency As regards ownership and property rights, it appears unity is no longer taken for granted. In Western market that over the last 12 to 18 months there has been a sea economies the benefits of monetary integration and in- change in Central and Eastern Europe in clarifying that dependent monetary control are well understood. The the goal of reform is to establish private property rights potential risks and costs of a fragmentation in monetary as a critical component of a market-based competitive authority in the Central and Eastern European countries economic system. The role of economists in establishing should be carefully considered before moving too far in this goal was perhaps surprisingly small, although Man- the direction of disbanding current monetary arrange- uel Hinds and Janos Kornai have recently come to the ments. forefront in postulating that without widespread private In the area of trade, there was general agreement that property rights successful reform of systems is not pos- the socialist economies should establish open trade re- sible. The lack of clear guidance from economists on this gimes with a significant export orientation. While this question is surprising in view of the strong evidence, at goal was generally accepted, in practice the tendency least for certain sectors in traditional market economies, may be in the opposite direction. Barriers to trade that indicates that efficient allocation of resources, espe- among the Central and Eastern European countries may cially efficient levels of investment, are only possible spring up as the CMEA arrangements break down, and, where private ownership rights are clearly established. at least within the Soviet Union, there is a rising inci- This conclusion is true for agriculture, housing and ur- dence of internal trade barriers that could very quickly ban land, small-scale industry and services. Perhaps the result in significant disruptions and efficiency losses. Ex- only area where there may be room for debate is to what ternal trade liberalization and maintenance of free inter- extent large-scale industrial firms and banking institu- nal trade should remain clear goals of reform, even as tions can remain in some form of social ownership. The temporary difficulties arise in transition. conclusions of this debate should be of intense interest Finally, there is the issue of what should ultimately be to those socialist countries that retain a strong commit- the role of the govemment in the newly reformed, former ment to universal, or near-universal, social ownership of socialist, economies. Should the state retain a large share the means of production, such as China, Viet Nam and, of responsibility for the provision of social and physical in- until very recently, the USSR. Of course, how to get from frastructure, as has traditionally been the case in Western a system of predominantly state or social ownership to Europe? Or should privatization be pursued even in this one of predominantly private ownership is also a major area, with the state giving up state control over utilities question. Nevertheless, having a clear goal in this regard and mining, transport and communications, education is a major step forward that most Central and Eastern and health? Again, what will be the role of government in European countries now appear to have taken. providing a social safety net during the transition as well In the area of labor market reform, it seems that a as in the longer term? Considering the rapid increase in clear understanding of and agreement on where the re- the incidence of poverty in some Central and Eastern Eu- forming socialist countries should and want to end up is ropean countries, more attention will have to be paid to lacking. There are some distinct models for labor market this question. 297 In sum, there are still a lot of questions to be an- 1980s, where it contributed to the subsequent onset of swered and a lot of work to be done. Nonetheless, this hyper-inflation. At present, it is particularly serious in conference has helped provide answers to some ques- Czechoslovakia, Hungary, Poland and Yugoslavia. tions while sharpening the pursuit of answers to others. The reliance on inter-enterprise credit represents an The effort was worthwhile. attempt by firms to resist and postpone the painful ad- justment to ongoing changes. From a political and eco- Comments by Jan Svejnar nomic standpoint, however, this initially innocuous activity can become dynamite. It usually starts as a large A number of presentations and discussions at this number of independent responses by individual firms, conference have stressed the importance of macroeco- some of which are increasing their payables and others nomic stabilization. Another point, made somewhat less of which are ignoring the growth of their receivables. forcefully, is that a substantial microeconomic adjust- Once the mutual indebtedness becomes large, the gov- ment is crucial if the overall package of policies is to suc- ernment is confronted with large-scale insolvency. With ceed. This observation is very important since a number unemployment growing and living standards on the de- of successful stabilization programs have become unrav- cline, this collective insolvency problem is likely to in- elled because of inadequate microeconomic adjustment. duce even economically conservative governments to As a result, the focus of these concluding comments is abandon restrictive credit policies and bail out the fail- on several microeconomic issues that need to be given ing firms en masse. special attention if the economic transformation in Cen- tral and Eastern Europe is to be successful. Labor Market Institutions and Distortions Capital Market Imperfections Four interrelated issues stand out with respect to the functioning of the labor markets in Central and Eastern The capital market in Central and Eastern Europe is Europe. First, it is argued that labor exercises consider- severely underdeveloped. There are very few banks, and able influence over wage- and employment-setting and, those that exist are undercapitalized and frequently act if unchecked, could use its power to undermine the re- in collusion with one another. The banks also have a very form. Second, heterodox macroeconomic policies re- limited number of loan officers capable of appraising and spond to this and related fears by imposing some form of evaluating investment projects. On the other side of the wage control. Third, a salient feature of socialism is that market is a large number of potential entrepreneurs who it has greatly diminished worker effort. Hence there is a will need access to credit to launch and develop their great need for wage incentives (rather than wage con- businesses. The situation is probably worst in Czechoslo- trols) to generate rapid growth in productivity and prod- vakia, with about 200,000 registered small businessmen uct quality. Finally, the state or socially owned eager to start operations and 10 commercial banks that enterprises are widely perceived to have considerable have virtually no trained staff. amounts of redundant labor, and there is a feeling that it The bottleneck in the financial sector has obvious is not being reduced rapidly enough. economic and social implications. In the Central and The labor market is very important: its institutions Eastern European context, the problem is exacerbated and functioning could make or break the economic by the paucity of small- and medium-sized firms, the transformation. It is often treated somewhat mechani- lack of significant personal wealth and the ongoing or cally at the macro level, but present circumstances in imminent lay-offs of redundant labor from the large Central and Eastern Europe suggest it needs special at- state enterprises. The limited ability of a large number of tention. In particular, the major problem is that the im- small entrepreneurs to launch (labor-intensive) busi- position of wage controls, while aimed at moderating nesses because of a malfunctioning banking sector will potentially excessive wage demands, usually reduces the have a considerable negative effect on the allocation of incentives for greater effort and work quality. resources, growth and social welfare. In considering the design of optimal labor market policies, one aspect that ought to be taken into account Inter-enterprise Credit is that worker participation in management is a phe- nomenon with strong appeal in Central and Eastern Eu- In response to the restrictive monetary policy, state rope. The system has worked relatively well in the or socially-owned firms in virtually all the Central and Federal Republic of Germany (West Germany) and Aus- Eastern European countries have resorted to inter-en- tria, to mention just the neighboring countries, and the terprise credit to counteract the credit crunch. The phe- example is visible. Moreover, spontaneous tendencies to nomenon was first detected in Yugoslavia in the mid- 298 establish participatory institutions are a tradition in the es by opening up the economy and strongly encourage region. For instance, although Hitler abolished partici- the entry of new (private) firms.4 These policies will go a patory schemes and trade unions, both re-emerged in long way toward getting rid of the short-run problem most of Central Europe after the war. The Communist without introducing significant new distortions. regimes eliminated them in the late 1940s and early Finally, the issue of labor redundancy merits a few 1950s, but whenever the central controls were lifted, words. The phenomenon is clearly a serious one that has workers' councils or other forms of participation sprang to be dealt with. The growth of new (private) firms is the up.2 It is plausible to surmise that a tendency is being ob- most promising way of absorbing laid-off workers and served that most likely will not be stopped by democrat- making involuntary separations socially acceptable. At ically elected governments. the same time, it is important to heed the evidence from The question that naturally arises is whether the phe- developing countries that the reduction of redundant la- nomenon is negative or positive from the economic bor in state-owned enterprises is a painful, often expen- standpoint. The economic theory on participatory and sive and especially drawn out process.5 In this context, unionized firms suggests that in competitive environ- the findingWin the Frydman and Wellisz paper in this vol- ments these firms do not behave very differently from ume that Polish enterprises reduced their labor force by their profit-maximizing counterparts. Empirical evi- 7 percent in 1989 and 10 percent in the first eight dence indicates that the economic effects are indeed not months of 1990 attests to a rapid decline in the redun- negative3; the functioning of the West German and Aus- dant labor force by intemational standards. trian economies has been consistent with this claim. A potential problem that could arise during the economic Notes transition is that insider power could generate perverse behavior by the firms in imperfect markets. However, 1. For a discussion of this issue, see Gomulka (1990). the West German and Austrian experience with the re- 2. This was the case in the former German Democratic Republic introduction of participatory schemes when these econ- and Poland in 1953, Hungary in 1956, Czechoslovakia in 1968, Hunga- ry and Poland in the 1980s, and Czechoslovakia in 1990. omies had imperfect markets in the late 1940s and early 3. For recent surveys, see, for example, Blinder (1990). 1950s is quite acceptable. Moreover, recent theories and 4. It is notable that until recently Yugoslav policy has failed on all empirical work suggest that participatory firms general- these fronts. ly react to external shocks by adjusting wages and other 5. The Chilean experience is an exception because the lay-offs were benefits rather than employment. This pattern is consis- carried out forcefully and sometimes with the aid of especially appoint- ed outside directors. tent with what has been observed in West Germany, Yu- goslavia and Poland, and it suggests that the presence of rents and shocks during the economic transition would References be reflected more on the wage than on the employment side of the labor market. Blinder, Alan, ed. 1990. Payuing for Productivity. Washington, D.C.: The policy implication of the above findings is that, in Brookings Institution. general, limiting the scope for rents and soft budgets is essential. Moreover, rather than destroying incentives Gomulka, Stanislaw. 1990. Theory of Technological Change and Eco- through wage controls, appropriate policies should set a nomic Growth. London and New York: Routledge, chapter 9. realistic capital tax (dividend), enforce competitive pric- 299 Participants Piotr Aleksandrowicz Vittorio Corbo Rzeczpospolita, Poland World Bank, Macroeconomic Adjustment and Growth Division (CECMG) Mark Allen International Monetary Fund, Warsaw Fabrizio Coricelli World Bank, Macroeconomic Adjustment Leszek Balcerowicz and Growth Division (CECMG) Polish Ministry of Finance Marian Crisan Wojciech Bienkowski Rumanian Investment Bank Warsaw School of Economics Marek Dabrowski Jan W. Bossak Polish Academy of Sciences, Institute of Economics World Economy Research Institute Warsaw School of Economics William Easterly World Bank, Macroeconomic Adjustment Pawel Bozyk and Growth Division Warsaw School of Economics Stanley Fischer Bogumila Brocka-Palacz World Bank, Office of the Vice President World Economy Research Institute Development Economics, and Chief Economist, Warsaw School of Economics and Massachusetts Institute of Technology, Economics Department Guillermo Calvo International Monetary Fund, Jacob Frenkel Research Department International Monetary Fund, Research Department Joshua Charap Roman Frydman Ministry for Economic Policy and New York University, Department of Economics Development of the Czechoslovakia Republic Alan Gelb Simon Commander World Bank, Socialist Economies Reform Unit World Bank, Economic Development Institute Janusz GoLebiowski Lucjan Ciamaga World Economy Research Institute, Warsaw School of Economics Warsaw School of Economics 301 Stanislaw Gomulka Roland Pac London School of Economics Institute for International Trade (Warsaw) Leszek Hajkowski Ryszard Rapacki World Economy Research Institute Warsaw School of Economics Warsaw School of Economics Andrzej Rapaczynski Miroslav Hrncir Columbia University, Law School Institute of Economics (Prague) Petru Rares Danuta Hubner Rumanian Agency for the Promotion of Investments Warsaw School of Economics and Economic Assistance from Abroad Erika Jorgensen Andrew Rasbash World Bank, Socialist Economies Reform Unit Delegation of the European Community (Warsaw) Janusz Kaczurba Werner Riecke Polish Ministry of Foreign Economic Relations Hungarian National Bank and Institute for Economic and Market Research (Budapest) Stefan Kawalec Polish Ministry of Finance, Warsaw Luis A. Riveros World Bank, Macroeconomics Adjustment Elzbieta Kawecka-Wyrzykowska and Growth Division World Economy Research Institute Warsaw School of Economics Roberto R. Rocha World Bank, Macroeconomic Adjustment Miroslav Kerous and Growth Division State Bank of Czechoslovakia Dariusz Rosati Miguel Kiguel Institute for International Trade (Warsaw) World Bank, Macroeconomic Adjustment and Growth Division Jacek Rostowski Polish Ministry of Finance Grzegorz W. Kolodko Warsaw School of Economics, Eugeniusz Rychlewski Research Institute of Finance Warsaw School of Economics Jerzy Kozminski Jeffrey Sachs Polish Council of Ministers Harvard University, Department of Economics Johannes F. Linn 1. T. Singh World Bank, Country Economics Department World Bank, Socialist Economies Reform Unit Nissan Liviatan Andres Solimano Hebrew University of Jerusalem and World Bank, Macroeconomic Adjustment World Bank, Macroeconomic Adjustment and Growth Division and Growth Division Monika Sowa Gabor Oblath World Economy Research Institute Institute for Economic and Warsaw School of Economics Market Research (Budapest) 302 Jack Spilsburry Stanislaw Wellisz U.S. Department of State, Columbia University, New York, and Polish U.S. Embassy (Warsaw) Ministry of Finance Jan Svejnar John Williamson University of Pittsburgh, Department of Economics Institute for International Economics (Washington, D.C.) Vito Tanzi International Monetary Fund, Marian Wojnar Fiscal Affairs Department Foreign Trade Research Institute (Warsaw) Ulrich R.W. Thumm Jozef Zieleniec World Bank, Europe, Middle East and Charles University (Prague) North Africa Department 303 THE WORLD BANK Because the reforms taking place in Central and Eastern Europe are unprecedented in political and economic history, the politicians and economists in the region are without models to guide them in the design and implementation of programs and policies. Much can be learned, however. from the experience of reforms that have been implemented on a smaller scale in developing countries. This book presents suggestions based on the World Bank's experience with adjustment programs and discusses some of the unique features of reform in Central and Eastem Europe. Most of the papers included in the book were presented at the Conference for Adjustrnent Lending: Lessons for Eastern Europe, held October 4, 1990, in Pultusk, Poland. The papers discuss such diverse elements of adjustment as the problems of macroeconomic stabilization, the development and reform of key markets, and privatization. The book will be of interest to economists, government planners and policymakers, sociologists, and students of the peoples of Central and Eastern Europe. Fabrizio Coricelli is an economist in the Country Economics Department of the World Bank. Vittorio Corbo, a consultant to the World Bank, is a professor of economics at Catholic University in Santiago, Chile. Jan Bossak, a consultant to the Bank, is from the World Economy Research Institute (WERI) of the Warsaw School of Economics. WORLD BANK PUBLICATIONS OF RELATED INTEREST The Transformation of Economies in Central and Eastern Europe Issues, Progress, and Prospects Alan H. Gelb and Cheryl W. Gray, Policy and Research Series 17 The Economy of the USSR International Monetary Fund, World Bank, Organisation for Economic Co-operation and Development, and European Bank for Reconstruction and Development Public Enterprise Reform The Lessons of Experience Mary Shirley and John Nellis, EDI Development Studies Financial Reform in Socialist Economies Christine Kessides, Timothy King, Mario Nuti, and Catherine Sokil, editors, EDI Seminar Series An Agricultural Strategy for Poland Bulgaria Crisis and Transition to a Market Economy A World Bank Country Study Czechoslovakia Transition to a Market Economy A World Bank Country Study Poland Economic Managementfor a New Era A World Bank Country Study ISBN 0-8213-1893-4