A ic tq . T ' '' - -= Stg_l ; -; 4 aA ~~~~~~~--.- .. I| _ ~ ~~~~~ ~ . .. - t' EDI DEVELOPMENT STUDIES Corporate Governance in Transitional Economies Insider Control and the Role of Banks Edited by Masahiko Aold Hyung-Ki Kim The World Bank Washington, D. C. Copyright 0 1995 The International Bank for Reconstruction and Development i THE WORLD BANK 1818 H Street, N.W. Wfashington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing February 1995 The Economic Development Institute (EDI) wvas established by the World Bank in 1955 to train officials concemed w.ith developmentplanning. policymaking investment analysis, and proectimpleraentationin mernberdevelopingcountries- Atpresent thesubstanceof the ED1's workemphasizes macroeconomicand sectoral economicpolicyanalysis.Through avarietyof courses, seminars, and workshops, most of which are given overseas in cooperation with local instihttions, the EDI seeks to sharpen analytical skl;ls used in policy analysis and to broaden understandiing of the experience of individual countrie, with economic development Al- though the EDI's publications are designed to support its training activities, mnany are of interest to . ;nuch broader audience. EDI materials, including any findings, interpretations, and conclusions. are entirely ithose of the authors and should notbe attributed in any manner to the World Bank, to its affili:ed org,nizations, or to members of its Board of Executive Directors or the countries they represent Because of the informality o. tis 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 theWorld Bankacceptsno responsibility for errors. Somesourcescited in this book may be informal documents that are not readily available. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisherat the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission tocopy portions fordassroom use is granted through theCopyright ClearanceCenter Inc-, Suite910,272 RosewtoodDrive,Danvers,Massachusetts 01923, U. S. A. The backlist of publications by the World Bank is shown in the annual Inder of Publications, which is available from Distribution Unit, Office of the Publisher. The World Bank. 181$ H Street, N.W., Washington, D.C. 20433, US.A., or from Publications, Banque mondiale, 66, avenue d'lna, 75116 Paris, France. Masahiko Aoki is professor of Japanese studies in Stanford University's department of econornics. Hyung-Ki Kim is chief of the Studies and TrainingDesign Division in the World Bank's Economic Development Institute Library of Congress Cataloging-in-Publication Data Corporate governance in transitional economies: insider control and the role of banks I edited by Masahiko Aoki, Hyung-KI KinL p. an.-(EDI development studies, ISSN 1020-105X) Includes bibliographical references and index. ISBN 0-8213-2990-1 1. Corporate governance-Case studies. 2. Banks and banking-Case studies. 3. Economic development-Case studies. 1. Aoki, Masahiko, 1938- . II. Kim, Hyung-IC, 1936- . Ell International Bankl for Reconstruction and Development IV_ Series. HD2741C776 1994 658.4-dc2O 94-42709 CIP Contents Foreword Vil Acknowledgments iY Overview Masahiko Aoki and Hyung-Ki Kim xi Part I Generic and Comparative Issues: Theory and Policy Implications 1 Controlling Insider Controk Issues of Corporate Governance in T1ransition Economies Masahiko Aoki 3 Setting a Conceptual Framework 4 Emergent Insider ConTrol in the Transition 7 Bank Syndication and Contingent Governance 15 Eclectc Approach-Probably the Best in the Trasition 19 Other Reasons to Develop Banking Institutions in the Transition 21 Banking Reform Needed in the Transition 23 Bibliography 28 2 Political Economy Issues of Ownership Transfonration in Eastern Europe G3&ard Rolancd 31 Political Economy Issues of Ownership Transformation 33 Political Economy Problems Facing an Agenda-Setting Government 35 Rent-Seeking and Ownership Transformation 44 How Relevant Is the Chinese Experience? 47 The Role of Banks 50 Conclusions 54 Bibliography 55 111 iv Contents 3 Corporate Governance in Transition Economies: The Tbeory and Its Policy Implications Erik Berglbf 59 Corporate Govemance in Transition 60 Corporate Govemance and Finance: A Conceptual Framework 65 Some Lessons for Transition Economies S1 What Can We Fxpect from Banks? 83 Some Implications for Banking Reform 90 Bibliography 91 Part LT Coruntry Stidies in Comparative Perspectives 4 Corporate Governance, Bnks andFiscalReforn in Russia John M. Litwack 99 The Russian Environment before 1992 and Insider Control 102 The Period of "Shock Therapy": 1992-94 105 Fiscal Reform and Prospects for Corporate Governance in Russia 111 Bibliography 117 5 Enterprise Governance and Investment Funds in Russian Privatization Noritaka Akamatsu 121 Russia's Privatization Scheme and the Power of Incumbent Management 124 Voucher-Based Mass Privatization and the Emergence of Voucher Investment Funds 132 Structure of VIF Groups and Their Involvement in Enterprise Governance 147 Structure of the Enterprise Governance Board and the Power of Outside Owners 157 Power of Commercial Banks and Foreign Investors and Their Relationship with VIFs 163 Bankruptcy, the Power of Creditors, and the Role of VIFs 171 Concluding Summary 179 Bibliography 181 Content v 6 Evoolution of Commercial Banking in Russia and the Implicationsfor Corporate Governance Elena Belyanova and Ivan Rozinsky 185 Commercial Banking Sector. 1993-lst Quarter of 1994 186 Structre Updated 187 Banks and Corporate Governance in the Postprivatization Period 197 Bibliography 213 7 Reforming Corporate Governance and Finance in China Yingyi Qian 215 Enterprise Reform Policies of the Govermnent 218 Changing Patterns of Governance 220 Chaitging Patterns of Savings and Finance 231 Current Issues in Governance Reform 234 Needed Banking Reforms 245 Concluding Remarks 250 Bibliography 250 8 Centralized Decentralization: Corporate Governance in the German Economic Transition Ernst-Ludwig von Thadden 253 Some Determinants of Economic Structure before 1989 256 Political and Economic Development after 1989 259 Centralized Ownership: The Treuhand 265 External Control of the THA 274 Decentralized Ownership: Postprivatization 279 Condusions 283 Bibliography 287 Part III Relesance and Lessons of the Japanese and German Experiences Introductory Note Masahiko Aoki 295 9 Cleaning up the Balance Sbeets: Japanese Experience in the Post-war Reconstruction Period Takeo Hoshi 303 Bank Reorganization and Recapitalization 305 Corporate Reorganization and Recapitalization 317 vi Contents Case Study 1: Industrial Bank of Japan 328 Case Study 2: Yasuda Bank to Fuji Bank 333 Case Study 3: Ajinomoto 338 Case Study 4: Dissolution of Nippon Steel 341 Case Study 5: NEC 346 Lessons from the Japanese Experience 349 Bibliography 355 10 The Privatization of Ex-Zaibatsu Holding Stocks and the Emergence of Bank-Centered Corporate Groups in Japan Hideaki Miyajima 361 GHQ's Design for Japanese Corporate Governance Structure 364 The Zaibarsu Dissolution: Transfer of Ex-Zaibatsu Stock and the Managerial Revolution 368 Emergence of Institution-Centered Ownership Structure 379 Emergence of Main Bank System as Delegated Monitor 389 Effective Monitoring through Bank-Centered Corporate Groups 395 Bibliography 399 11 Savings Mobilization and Investment Financing during Japan's Postwar Economic Recovery Juro Teranishi 405 Three Characteristcs of Period II 406 Corporate Finance in the Recovery Period 411 Maturity Transformation by Financial Institutions 424 Concluding Remarks 430 Bibliography 432 12 Sbareholder Voting and Corporate Governance. The German Experience and a New Approach Theodor Bauxms and Philipp v. Randow 435 Shareholder Voting and Collective Action 435 A Market for Voting Services 438 Summary 455 Bibliography 455 Index 459 Foreword This book presents the results of a research project on corporate gov- ernance issues in transitional economies from a new perspective based on comparative institutional analysis. The project grew out of a con- cern about three issues: the emerging phenomenon of insider control, the possible role of banks in corporate governance, and the desirability of taking a comparative analytic approach to finding solutions. By examining and comparing individual country studies, policy- makers may be able to avoid simplistic generalizations or theories based on the observation of a single economy. By analyzing the workings of diverse systems, they may also be able to uncover those factors that contribute to or constrain the effectiveness of particular governance structures. This kind of comparative analysis may serve in the social sciences as a kind of proxy for laboratory experimentation. This work was prepared as part of the Economic Development In- stitute's (EDO multiyear Program for the Study of the Japanese Devel- opment Management Experience, which is financed by the Policy and Human Resources Development Trust Fund established at the World Bank by the Government of Japan. The program is managed by EDI's Studies and Training Design Division. Vinod Thomas Director Economic Development Institute The World Bank vii Acknowledgments We would likc to express our appreciation to our colleagues both in and' outside the World Bank for their invaluable contributions to this project. As we set out to identify and articulate the relevant issues during two design and review workshops, many World Bank staff working on corporate gcvernance issues in transitional economies helped to shape this study. In particular, we would like to thank Cheryl Gray, Djordjija Petkoski, Gerhard Pohl, Inderjit Singh, and Alexander Tsapin. We also benefited from the wisdom of the discus- sants at the workshops and the reviewers of the draft manuscript, including John Earle, Millard Long, John McMillan, Gerhard Pohl, and Dimitri Slavnov. Their challenging comments helped to clarify our thoughts and analysis. Elizabeth Goldstein and Bruce Donald served as rapporteurs at the workshops and produced faithful records of complex discussions. We are also grateful to Yushiru Yamamuro, chairman of the Board of the International University of Japan, who shared with us his experiences in dealing with the recapitalization of Japanese banks after World War I. Caroline McEuen provided valuable editorial and production support in preparing the manuscript for publication. Finally, the tWo workshops were skillfuliy organized by Deborah Carvalho at the St-anford Center for Economic Policy Research and by Latifah Alsegaf at the World Bank. We are also grateful to the Stanford Center for Economic Policy Research for cosponsoring this project. ix Overview During the initial phase of transforming socialist planned economies, there was a naive optimism that the transition to a market economy could be readily achieved by the privatiation of the state-owned enterprise, combined with the introduction of the equity market, which would also serve as the market for corpo.rte control. This belief reflected a textbook notion of the capitalist economy, but recommen- dations for economic reforms based on such theological belief have proven to be unrealistic or simplistic. It has been increasingly recog- ni2ed that the isste of the transition or evohution of the system has not been well understood because of economists' preoccupation with the idealized model of the market economy, as well as their naivete regard- ing the effects of the polidcal economy and the broad diversity cf trans- formation issucs, Further, it has come to be recognized that advanced market economies are not necessarily structured according to the text- boolk description. The Anglo-American economy is perhaps the closest to the text- book model of market economies, but the economies of continental Europe and Japan operate very differently from the Anglo-American model. In Japan and Germany, banks, rather than the market for cor- porate control, have played unique roles in the corporate governance structure. The three largest banks in Germany own a large share of the stocks of publidy traded corporations. They also vote on behalf of indiviidual stockholders. They normally sit in the Aufsichtsrats (super- visory boards) of large, publicly traded corporations, wlhich select the management boards. In Japan, the size of stockholdings and the extent of board participation by the banks are, on the surface, not as large as in Germany. Yet it has been observed that the financial difficulties ex- perienced by large corporations have almost surely triggered bank inter- vention and, very often, the shift of management control to the bank. The coexistence of the alternative models for corporate control in the developed countries suggests that their possible "lessons" for the transitional economies are not so obvious. It makes little sense to judge xi xii Oveview the merits of each corporate governance model and its applicability to the transitional economies withoiit noting a country's development stage and the history of its institutions and conventions. In designing corporate governance structures for the transitional economies, econo- mists are required to identify the specific conditions under which each corporate control model (or the combination of diffecrent models) works, the availability of these conditions in the transitional economies, and the most efficient approach to achieve these conditions. This volume presents the results of a research project on corporate governance issues in transitional economics from the perspective of the c.ew comparative institutional analysis. In this approadc, banks and other outside institutions can play an important role in corporate gov- ernance. This contrasts with the traditional approach, where enabling stockholders to exercise control is the goal of efficient corporate gov- ernance. T'he traditional analysis suggests that the problem of creating corporate governance in transitional economies should be addressed by promoting the development of efficient securities markets, nurturing the growth of active and influential stockholders, and legislating corpo- rate laws that would assure stockholders a controlling position on cor- porate boards. But a series of events in transitional economies has made it clear that the matter is not so simple. We cannot ignore the path- dependent, or evolutionary, nature of the transition process. The legacies of socialism, the increased autonomy of managers of state- owned enterprises in the last phase of the communist regime, and the strong political power of the workers in many transitional economies, including Russia, Poland, and China, seem to have left strong con- straints on the privatization process of the succeeding transitional economies and the nature of the evolving corporate control structure. In many transitional economies the phenomena of insider control are becoming evident. By insider control we mean de facto or de jure capture of controlling rights by the managers and the strong representa- tion of their interests in corporate strategic decisionmaking, often in collusion with the workers. Some recent privatization programs have, in effect, legitimized the de facto control of insiders by transferring the majority of the ownership of many large enterprises to incumbent managers and workers. In Russia, by November 1993, 91 percent of the privatized enterprises had adopted the so-called type II privatization program, in which employees (including managers) buy up to 51 per- Overview xiii cent of shares at the pre-inflation book value. In many privatized firms, board meetings and shareholders' meetings are dominated by the man- agement, who also find ways to prevent workers from selling their equity holdings (see Chapter 6). In Poland, widespread asset stripping by state enterprise managers took place when the government started to loosen control over the state-owned enterprises in the late 1980s (Lipton and Sachs 1990). After privatization, insiders, especially the manag, nent, continued to preserve their privileged status, and workers' representatives on the supervisory board continued their activist Solidarity tradition and played an important role in enterprise decisions. In China, evidence from some 4,000 experimental share- holding companies suggests that employee-shareholders have benefited greatly from unfair profit division between themselves and the state, and employee-shareholders have continued to demand excessive wage increases and in-kind benefits. In many collectively owned Chinese enterprises, spontaneous and formal privatization-unauthorized transfer of assets to formal or de facto private ownership-has become common (see Chapter 7). In the orthodox stockholder sovereignty model, the problems of management slack, incompetence, and moral hazard are corrected by outside stockholders through an efficient capital market for corporate valuation and control, as well as institutions such as competitive labor markets for managers and for workers' I2' Jr services. In the transition- al econromies, however, both competitive capital and labor markets are lacking. Managers have es:a6lished strong control within their enter- prises; there is no external agent with the decisive power to dismiss them for poor management performance or moral hazard behavior. Workers have strong attachments to their employing enterprises, which protect their vested interests and jobs. Insider control has evolved out of this relationship. Outsiders would then anticipate substantial agency costs to investing in insider-controlled enterprises. Therefore, the funds necessary for restructuring formerly state-owned enterprises would be difficult to come by from the capital market (insiders' dilema"-see Chapter 3). Faced with a majority or substantial share ownership by the insiders (managers and workers), any external pressure over strategic decisions that might adversely affect the job security and other benefits of the insiders would surely face tremendous resistance. If the securities market is thin and the workers and managers have strong XiVJ OVe7JieWV attachments to their employing firms, market signals of corporate valuation would be garbled and frill of noise. Further, exercising corporate control through the market (the takeover mechanism) would simply not be feasible unless insiders gave up their shares. The design of the corporate governance structure in transitional economies must face this reality. Recognition of the strong tendency toward insider control compels us to search for an alternative external monitoring mechanism over enterprises, which may work even if outsiders are not the dominant stockholders and managers and workers do not voluntarily give up their vested interests and rights. An obvious candidate to implement such a mechanism would be the bank. In the postwar high-growth stage of both the German and Japanese economies, banks maintained contin- uous relationships with corporate clients and occupied unique positions in corporate governance. Their role has been somewhat different from that played by anonymous stockholders. There are phenomena in both economies that would suggest that bank monitoring might be con- sistent with, even complementary to, a certain degree of insider control. In Germany, worker representation on the supervisory board (Aufcicbtsrat) through codetermination is matched by the high concen- tration of voting rights in a few large, universal banks. In Japan, the main bank system, in which investors delegate the monitoring of the enterprise to a single bank, emerged after the demise of the planned war economy and the massive sales of stock of major Zaibatsu corpora- tions to insiders. This bank, called the main bank, selectively inter- vened in the internal management in the event of financial distress. The idea of applying the bank-oriented monitoring system of the German and Japanese model to transitional economies has been pro- posed rather casually in many writings and public discourses. Its genuine applicability to transitional economies is, as yet, far from obvious. The German and Japanese systems evolved in different histori- cal and institutional circumstances and there are major differences between the two. The role of banks in the corporate governance struc- ture in each economy may be complementary to other institutional facets that may not exist in the transitional economies (for example, participatory work organizations, competitive product markets, or neu- tral government regulatory power). Overview xv The old banking institutions inherited by the transitional economics from the planned era suffer from bad debts and bad reputations. Can the spin-offs of the former state banking sector, privatized and recapital- ized, be transformed into a solid banking system? Will the emerging new banks evolve, eventually, into accountable financial intermediaries to which monitoring of enterprises may be delegated and trusted? Al- ternatively, should the nonbank intermediaries, such as mutual funds, be nurtured as potential corporate monitors in spite of the temporary phenomena of insider control, or should the banking sector and the securities markets bc developed in complementary ways? These questions do not seem to be readily answerable by mechani- cally extrapolating the experiences of existing hank-oriented systems to the financial system design of transitional economies. On one hand, we need to carefilly identify the historical conditions prevailing in existing bank-oriented economies that were conducive to the evolution of such systems, and the institutional environments that might hlave been com- plementary to their effective operation. On the other hand, we need to carefully observe current developments in transitional economies and identify conditions that may (or may not) call for, and facilitate the development of, the banks as potentially active players in corporate governance. By pooling rich individual co-untry studies and cross-examining and comparing their implications, we may be able to avoid premature gen- eralizations or theorizing based on the observation of a single economy. By comparing the workings of diverse systems, we may also be able to uncover latent factors that are conducive to, or constrain, the workabil- ity of particular governance structures. Comparative analysis may thus serve as a substitute for the laboratory experiments difficult in the social sciences, although the potential sample nurnbers are extremely limited. A concern with three points-the emergent phenomena of insider control, the possible role of banks in corporate governance, and the de- sirability of a comparative analytic approach-sets the common ground for our collective research, which resulted in this volume. The volume is composed of three parts. Part I contains three chapters that deal with generic and theoretical issues of corporate governance in transitional economies from the perspective of comparative institutional analysis. xvi Ozerview The five chapters comprising Part II deal in depth with various aspects of the corporate governance problem in three transitional economies: Russia, China, and the former German Democratic Republic. Part m contains four contributions dealing with aspects of Japanese and German experiences relevant to the transitional issues. Let us briefly introduce each chapter. In Chapter 1, entitled "Controlling Insider Control: Issues of Corporate Governance in Transition Economies," Masahiko Aoki first reviews the tendency toward insider control in transitional economies and explains why the model of stockholder sovereignty may be ineffec- tive in coping with this problem. He emphasizes complementary rela- tionships between corporate governance structure and other facets of institutional arrangements of the economy, especially internal organiza- tion of the enterprise. He presents a theoretical possibility of con- tingent governance in which the control rights can shift automatically from the insider to the outsider (the lead bank of a consortium) contin- gent on the financial distress of the enterprise, and indicates how the expectation that such an intervention can occur effectively controls various incentive issues for the insider-controlled enterprise. This scheme appears to be more applicable to financially viable insider- controlled firms in the posttransition period rather than as a solution to the transition problem itself. Since the direction of the institutional development of internal organization in transition economies remains fuill of uncertainty, the paper advocates an evolutionary approach to corporate governance design in the transition. It argues that only t:he organic development of an institutional duster will ultimately deter- mine the relative importance of the banking institutions and capital markets in corporate governance. In Chapter 2, "Political Economy Issues of Ownership Transforma- tion in Eastern Europe," G6rard Roland takes up the issue of political constraints on privatization and restructuring arising from the legacies of the socialist system. He makes an important analytical distinction, not necessarily mutually exclusive in practice, between the strong governrnent, with agenda-setting power, and the weak governrment, sus- ceptible to the rent-seeking activities of various groups. Using this framework, he argues for the importance of correctly sequencing the privatization procedure so that the transition is politically irreversible and restructuring is effective. Massive giveaways or distribution of Oveuiew xvii corporate assets may well result in irnsider control (as in Russia and Poland) or the concentration of economic power (as in the Czech Republic), the result of which may be the deterrence of economically viable restructuring and de facto renationalization (rescue by the government). He recognizes advantages of the gradualism in pavatiza- tion pursued by the agenda-setting Chinese government. He concludes with several concrete proposals for banking reform from both agenda- setting and rent-seeking perspectives. Recent academic thoughts on the subject of corporate governance are much more subtle than the single-minded advocacy of stockholder control over corporate governance. Unique roles for debt contracts and bank intermediaries are increasingly recognized. In Chapter 3, entitled 'Corporate Governance in Transition Economies: The Theory and Its Policy Implications," Erik Berglof introduces a valuable conceptual dis- tinction between 'arm's-length" and "control-oriented" financing and argues that these two kinds of financing are actually complementary and mixed in various ways, depending on national conditions. Under the condition of the illiquidity of asset markets and ambiguity about property rights, he predicts that control-oriented financing will play a dominant, albeit not exclusive, role in the transition. He presents a nuanced and balanced argument for positive and gradualist intervention by the government to transform the frgile banking sector into credible financial institutions. In Chapter 4, "Corporate Governance, Banks, and Fiscal Reform in Russia," John Litwack argues that the commitment and trust in econo- mic relationships developed through long-run personal and mutually beneficial ties have a long tradition in Russia, which may favor the development of a bank-oriented financial system as opposed to an American-type arm's-length model. He notes, however, that the current practice of using the commercial banking system for the administration of implicit subsidization distorts the incentives of banks to monitor, while discretionary taxation encourages managers of enterprises to mnimize legal profits by diverting resources to hidden consumption by insiders. He argues that, given the magnitude and scope of the necessary restructuring in Russia, the reduction of subsidies should be gradual and based on optimal sequencing, but for an accountable banking sys- tem to develop, subsidies should not be made through the comnmercial bank network. xviii Overview Chapter 5 by Noritaka Akamatsu, "Enterprise Governance and In- vestment Funds in Russian Privatization," provides detailed, up-to-date institutional descriptions of privatization schemes, Voucher lnvestment Funds (VIFs), corporate and bankmptcy laws, and the emergent indus- trial financial conglomerates in Russia. In conjunction with Chapter 6, it gives a most comprehensive overview of the current Russian situation surrounding enterprises and the financial system. He describes the rivalry between the GKI (the State Property Management Committee), responsible for privatizarion schemes and capital market regulation, and the Minisuy of Finance, responsible for commercial banling regulation. He makes an interesing point that, as the former tries to promote in- vestment banking and restrict commercial banks' activties in capital markets, the mixture of Anglo-American-style capital markets and a German-style banking system may well result in the Japanese model. Chapter 6 by Elena Belyanova and Ivan Rozinsky, "Evolution of Commercial Banking in Russia and Its Implications for Corporate Gov- ernance," is a rare contribution by academicians-cum-practitioners from Russia regarding the recent development of the banking sector in Russia. Based on recent statistical data and questionnaire studies, the chapter provides a valuable glimpse into the relatively unknown work- ings of the newly emergent banks, as well as spin-offs of former stare banks. It depicts the Russian banks as becoming much more important financial institutions in comparison with investment funds, and predicts that the best part of the banking sector, the so-called "hard currency islands," will play a significant role in mid-term investment financing and corporate governance in the future. Because of situational and historical differences from Japan, however, they argue that a German- style system, based on more formal controlling instruments, is more likely to be viable in Russia. From this perspective, they criticize the recent regulation on banks' holding of industrial shares. Chapter 7 by Yingyi Qian, "Reforming Corporate Governance and Finance in China," provides a comprehensive description and analytical discussion of the most recent corporate governance reforms in China He portrays the paradox of Chinese reform: that the communist leader- ship was able to delegate more decisionmaking authority to managers precisely because of their retention of control over personnel appoint- ments, and thus control of managers' career incentives, but that this increased delegation leads to a tendency toward de facto insider control. Overview xix He compares proposals for reforming corporate governance in China within the political constraints of gradualist privatization. He concludes that reforming corporate governance should not follow a single model in China, although a sound banking system would certainly constitute an important element of any reform to counterbalance insider control. From this perspective he outlines a proposal to spin off the vast network of the existing central bank as multiple regional commercial banks. Chapter 8 by Ernst-Ludwig von Thadden, "Centralized Decentral- ization: Corporate Governance in the German Economic Transition," describes and assesses the privatization process mediated by the Treuhandanstic, which had privatized about 18,000 businesses or parts of businesses by the end of March 1994. More than half of these new entities were purchased by western German firms in a manner analo- gous to the working of takeovers in competitive capital markets. The presence of strong centralized control by the Treuhandanstalt prior to privatizaion was instrumental in curbing the insider control problem. The chapter argues that the complete independence of, and institutional time limit imposed on, the activity of the Treuhandastalt (it will cease its operations at the end of 1994 by law) and the career concerns of its managers effectively precluded any long-term gains possible from collu- sion and lobbying by interest groups. Another interesting point. is that West German banks did not play any significant role in the proision of risky resructuring finance to Treubandanstalt firms prior to privat- nation. That this is the case, even under the exceptionally favorable instutional environment of East Germany, suggests that the role of banks may be limited in the transition process, if not in posttransition corporate govemance. Part III turns to the experience ofJapan and Germany. After a brief introductory note concerning broad historical contexts for the emer- gence of a bank-oriented system in Japan, Chapter 9 by Takeo Hoshi, "Cleaning Up the Balance Sheets: Japanese Experience in the Postwar Reconstruction Period," describes how Japanese banks and enterprises resolved a serious postwar, economywide insolvency problem caused by the repudiation of wartime compensation by the government. His care- ful case traces how the bad loans of banks and losses of enterprises were cleaned up without interfering with ongoing business through the separation of old and new accounts, the cancellation of debts and xr OvenLww capital in the old accounts against losses (bad loans), and the subsequent recapitalization through new equity issues. Because the balance sheet cleanl-up was performed in dose cooperation between enterprises and single partner banks, sometimes against the interests of workers, the reorganization process had a significant impact on the subsequent devel- opment of the corporate go-vernance structure. Since the insolvency and recapitalization problem is an important issue in today's transitional economies, this chapter may be very suggestive and relevant for policymakers. In Chapter 10, "The Privatization of Ex-Zaibatsu holding Stocks and the Emergence of Bank-Centered Corporate Groups in Japan," Hideaki Miyajima utilizes hitherto unused original documents of the privatization following Zaibatsa dissolution in the postwar period of Japan and describes its consequences. As in the current privatization process in Russia, the initial objective of the occuparion army was the democratization of stockholding, supplemented by employee owner- ship. The subsequent slump of the stock market and the managerial effort to fend off outsiders' control eventually led to the formation of bank-centered cross-holding among related enterprises. He presents some empirical evidence to indicate that the main bank, which had a multidimensional, close relationship with the client enterprise, fanctioned as an effective monitoring device to counteract insider control in the reconstruction period. In Chapter 11, "Savings Mobilization and Investment Financing during Japan's Postwar Economic Recovery,' Juro Teranishi discusses how the banking sector evolved as an effective intermediary, channeling savings into investible hinds in the high-growth period after the mid- 1950s. He argues that the interim period (1950-55) berween the immediate postwar period (as dealt with by Hoshi) and the high- growth period was characternzed by a low share of investment to GNP and a heavy reliance by large enterprises on internal financing. Using nrch statistical data, he shows that the establishment of an effective system of maturity transformation of short-term savings to long-term financing was instrumental and responsible for ushering in the period of high growth. He analyzes the workings of this system, in which rationing of corporate bonds to the banks and the accommodating supply of Bank of Japan credits to the banks at below market rates Overview ni played very important roles. From this analysis he opposes the view that the postwar system is a simple inheritance of the wartime system. Chapter 12, "Shareholder Voting and Corporate Governance: The German Experience and a New Approach,' is by corporate and bank- ing law scholars in Germany, Theodor Baums and Philipp v. Randow. They describe the German experience, in which a substantial propor- tion of voting rights in large corporations are exercised by leading banks on the basis of their own shares, their custodial shares, and shares held by investment companies that are bank subsidiaries. They point to the danger of conflict of interest inherent in such a system and propose an institutional solution-the creation of a specialized proxy agent. The conflict may arise between the banlk's interest as the representative of value-maximizing small shareholders and the bank managers' own interests in expanding their business with corporations through underwriting syndicates, lending, and so forth. As Akamatsu indicates in Chapter 5, there is already a similar potential conflict in Russia and other Eastern European economies, where banks also con- trol investment funds through subsidiaries, and this analysis of the German experience may be relevant and instructive. These introductory paragraphs only touch the surface of the analyses and presentations of empirical evidence in what is a set of deep and richly nuanced studies of the issues. Without flinching from too bold a generalization, we may succincdy sunmmarize the consensus view drawn from the project as folows: * Sole reliance on the stockholder control model of corporate governance would be of limited merit in transitional economics because of the evolutionary tendency toward strong insider control. * The comparative analysis indicates that, while the tendency toward insider control is generic in transitional economies, its degree and scope varies across economies, depending upon evo- lutionary and institutional factors (the strength of the govern- ment authority before and after the demise of communist control, the development stage and the historical legacy of the economy, and the political-economic processes involving many interest groups during the transition, to name but a few influ- xxii Overview ences). Therefore, institutional responses to insider control should also be diverse. At least on a theoretical level, however, strong banking institutions can be considered effective for controlling insider control, and public policies encouraging such developments need to be recommended. But the effectiveness of the banking sector may depend upon the simultaneous devel- opment of other complementary institutions. Because there is much uncertainty in this regard, the banking system and the capital markets should not be taken as alternative choices in the transition- Although sound banking institutions can be designed to play an effective monitoring role in the corporate governance structure of viable insider-controlled firms, the role of banks in the tra- sition may be limited because of their undercapitalization, the low level of monitoring capacity, the legacy of soft budgeting, and so forth. Therefore, government subsides may need to be continued in the restructuring process, but they should be separated from the commercial banking sector for its sound development. Many subtle lessons can be drawn for corporate governance and financial system design in transitional economies from the expe- niences of Japan-particularly in how insider control can be controlled, how banks are to be motivated to monitor, how bad debts of banks and enterprises are to be resolved through their recapitalization at the time of restructuring, and how the banking institutions are to be nurtured as an effective vehicle for transforming saving into long-term financing. Obviously, however, the direct transplantation of the Japanese system is out ofthe question. The developmental, historical, institutional, and international conditions are different in many important respects between postwar Japan and the current transitional economies. Masahiko Aoki Hyung-Ki Kim Overview =iii Reference Lipton, David, and Jeffrey Sachs. 1990. "Privatization in Eastern Europe: The Case of Poland.' Brookings Papers on Economic Accivites 2: 293-341. Part I Generic and Comparative Issues: T'heory and Policy Implications I Controlling Insider Control: Issues of Corporate Governance in Transition Economies Masabiko Aoki This chapter identifies and discusses some fundamental issues of cor- porate governance in the transition economy. In the first part we present an overview of the generic tendency toward insider control in transitional economies. By insider control, we mean the capture of sub- stantial control rights by the manager or the workers of a formerly state-owned enterprise (SOE) in the process of its corporatization. There are variations in the degree and scope of insider control across transitional economies, depending on national conditions. The tendency is generic, however, in the sense that it is an evolutionary outcome of communist legacies. We argue that the mechanical application of the neoclassical model of stockholder sovereignty for corporate governance In writing this chapter, I benefited from comments and discussions by Erik Berglbf, John Earle, Hyung-Ki Kim, Gerard Roland, Dimitri Slavnov, Yingyi Qian, and other participants in the project on "Corporate Governance in Transition Economies: Insider Control and the Role of Bank" sponsored by Economic Development Institute of the World Bank and the Center for Eco- nomic Policy Research at Stanford University. Bruce Donnell provided research assistance. 3 4 Corporate Govemance in Transitional E conomnies design in the transition is not effective in coping with the insider control problem; worse, it may even prolong the transition process. The second part of the chapter shows that there can be an alterna- tive model of corporate governance to monitor and control incentive issues unique to the insider-controlled enterprise. The essential idea is to rely on the development of banking institutions that can selec- tively intervene in the insider-controlled enterprise at the time of financial distress. The model considers the problem of how the bank is motivated to i11onitor the insider-controlled enterprise while diversi- fying lending risk to some extent. The second part of the chapter is purely theoretical; the precondi- tions assumed for the workability of the model do not seem to exist in transition economies. The fundamental question of which model is superior-the model of stockholder sovereignty with competitive capital markets or the model of the governance in which control riglhts shift from the insider to the outsider (the bank), contingent on the financial state of the enterprise-cannot be answered in isolation from other institutional characteristics of the economy, particularly the mode of internal organization of enterprises. We argue that the future course of the transition economy in this respect is too uncertain to predict. Therefore, we advocate an eclectic approach to corporate governance in the transition, including the simultaneous development of the capital markets and banking institutions. They are presumably complementary in their roles in the development of sound corporate governance in transitional economies. It seems certain, however, that sole reliance on the neoclassical model of stockholder sovereignty will be untenable. With this in mind, the last part of the chapter discusses banking reform in transition economies. Setting a Conceptual Framework In order to discuss issues of corporate governance in transitional economies, let us first make clear what we mean by the transition: transition from which regime to which regime? In orthodox neoclas- sical economics, the latter should be something close to the neoclassical ideal of a regime of perfectly competitive markets. If all marketable factors of production are valued in competitive markets, the allocative Controlling Insider Control .5 efficiency of the economy would be assured by the maximization of re- sidual after payments to those marketable factors. If the daimant of residual is identified with the stockholders, who do not gain any bene- fits other than the residual, they would be unanimously interested in maximizing the stock value of enterprises as reflecting the discounted sum of expected future flow of residual. Therefore, what is needed is to value enterprises competitively and effectively correct the manage- ment of enterprises if their values are lowered. The corporate gover- nance structure that assures the sovereignty of the stockholders, combined with the competitive stock market, provides the necessary and sufficient institutional framework for that purpose. The task of the transition is to jump as quickly as possible to the regime in which such a framework prevails. This neoclassical paradigm is crystal clear in its logic and useful- for some purposes, but in our opinion its mechanical application to transitional policy may not always yield good results because of its incompatibility with historical constraints. We want to propose more practical and indusive definitions of the transition and the posttransi- tion that allow for more diverse approaches. We believe it important to recognize two central issues in considering corporate governance. First, the conditions inherited from the communist regime and those extant at the outset of the transition constrain the feasible options for corporate governance design in the transition process both political- ly and economically. Second, the neoclassical stockholder-sovereignty model of corporate govemance may not be the only efficient solution. Any corporate governance structure may be in complementary relation- ships with other institutional arrangements of the economy-such as the internal work organization of the enterprise, the labor market and financial market institutions, and so forth-and the performance of a governance structure cannot be judged independently of how those other institutions are arranged. More specfically, the stockholder- sovereignty model can be efficient when it is surrounded by a cluster of complementary institutions of a particular kind, such as the hierar- chical work organization and competitive labor and capital markets. We cannot preclude the possibility that for another cduster of institu- tions including the team-oriented work organization, another type of corporate govern-ance could be more efficient. Recent results of compar- ative institutional analysis indicate that these two systems may not be 6 Corporate Governance in Transitional Economies efficiency-rankable independent of technological parameters of the economy. Accordingly, the transition path could also be diverse. We will elaborate these points gradually. Anticipating the diversity of institutional arrangements, we begin with a less specific definition of the time-line, composed of the commu- nist regime, the posttransition regime, and the intermediate transition process. Because our immediate concern is about corporate governance, we deal only with ownership and management of enterprises as defin- ing factors. First, we refer to the communist regime as the period when the following conditions prevail: * (C1) All enterprises are owned by the state, and their continu- ity is at the discretion of the state. * (C2) The management (directors) of enterprises are appointed by the state organ controlled by the Communist Party. The transition process is defined as the time period characterized by the following conditions. - (T1) All enterprises are transformed into corporations (corpo- ratization or "commercialization"), but their ownership struc- tures are in the process of being defined. * (T2) The state has lost the discretionary power to appoint and dismiss the management of enterprises, yet no definite power to do so has emerged. Finally, the posttransition regime is defined as the one satisfying the following conditions. * (Pt) The corporate governance structure has been well defined for each enterprise and the share ownership structure has be- come stable in a statistical sense. * (P2) The management of enterprises is chosen through due process, defined by the corporate law. There is a credible mechanism operating to replace poorly performing manage- ment. This time-line construction is purely conceptual and its mechanical application to any economy may entail some classification problems. For instance, Poland introduced the State Enterprise Law in 1981, while the Communist Party stil held political control. It enabled the Controlling Insider Control 7 workers to appoint the managing director of enterprises through demo- cratically elected workers' councils. At the same time, the state retained the power to create and liquidate enterprises. In China enterprises are now being corporatized, and varied ownership structures are being tried (see the chapter by Qian in this volume). The selection of the manage- ment, however, is still placed under the personnel administration of the Communist Party. According to the above definitions, we cannot say whether or not Poland in 1981 and China today are in the transition process. Ad hoc and somewhat inconsistent it may be, but we say that both Poland after 1981 and China today are in the transition process. We expect that the maass corporatizaion of enterprises in China will eventually lead to depoliticization of management appointments. Note that we do not necessarily identify the transition with the privatization of enterprises. The majoriry stocks of a significant num- ber of corporatited enterprises may remain owned by the state (as in Hungary). Also note that we do not include the market control of cor- porate governance as a condition for the arrival of the posttransition regime (for example, we do not observe it in Japan). We intend to include in that regime cases in which the majority or minority blocks of the stock of enterprises are stably owned by the state or insiders (workers, managers, enterprise pension funds, and the like) when there is a credible mechanism to punish poor management, if not through takeover. We deem that the existence of such a mechanism is impera- tive for an economic system to be viable without discretionary stte intervention. One of the purposes of this chapter is to inquire: If insider control is likely to evolve in many enterprises in the transition process because of conditions prevailing at the demise of the communist regime, what public policies would be desirable and politically feasible for transiting to an efficient posttransition regime? This inquiry will inevitably entail another important question: Is it possible for a posttransition regime to include a significant number of enterprises efficiently controlled by insiders? Emergent Insider Control in the Transition Insider control (either by the manager or the worker) appears to be a generic potential in the transition process, evolving out of inheritances S Corporate Governance in Transitional Economnies of the communist regime. \Vhen the stagnation of communist regimes deepened in the 1970s and 1980s in Central and Eastern Europe, central planning bureaucrats tried to cope with the problem by relinquishing most of the planning instruments to the management of SOEs. The di- rectors built up an irreversible jurisdictional authority within their own SOEs. The gradual retreat of the central planning authority ended with its sudden dismantling. The managers of the SOEs who had already carved out substantial controlling rights from the planning apparatus further enhanced their rights in the vacuum created by the collapse of the communist state. There seems to be nobody who has obvious legal or political power to dismiss the managers of ex-SOEs while they have the support of their workers. The other quality of the communist regime that constrained the worker's freedom of job choice was their de facto job security. They were provided with medicare, child care, leisure facilities, housing, paisions, and so forth by the employing SOBs or the state. Workers had strong stakes at the employing enterprise. After the coUapse of commulnism and the end of its "egalitarian" ideology, the workers were threatened by the possibility of losing those vested interests. Their fear may be greater the more uncertain the outcome of corporatization of their enterprises. Their possible opposition to massive privatization may have to be overcome by virtually giving them a substantial portion of enterprise assets. Needless to say, the actualization of the potential of insider control varies across economies. We define insider control as a majority or substantial block-holding by the insiders in the case of privatization, or strong assertion of insider interests in strategic enterprise decision- making when the enterprises remnain owned by the state. Among possi- ble factors conditioning the extent of insider control, the most impor- tant ones are the degree of management autonomy and the workers' strength against communist control in the final stage of the communist regime, and the political autonomy of the privatization authority against the various interest groups in the transition process. At one end of this continuum, there is Poland. As already noted, even before the fall of the communist regime, the workers councils, composed of fifteen members elected by the employees, had attained a powerful position analogous to the board of directors in capitalist corporations, indcluding the right to appoint the director and approve Controlling fnsider Control 9 the annual plans of the enterprise. Once the transiton phase began, the workers quickly moved to capture control of the assets of the enter- prises before any market-based privatization plan was to be put into effect. The most common form of state property transformation worked as follows. Rather than corporatizing, the viable SOE was U1iq- uidated' and a new company, in which the majority of the workers of the liquidated SOE became stockholders, leased or bought the assets. The much-publicized massive privatization through state-sponsored in- vestment funds, an artifact of the neodlassical dogma of stockholder sovereignty, has thus been virtually defeated by the coalition of workers and managers of the better enterprises. Russia is a case of strong manager control. The director of the SOE, who had already built a virtually autonomous empire in the com- munist regime, became almost invincible after the dismantling of the party and its planning apparatus. For any privatization plan to be implemented, it would have had to recognize this de facto control power of the director (see the chapter by Litwack in this volume). The State Committee on Property Administration (GKI), which was charged with mass privatization, has become the most politically suc- cessful reform authority in Russia through its generous accommodation of insiders' interests. The details of the scheme are described in the chapter by Akamatsn in this volume. Simply put, according to the scheme the privatizaton of the SOE proceeds through three stages. In the first stage, mandated by a Yeltsin decree of July 1991, the SOEs were to be corporatized and made legally autonomous entities, although all the shares were held initially by the state (the Federal Property Foundation) and adminis- tered by GKI. In the second stage, the insiders (the workers) chose an option for their privatization benefits from three variants specified by GMI, and the local committee of GKI approved an adopted plan. At this stage, the managers and the workers obtained a large share free-of- charge, or a majority share purchased at discounted value, depeo.ding on the adopted variant. In the third stage, the remaining shares were auctioned for vouchers that had been given to every citizen of Russia, sold in a package to investment tenders, and kept by state control for the next several years. The full implications of the scheme have not been worked out yet, but so far the insiders have overwhelmingly selected an option to guar- 10 Corporate Governance in Transitional Economies antee them a majority share-that is, the option that gives managers and workers together individual ownership of 51 percent of the equity at a low purchase price (at 1.7 times the July 1992 book value of assets). The managers can also increase their shares by purchasing vouchers in the market or by buying back shares from their own workers and markets (the workers aie now given incenrives to sell their shares tax- free). At the same time, investment funds that participate in voucher auctioning are limited initially in their ownership in one privatized SOE to 10 percent (raised to 25 percent after January 1994). The board of directors of the newly privatized SOE, before the first meeting of shareholders (which has to take place within one year after privatiza- tion), is composed exclusively of the general manager and worker representatives, except for representatives of the local GMI and the Property Foundation. As a result, the insiders, particularly the managers, have built solid controlling power in their enterprises.' At the other end of spectrum _s the former German Democratic Republic (Ea Germany), whose privatization process under the cen- tralized privatization authority, the Treuhandnstaldt /;HA), is de- scribed in detail in die chapter by von Thadden in this volume. Even ir East Germany, however, asset stripping by the insiders was an immi- nent danger at the rime of the demise of the communist regime. The only factor that prevented the subsequent development of insider control was the authority given to the TEA. von Thadden sho-ws that the institutional commitment to complere privatization by the end of 1994 prevented the THA from being susceptible to influence by the insiders. The recruitment of professionally capable TIA managers from western Germany is also a positive factor unique to East Germany. The 1. In a decree issued by Yeltsin on December 24, 1993 (N2284), some measures were introduced to curb the tendency toward strong insider control. Requirements for worker payment for benefit options were made somewhat stringent (for example, the amount of the first installment for stocks bought with a discount was increased). It was also stipulated that the number of representatives of employee-stockholders may not be more than one-third of the board of directors, and that managers or workers may not sik on the board representing the interests of the state. That these provisions were thought necessary may suggest that such practices had been widespread, without check. Controlling Insider Control 11 privatization of the SOEs was to be completed predominatmly through the partial or whole acquisition of assets by West German (former Federal Republic of Germany) corporations. la that sense, the privat- ization in East Germany may be said to be comparable to a takeover in capital markets, although it was mediated by the centralized pri- vatization agency. Even in this case, however, the end result of the transition would be the absorption of the SOEs into the West German corporate governance structure, which is different from the neodcasical model of stockholder sovereignty. This governance structure is charac- terized by insider (worker) participation in the supervisory councils through the legal requirement of codetermination. The Czech Republic and Hungary provide intermediate cases. The insiders were weaker in the era of the communist regime in comparison with Poland or Russia, and the political power of the state (thie pn- vatizaton agency) in the transition process is weaker in comparison with East Germany. As a result, the tendency toward insider control has not been clearly resolved. Privatization in the Czech Republic is widely viewed as an ideal example of an approximation of the neoclas- sical model of outside stockholders' control through "vouchere privat- ization. The marter does nor seem to be so simple, however. The pri- vatization process is initiated with the decentralized submissions of a -privatization project," which can be done by anybody. The Ministry of Privatization has the centrlzed power to select a project. The Ministry has a political preference for projects including the compet- itive bidding of shares for vouchers. Nevertheless, project proposals for direcL sales of assets to a new company formed by a group of insiders are also possible. According to data from the Ministry of Privatization, and quoted in Frydman, Rapaczynski, and Earle (1993, p. 84), only 53 percent of the total book value of privatized enterprises have gone to vouchers. The fist preference of managers who were able to submit the most informative plans is said to be buyout (Frydman, Rapaczynski, and Earle 1993 p. 81). The tendency toward insider control surely exists, but has been moderated by the centralization of project selection. In Hungary, a self-management system similar to the 1981 Polish scheme was introduced in 1934 (Law on Enterprise Councils), although the relative authority of managers in relation of the workers was stronger. The free-market-oriented postcommunist government adopted 12 Corporate Govrance in Transitional Economies a decentralized privatization scheme that gave the initiative to privatize to the enterprise councils, subject to approval of the State Property Agency (SPA). In contrast to the semicentralized Czech approach, this scheme seems to have proviided more room for maneuvering by the managers to retain control and to fend off 'outsider? intervention. Privatized enterprises tend to be cross-owned by other enterprises, baks, and the state (SPA). Unfornmately, because of the unavailability of data, the extent of cross-holding is not precisely known, but some- thing simnilar to the corporate grouping in Japan may be emergent. Thus, although there is a variation in its degree, the tendency toward insider control is manifested everywhere in Eastern and Central Europe except for the newly emergent entrepreneurial enterprises and joint ventures with foreign corporations. This is an evolutionary out- come of legacies of the communist regime, which can be moderated only by a strong privatization agency. But an attempt to introduce out- side stockholder control does not seem to effectively counteract this tide. Privatization of SOEs is ex ante constrained by the legacy of socialism, and in most it is cases ex post constrained by the weakness of the privatization agency in relation to the inside interest groups (see the chapter by Roland in this volume). This lesson may be instructive for China, which has now begun ex- perimenting with various corporate governance structures. Even in China, where the Communist Party has retained solid control over the personnel selection and dismissal of management of the SOEs, the evolutionary tendency toward insider control is not unknown. The chapter by Qian illustrates this point in detail. One cornmonly ob- served method for enhancing insider control is the spin-off of sub- sidiaries by the management of the SOEs and the leasing or sale of assets to these subsidiaries, created for this purpose. The "non-state- owned" enterprises thus created, together with smaller township and village enterprises, constitute the essential carriers of vital entre- preneurial initiatives in present-day China This state asset stripping by insiders is often regarded as illegitimate by citizens and, unless placed under a transparent due process, this 'privatization process may provoke political backlash. The mechanical application of the neo- classical paradigm, however, such as voucher privatization by invest- ment funds, does not seem to be an alternative solution. We next pre- sent theoretical reasons for this. Controlling Insider Control 13 Inadeqttacy of the Investment Fund Scbeme for the Transition Many economists have argued that the creation of investment funds (IF) that hold a substantial block of shares of the privatized enterprise may serve as an effective external check on insider control. The hope is that the IF would be interested in capital gains made possible by efficiency-enhancing restructuring, while at the same time being capable of exercising sufficient pressure or control over the management to implement the restructuring (see the chapter by Akamatsu for a detailed description of the IF in Russia and a comparison with those in the Czech Republic). Nevertheless, it generally seems to be the case that the effectiveness of IFs in external monitoring has been limited. First of all, facing a substantial insider holding of shares, even a block of shares held by the IF may not be sufficient for effectively controlling the privatized SQE (as in Russia). To advance the case for the active role of the IF in corporate governance, a well-known proposition by Sbleifer-Vishny, which points to the importance of a blockholder, is often mentioned. What the proposition asserts, however, is that the existence of a block- holder is "necessary" for overcoming free-riding by small, passive share- holders in disciplining inefficient management by a takeover. Differ- ing from the presumption of the Shleifer-Vishny model, the Russian situation is characterized by the existence of a large body of inside shareholders. How can the minority IF overcome this imbalance of power at a critical moment when the possible dismissal of the inside managers (and massive labor shearing) should be placed on the agenda) Second, the IFs were formed as privatization intermediaries and funded primarily by vouchers that were entrusted by investors or pur- chased by the IF. They are therefore under pressure to realize reason- able dividends for investors. If the markets for shares develop, however, they paradoxicaly may not be interested in monitoring and restruc- turing individual enterprises. By investing in the market index, like funds in developed, securities-oriented economies, the IF would be able to perform at least as well as the market (this possibility is also emphasized by Phelps and others 1993). To respond to this concern, it has been proposed that the range of portfolio selection of the IF be resticted. But such a move would be inconsistent with the IF function of expected profit nmaximization, and it would effectively transform the 14 Corporate Governance in Tasitional Economies IF into a holding company. The next issue raises a question about the ability of the IF to function as a holding company. Third, the privatized SOE may be in desperate need of additional funding for restructuring, but the IF, as a share redistributing inter- mediary, may not be able to readily mobilize financial resources to meet such needs. Even if the IF could mobilize new financial resources, the insider majority conmrol may imply tremendous agency costs for equity financing. The management and the workers may be interested in consuming on-the-job potential residual before it is distributed as dividends. The IF may be ableto mediate bank loans because the IF is often controlled by a holding company that also controls a bank (as in Russia) or is owned by a bank (as in the Czech Republic). In this case, however, the conflict of interest issue needs to be addressed (see the chapter by Akamatsu). For example, the assets of the IF may have been heavily invested in a failing enterprise and the holding company/bank may be interested in funneling funds to salvage it at the depositors' risk. These discussions are not intended to deny any role for the IF in the govemance structure in the transition process. On the contrary, it may be an indispensable institutional component in counteracting the ill-effects of insider control. The point is to argue that the IF alone may not effectively resolve the problem of corporate governance design in the transition process posed by the evolutionary tendency toward insider control. To attempt to rely solely on the IF may actually prolong the transitional process by encouraging inefficient influence of insiders to reduce outside intervention. The management may sabotage restructuring by colluding with the workers to fend off the outsider intervention. Further, public policy that may be needed to foster the development of other institutions, such as the banks, may lag behincL The transition economy has to face the evolutionary tendency toward insider control, and to do so an application of the abstract neoclassical model or the straightforward transpiant of the Ango- American model seems to be of limited value. In the next section, we propose an alternaive model of external control of the insider control enterprise based on the idea of "selective intervention.' Following the presentation of the theoretical model, we discuss whether it suggests any public policy approach toward the insider control problem in the transition process. Controlling Insider Control 15 Bank Syndication and Contingent Governance The insider-controlled enterprise may have unique incentive problems, even if it is potentially productive. The management may try to borrow to build an empire (i the case of management control), to spend on nonproductive projects to enhance worker benefits, or to construct plants excessively equipped with machines to increase per capita outputs, while staying away from risky entrepreneurial projects (adverse selection problems). The workers may shirk to free ride on each others' efforts when team work is involved (moral hazard prob- lem). The insiders may have incentives to consume as much of the revenue of the enterprise as possible on the job or in the form of supracompetitive compensation before repayrments to lenders or divi- dend payments to shareholders are made. Poor management may be tolerated out of collegial compassion. To cope with the possible inefficiencies of such incentive problems, some external agents must play active roles in monitoring the insider- controlled enterprises. To filtate the following discussion, it is useful to distinguish conceptually three phases of investors' monitoring in reference to the timing of investment: ex ante monitoring, in which potential new projects and/or new dients are evaluated to cope with the problem of adverse selection; interim (ongoin) monitoring, to uncover moral hazard problems arising from the divergence of interests between outside investors and the insiders, as well as free-riding among the insiders; and, fmally, ex post monitoring, to verify the true financial state of the enterprise, to assure the (re)payment of debts or dividends, and to punish the management in the event of a failure to do so. This section presents a purely theoretical model of corporate gover- nance that resolves incentive problems of insider control by integrating the three stages of monitoring by a single bank, while other investors can diversify risk. The model is derived by a modification of the bank- ruptcy procedure proposed by Bebchuck (1988) and Aghion, Hart, and Moore (1992), which attempts to strike a balance between the merits of equity and debt contracts as controlling instruments. The novelty of the model presented below is to explicitly consider the incentives of the monitor-in this case, a bank-to monitor the insider-controlled enter- prise in an integrative way, while preserving the essential feature of their model. 16 Corporate Governance itt Transitional Economies Suppose that when the viable insider-controlled enterprise is in need of external long-term investment funds, a bank that has had a long-term relationship with the enterprise organizes a loan syndicate with many other banks. This lead bank (LB) may own a minority share of the borrowing enterprise up to a certain limit (say, 5 percent). The LB is assumed to perform the commercial banking function by running the major payment system accounts as well as the deposit accounts of the borrowing enterprise. These two attributes may provide the infor- mation advantage necessary for the bank to be an LB. The question is how this advantage can be utilized for the reduction of agency costs of the external financing and monitoring costs, rather than allowing the LB to exploit its private benefits at the cost of other banks and investors. Suppose that the LB is limited to provide only a minority share, say 20 percent, of the syndicate loans, but the LB must guarantee the repayment of the claims of other member banks (this stringent require- ment will be relaxed later). This imposes a heavy responsibility on the LB. In return, however, the LB may charge a syndicate management premium, as well as enjoying the benefits of running the payment set- tlement and deposit accounts of the borrowing enterprise. Meanwhile, the IF and other investors may be actve in share markets, evaluating the performance of the enterprise through trading on the basis of their own interim monitoring. When the enterprise becomes unable to meet repayment obliga- iions, the LB is obliged to buy the defaulted claims of syndicate mem- ber banks. The LB then completely writes off these debts and converts them into new equity. The LB either auctions off the new equity rights to reorganization specialists or holds themn for a specified period of rime (for instance, three years). In the latter case, the LB is engaged in re- structuring the defaulting enterprise by replacing the managers, laying off workers, liquidating some assets, and so forth. If the insiders refuse to cooperate, the LB could threaten to invoke the liquidation proce- dure. After restructuring within the said period, the LB may sell the amount of shares beyond the normal limit (5 percent, for example) for possible capital gains from restructuring. In either case, the restruc- tured enterprise would be transformed into an outsider-controlled enter- prise, while the insiders are penalized for debt default by the loss of their share values, and possibly the loss of employment continuation Controlling Insider Control 17 values as well. If there is no prospect for capital gains from restructur- ing, the LB may decide to liquidate the enterprise. In this case the un- covered value of debts would be born by the LB. The scheme has merit. First, in contrast to the Aghion-Hart-Mo ore model, thie postbankruptcy procedure is administered by the LB rather than by the court, which may lack expertise in ex post management of the bankrupt enterprise. The LB is clearly advantaged in information useful for ex post management, but prcvented from using it at the ex- pense of other creditors' interests because of its repayment guarantee. in our scheme the LB is also responsible for ex ante mon:toring to cope with adverse selection of the borrowing enterprise and for interim monitoring to control its moral hazard. The LB would be motivated to eamestly perform ex ante and interim monitoring in order to avoid the heavy costs of liquidation and/or restructuring arising from debt guar- antees for other member banks. Second, generaly speaking, from the point of view of the bank, there are tradeoffs between the diversification of lending risk and incen- tives to moriitor. An arm's-length relationship between the bank and the borrower may allow the bank to diversify risk, provided that risk is distributed independentiy of bank action. But risk diversification may dilute the incentives of the bank to monitor the enterprise ex ante and interim. At the same time, the exclusive lending relationship will not only expose the bank to idiosyncratic risk, but may also dilute its com- mitment to cx post monitoring because once lending is made, the con- tinuation of the enterprise may become ex post desirable even for the bank. As noted above, in our scheme the LB is certainly motivated to monitor. But what about the risk diversification opportunities of the LB? Does the proposed scheme not amount to the same thing as the LB bearing full risk costs, as in exclusive relational lending? That is, is the LB not exposed to the same degree of idiosyncratic risk as the relational bank? Why then is syndication worth the trouble? Suppose that a sufficient number of qualified banks that have the required monitoring capability exist to allow for workable competition among them (roughly, the number is not too small nor too large). Sup- pose that each of them has a mutually exclusive group of customer en- terprises for which they function as LBs. These banks become ordinary syndicate members for nonaffiliated enterprises for which other banks 1S Corporate Governance in Transitional Economies act as LBs. In other words, therc is "reciprocal delegation of monitor- ing" (Sheard 1994) among the banks. Other minor banks may partici- pate in any syndicate as ordinary members. Responsible monitoring by the LB saves the duplication of monitoring costs, particularly by minor banks. Now let us modify the described scheme in such a way that the group of qualified banks mutually agree to rebate a fraction of the LB's guaranteed repayment to them in case the revenue of the LB from liq- uidation of assets or the prerestructuring auction falls short of a certain level. Such reciprocal arrangements may spread the cost of risk-bearing among qualified banks, while somewhat diluting the incentives for the LB to monitor. The modified syndicate arrangement is a device to strike a balance between the conflicting requirements of risk diversifica- tion and incentive provision. The third advantage of the scheme is that the risk of bad perfor- mance by the borrowing enterprise is not distributed independently of insider and bank actions. The risk may be reduced by more intensive ex ante and interim monitoring of the LB. The requirement of syndi- cation severs qualified banks from exclusive relationships with the borrowing enterprise. This would reduce the hazard of the banks being captured by the interests of the customer enterprises and make them more independent in their judgements at the ecx ante monitoring stage. Fourth, after the initial investment is sunk, the continuation of a bad project might become ex post profitable, if bad debt were written off. In such a situationa if the relationship between the bank and the enterprise is that of the exclusive relational bank, they would be induced to renegotiate. The enterprise's insiders may have strong motives to negotiate for the survival of the enterprise to save the loss of employment continuation values if the labor market is imperfectly competitive. The bank may be induced to accept an insider's conces- sion, while keeping the insider's control intact. Such a prospect would dilute the ex ante and interim incentives of the insiders. In our scheme, the shift of control rights can be automatically triggered by a dtebt- equity swap when the insider-controlled enterprise defaults. Insider control is maintained contingent only on the financial viability of the enterprise. The corporate governance structure implied by the scheme may thus be called contingent. Controlling Insider Control 19 The LB or the reorganizer (aFs, or another enterprise) that acquires the shares in the restructuring auction has incentives to restructure for capital gains. As opposed to the creditor rescue operation, the restruc- turing agent can secure future returns to restructuring costs without fearing the emancipation of the rescued enterprise. Thus, premature liq- uidation (Type I error) may be avoided, wvhile the threat of punishing poorly performing insiders is made credible. This is the essential feature of the Aghion-Hart-Moore model. The fifth major advantage of the scheme is that the contingent gov- ernance structure may have positive incentive effec.ts on the insiders. The contingent governance implies that as long as the insider-controlled enterprise is financially healthy and able to repay its debts without any problem, the insiders remain as residual claimants. If they always remain residual claimants, regardless of the financial state of the enterprise, the moral hazard of free-riding among insiders would be- come a problem, or sheer bad luck might make the insiders lose large asset values (in financial and human capital) in liquidation. These possibilities may be prevented by the participation of the LB, which may restructure the failed yet viable enterprise, but imposes harsh penalties when the postbankruptcy situation is hopeless. if the contingent governance is efficiently designed, the insiders may develop incentives to accumulate the internal financial resources to become autonomous from possible external intervention. The rela- tive autonomy of the enterprise from external loans would in turn stimulate insiders' incentives for greater effort, because the fruit of their effort would accrue to them as residual claimants. Thus, once the contingent governance is put in effect, the virtuous cycle of insider control and enterprise growth may be generated up to a certain thresh- old point. (The possibility of such dynamics, together with the efficien- cy property of the contingent governance structure, is analyzed in Aoki 1994.) Eclectic Approach-Probably the Best in the Transition We have shown a theoretical possibility of a governance structure that can cope with the unique incentive problems arising from insider control. But we have assumed that the insider control enterprise is 20 Corporate Governance in Transitional Ecoonomies viable and that there is a banking sector comprising a sufficient number of banks capable of assuming the heavy responsibility of lead banks. These last two conditions do not hold in the transition process. The pzivatized enterprise may need to be restructured, for which outside financing is needed. Nevertheless, incentive problems of insider control may be rampant and any bank may consider it too risky to assume the responsibility of the LB. First of all, no bank may have either the financial resources to bear the responsibility or the capacity to momitor. The merit of the theoretical exercise in the last section was to show that a corporate governance structure alternative to the neoclassical model of stockholder sovereignty is conceptually possible in the post- transition regime. The contingent governance structure may not be just a passive reaction to the insider control problem. Rather, it may have an active raison d'etre: to facilitate the development of a ream-oriented production organization characterized by lateral cross-functional coordi- natiorn, joint task r2sponsibilities, mutual help, and the like in which worker skills and shared knowledge become specific to the organiza7- tion. According to a recent achievement of comparative institutional analysis, we cannot unequivocally rank the team-onented work organi- zation and the traditional hierarchical organization according to efficiency criteria. The former may perform better in an industry where coordination among tasks is relatively more important because tasks are complementary, while the latter organization may perform better i the industry where flexible reallocation of scarce corporate assets among tasks is important (Aoki 1994b). The kind of financial institution that would be desirable to develop would depend on the prevailing work organization. If workers' skills and shared knowledge become organization-specific, and thus not indi- vidually marketable, stock value maximization may not be consistent with intcrnal and allocative efficiency. This is because the value maxi- mization criteria presupposes that all factors of production other than fixed factors are market valued. If the insiders constitute an immobile factor of production as a team, the enterprise ought to strike a balance between insiders' interests and outsiders' interests. Gamnes between them may no longer be zero-sum, however, and there may be gains from cooperation (Aoki 1984). Insiders' shared rights of control in cor- porate governance (as in West Germany) or outsiders' selective inter- vention through the contingent governance (as in Japan) may facilitate Controlling Insider Control 21 an approximation of the cooperative solution. Thus the bank-oriented financial system may be complementary to the teamn-oriented organiza- tion, while the market-oriented financial system is complementary to the hierarchical organization, as the neoclassical paradigm asserts. In what direction will the evolution of the transitional economies lead? One scenario may be that the hierarchical aspect of the work organization in the SOE would reform itself so that task assignments in the organization are made more on the basis of individual skills. For this direction, the complementary development of the capital market institutions would be necessary, because the efficiency of such an organization can be best valued by residual after competitive payments for individual skills. Another scenario may be that the collegial aspect of the work organization in the SOE would develop into an efficient, team-oriented work organization. For this direction, the development of banking institutions might be complementary. Still other scenarios may be feasible. The transition economy may take advantage of the latecomer being able to develop a hybrid by combining the two types of organizations. Or, if the transition economy fails to develop proper governance and financial institutions, it might be locked in permanent stagnation. Nobody seems to be able to predict with certainty which scenario is the most likely. We may posit that an eclectic approach is an option in the transi- tion process. That is, instead of pursuing solely the possibility of exernal control of the enterprise through the development of capital markets, or that of banking institutions, it is better to foster their simultaneous development in the transition. Only spontaneous develop- ment of organizations through competition would determine the domi- nant system in the posttransition regime. Other Reasons to Develop Banking Institutions in the Transition In the model of syndicate lending presented above, ex ante, interim, and ex post monitoring of the enterprises are all integrated and dele- gated to a single LB. In contrast, in a highly advanced securities- oriented financial system, such as the Anglo-American economy, these three phases of monitoring are dispersed among various intermediaries, 27 Co-DOrate Governance in Transitional Economies information-processing agents, and corporate and legal institutions possessed of different specialized expertise. For example, ex ante monitoring is performed by investment banks for large enterprises, venture capitalists for entrepreneurial start-ups, and commercial banks for smaller firms; interim monitoring is performed by rating firms, commercial banks, funds of various types, marker arbitrageurs, and so forth; ex post monitoring is done by accounting firms, the bankruptcy court and reorganization specialists, takeover raiders, LBO partners, and the like. In general, the transition economy that has evolved from a state of an absence of financial markets initially lacks the accumulation of such diverse monitoring resources. The integrated delegation of the three phases of monitoring to a single bank is a way to economize in the use of scarce monitoring resources. Further, there is a positive reason for the integration of the three phases of monitoring, rather than their decentralization, to bring about better results. In the transition process, ex. ante monitoring of the corporatized SOE, if not that of a new, innovative stan-up, would be unlikely to require highly sophisticated project analysis. The urgent problem with the corporatized SOE, privatized or not, is to restructure itself rather than to initiate new projects at the technological and com- mercial frontier. If this is so, the relevant ex ante monitoring would be more on the organizational capability of the corporatized SOE to absorb, adapt, and improve the excisting organizatonal, engineering, and commercialization know-how. For that, the bank that would maintain a long-run relationship with the borrowing enterprise may be in an informationally advantageous position because it can feed back informa- tion available from interim monitoring to the assessment of relevant organizational capability. It may also be the case that ex ante monitoring and interim momi- toring are complementary to ex post monitoring. Even if the financial state of the enterprise is critically worsened, the IF may be less adept at finding the problem when the accounting methods are not very in- formative and disclosure requirements are lax. Even if they are competent enough to find problems at an early stage, they may encounter resistance in the insiders to yielding control power. In contrast, the bank would mediate daily payment settlements for the customer enterprise as well as roll over short-term loans or discount trade bills necessary for financing working capital. Such operations Controlling Insider Control 23 would give the bank a power similar to that of being able to partially open the books. At the time of automatic transfer of control triggered by debt contracts in the event of repayment default, the bank may utilize knowledge accumulated through interim monitoring to exercise its judgement of whether the enterprise has a chance to survive or would be better served by liquidation. The merits discussed of the integration of the three phases of moni- toring presume that a single bank would credibly commit ex ante to interim and ex post monitoring. Such a commitment is not credible in the highly developed, market-oriented system where debt instruments are easily marketed. The initial investors may get rid of their claims in the market rather than bear bankruptcy or rescue costs ex post if they are in a position to find possible problems with borrowers at an early phase. As already noted, however, the necessity of developing a sound banking sector should not be taken as precluding the simultaneous functioning of financial markets. On the contrary, competitive and informative financial markets can be complementary to bank monitor- ing. Instead of t.e formation of syndicates, the bank may underwrite and guarantee bond issues of the customer enterprise. The price forma- tion of the secarities of the enterprise in the market can compete with the interim monitoring of the bank, pointing out its mistakes or reme- dying possible moral hazard. If the IF develops restructuring expertise in the posttransition regime, it can bidc for the equity that the LB auc- tions off after the debt-equity swap operation. The point is, however, that the role of a sound banking sector, composed of a reasonable num- ber of qudified banks, could not be fully substituted for the financial markets when the magnitude of insider control is substantial. Banking Reform Needed in the Transition Although the case for the development of a sound banking system in the transition process appears to be strong, the current state of banking institutions in transition economies seems to be far from that needed to perform the kind of tasks suggested above. Is there any hope that they will develop the capacity and incentives to do so? What kind of banking regulations are to be instituted in the transition process to en- courage such developments? 24 Corporate Governeance in Transitional Economies In transitional economies, most of the banks are either successors or spin-offs of the former state banks or newly established agent banks of corporatized SOEs. Large commercial banks in Central Europe were created in the last few years with the split of the former state mono- bank into a central bank and a number of commercial banks. The loan portfolio was distributed to the new commercial banks along regional (Poland) or sectoral lines (Czech Republic, Hungary). Most of the deposits are with specialized savings institutions, channeled to com- mercial banks in the form of refinancing credits through central bamks. The spin-offs of the former state banks are still owned by the state and their privatization is under preparation (Poland, Hungary), or their majority ownership has now been privatized (Czech Republic). Private commercial banks have been established recently, yet the former state banks are still dominant in assets. In Russia there are now approximate- ly 1,700 independent commercial banks (see the chapter by Belyanova and Rozinsky for a detailed account of the present-day Russian banking institutions). Among them, about 700 banks, including most of the larger banks, are spin-oifs of the former Soviet specialized banks, which are now primarily owned by former SOEs. For these banks the sets of shareholders and borrowers are the same. The state bank in the centrally planned economy was not an auton- omous financial institution, but an administrative instmment of central- ized planning to control the SOE. As is well known, one of the most important causes of the failure of centrally planned economies -was the soft budget con-straint on the SOB because of the lack of commitment by the state bank not to refinance ex post inefficient projects. Financing the existing SOE became automatic because of the political necessity of maintaining employment (latent insider control problem), the rising bargaining power of managers, and possibly because refinancing made economic sense once the initial investment was sunk. Insolvency cri- teria did not exist in the cornmunist regime, and thus soft credits could not be distinguished from outright giveaways. A possible problem with the spin-offs of former state banks is the continuation of soft credits as a form of inertia. Half of all commercial bank loans ex-tended in 1992 in Russia were in the form of directedi credits fusnded by the Central Bank or the budget and channeled through these banks (World Bank 1993, p. 2). Controlling Insider Control 25 The newly created agent banks may have their own problems. In R:ussia more than 1,000 agent banks have been created from nothing since 1990. They were usually created by enterprises or by groups of enterpnses to manage their cash flows and to perform payment system transactions on their behalf. These banks also make loans, primarily with funds from enterprise deposits and interbank markets, as well as funds in the process of collection. A possible problem with the agent bank is that, as with most relational banks in developing economies, they are captured by the interests of the parent enterprise and cannot act as independent sources of monitoring. Their credits may be exposed to risks that are too idiosyncratic. For banks to operate on a sound basis, it is necessary that their assets are sufficiently diversified. When the funding basis of the bank is thin, however, as is the case with most agent banks, it is difficult to diversify lending while meeting the funds requirements of the parent enterprises. The formation of a loan syndicate may be a possible re- sponse. The difficulty is that the sheer number of agent banks and their small size seems to prevent the development of syndicates, because the question of which bank ought to bear the responsibility for syndicate organization (ex ante monitoring), interim monitoring, and how to set the priority for daims cannot be settled easily. In spite these problems, however, the banking sector in transitional economies appears to be gradually evolving as a viable institution. Peter Dittus (1994) recognized the increasing spread between the deposit and lending rate and the recent noticeable decline of net lending to enter- prises in Central European economies. By careful examination, he ten- tatively concludes that the decline of lending is not a result of a credit crunch from the government deficit, and it can be regarded as a hardening of budget constraints for the enterprises. Hle cautiously notes: "clearly, the environment in which banks are operating and their behavior have changed much more than seems to be commonly acknowledged. It has also become evident, however, that the difficulties remain to be overcome are substantial" (p. 34). The chapter by Belyanova and Rozinsky in this volume indicates that the difference between better spin-offs of the ex state banks and newly created banks are beginning to be blurred, and some of them seem to be evolving as viable institutions in spite of the problems. 26 Corporate Governance in Transitional Economies What difficulties are to be overcome in order for better banks to evolve into active monitors of insider control? Let us try to identify some basic problems to be addressed. The first is the dilemma between risk diversification and monitor- ing. In order to be free from soft credits and the excessive exposure to idiosyncratic risks associated with relational lending, it is desirable that the bank diversify its loans. As noted earlier, one method to achieve this is to form syndicates. At the same time, the formation of loan syndicates may dilutc incentives for the bank to monitor. How can we resolve this dilemma? The second problem is related to the social costs of bankruptcy. As we have hinted, one possible advantage of a bank-centered monitoring mechanism is that the default of debt repayment can trigger the auto- matic shift of control rights from the insider to the creditor bank, even if the latter does not own a block of shares. The mechanical application of bankruptcy procedures would be unproductive given the current state of transitional economies.2 The newly corporatized SOE seems to need outside financing, and sometimes subsidies, to be viable and per- form the neceary restructuring. How can such finance and subsidy be made without perpetuating the soft credit relationship between the bank and the enterprise? As previously noted, there are some 1,700 banks today in Russia. This number is simply too large, and the average size of banks is too small to induce risk diversification and delegate responsible monitoring to single banks. Nevertheless, that more than 1,000 banks devoid of the traits of the old state bank system have emerged from scratch in only a few years may be considered as a posirive sign of the potential for vig- orous evolutionary change. It is said that some of the new banks were 2. Factors pointed out by many authors as working againsL the mechanical application of the bankruptcy procedure in the transition: without a sound payment system, many viable enterprises may be forced into bankruptcy by a mere chain reaction; the asset registry does not exist and privaTe ownership in land is not legally well defined; the bankruptcy procedure may involve costs of maintaining a system of commercial courts; and the lack of expertise and discipline in receivership, and the absence of dear rules regarding claim subordination, may also incur additional social costs of bankruptcr. Controlling Insider Control 27 organized and run by young, competent people (see the chapter by Belyanova and Rozinsky in this volume). Such a situation may suggest that, once a prudent and comoetitive regulatory framework is provided and a stable macroeconomic policy environment is set, some of the existing banks may be given an opportunity to develop as banks accountable for external monitoring. To emancipate banks from fragmented, exclusive relational bank- ing, there is a need to drastically increase the minimum capital require- ments of banks. Such regulation would provide an impetus for acqui- sition and mergers among banks. Further, it would be desirable to limit the lending of the bank to a single enterprise-for example, to one- quarter of the bank's capital. Such measures would induce banks to restrain the volume of relational lending. Nevertheless, our purpose is not to promote arm's-length banking. If portfolio diversification by banks were merely to accelerate arm's-length relationships, a vacuum would remain for external monitoring of the insider control enterprise. Because many banks are now owned by (a group of) enterprises, the movement toward arm's-length banking following the Anglo-American system may not be likely. Through the process of merger and acquisi- tion, originally close bank-enterprise relationships may be diluted, but maintained with some dismnce. The enterprise would likely hold major payment system accounts only with a few banks. Those banks would be likely candidates for the role of lead banks if Iending diversification should lead to organized syndication. In the process of past hyperinflation, bad debts of enterprises appear to have been largely wiped out in Central and Eastern European transitional economies, but it has not solved the recapitalization prob- lem of banks. On one hand, enterprises appear to rely upon intricate networks of trade credits rather than banks credits. Default on trade credits by one large enterprise may trigger chain reactions. On the other hand, banks appear to rely on lending based on interbank markets and funds in the collection process, but they have not acquired a solid deposit basis yet, except for deposits by foreign currencies. Spin- offs of state banks also rely upon the central bank's directed credits as lending sources. One solution to cope with all these problems may be to induce the development of an interbank payment settlement system based on trade bills drawable on partner banks by enterprises. The central bank should then gradually lirmit its capital infusion to the 28 Corporate Governance in Transitional Economies banking sector to "neutral" rediscounting of eligible trade bills at the window rather than directing credits to particular enterprises by discretion. Such a development would not only resolve the problem of supplying money on a sound basis, but also increase the capacity and incentives of the banks to monitor customer enterprises. Nevertheless, necessary state subsidies should be made through the budgetary process, separated from the commercial banking sector. Only through such a neutral stance of the central bank and insulation of the commercial banking sector from discretionary subsidization can the soft credits of banks be reduced. Bibliography Aghion, Philippe, Oliver Hart, and John Moore. 1992. 'The Econo- mics of Bankruptcy Reform." Journal of Law, Economics and Organization 8: 523-46. Aoki, Masahiko. 1984. Tbe Co-Operative Game Theory of the Firm. Oxford, U.K.: Clarendon. Aoki, Masahiko. 1994a. "Contingent Govemance of Teams: Analysis of Institational Complementarity." International Economic Review 35: 657-76. . 1994b. "An Evolutionary Parable of the Gains from International Organizational Diversity." Stanford University. Mirnmeo. Bebchuk, L. 1988. 'A New Approach to Corporate Reorganizations." Har-vard Law Review 101: 775-804. Berg, Andrew. 1994. "The Logistics of Privatization in Poland." In Olivier Jean Blanchard, Kenneth A. Froot, and Jeffrey D. Sachs, eds., The Transition in Eastrn Europe. Chicago, Illinois: The University of Chicago Press. Dittus, Peter. 1994. "Corporate Governance im Central Europe: The Role of Banks." Bank of International Setlements, Basle. Mimeo. Controlling Insider Control 29 Frydman, Roman, Andrzej Rapaczynski, and John S. Earle. 1993. The Privatization Process in Europe. Prague: Central European University Press. Phelps, E. S., Roman Frydman, A. Rapaczynski, and A. Scheifer. 1993. "Needed Mechanisms of Corporate Governance and Fince in Eastern Europe." European Bank for Reconstruction and Devel- opment, Working Paper #1, London. Sheard, Paul. 1994. "Reciprocal Delegated onitoring in the Japanese Main Bank System.- journal 0r '-anese and International Economies 8: 1-21. World Bank. 1993. "The Banking System in Transition-Russia." Europe and Central Asia Department Report 12763. Washing- ton, D.C. Political Economy Issues of Ownership Transformation in Eastern Europe G6rard Roland It is probably no exaggeration to state that analysts of transition in Eastern Europe have strongly underestimated the political constraints to reform. The fat of communism and the advent of democratic re- gimes in 1989 were widely applauded by the people of these countries, raising the general level of expectations. Analysts thought that a drastc and rapid move to the market economy would most likely be accepted, despite the inevitable transitional pains imposed on the population. In practice, things have been much more difficult amd signs of political opposition and backlash have increased, as witnessed by recent elections in Poland, Russia, and Hungary. Understanding correctly some of the basic political economy issues of transition is thus crucial for formulat- ing realistic policies of transition in these countries. In this chapter we focus on the political economy issues of privat- ization and restructuring. Various questions are raised in that context. How serious are the political constraints to reform? What is their influ- This paper was written for the World Bank project, "The Role of Banks in Corporate Governance in Transforming Economies." An important part, of this paper is derived from joint work with Mathias Dewatripont. I benefited from discussions with Erik Bergldf and Janos Kornai. I thank John Litwack and Juro Teranishi for their helpful comments and suggestions. 31 32 Corporate Governance in Transitional Economies ence on the design of reform programs? To what extent can the differ- ences in privatization policies in various countries be explained by differences in political constraints? The answers to these questions depend very much on the view one has of political decisionmaking, and in particular of the relative power or weakness of governments. We define strong governments as governments with the power to set the agenda: they are free to design and propose reform packages submitted to a democratic vote. An agenda-setting government faces political constraints but is strong in the sense that, by virtue of its agenda-setting power, its proposals are not subject to amendment but only to approval or rejection. We define weak governments as governments that are con- tinuously the prey of rent-seeking activities by all sorts of lobbies and interest groups. In that case, the lobbies tend to set the political agenda and succeed in influencing the government to adopt measures they put on the agenda. Fewer policy conclusions can be drawn if one adopts the rent-seeking perspective, but it offers a positive analysis that can be relatively sobering to policy expectations. It is important to emphasize that our definition of "strong" and "