0i"n Economic Development Institute of The World Bank venting Bank -C- se: aLessons-- fr-Rn R-ecn Global B n0h k: FailuWres:: --: 0 --~ :B -0\ 3 Sept. 1q9q -Ger rd Cpr, Jr.- W il ja C. Hunter Dafly M.Leipziger -T-I -~D D XIEVELO}PMENT -STUIES--- - Other EDI Development Studies (In order of publication) Ente?priseRestructuringandUnemploymentinModelsofTransition Edited by Simon Commander Poverty inRussia:PublicPolicyandPnvateResponses Edited by jeni Klugman EnterpriseRestructuyingandEconomicPolicyinRussia Edited by Simon Commander, Qimiao Fan, and Mark E. Schaffer InfrastructureDelivery:PrivateInitiativeandthePublic Good Edited by Ashoka Mody Trade, Technology,andInternationalCompetitiveness Irfan ul Haque Corporate Governance in TransitionalEconomies: InsiderControlandtheRole ofBanks Edited by Masahiko Aoki and Hyung-Ki Kim Unemployment, Restructunng,andtheLaborMarket in EasternEuropeandRussia Edited by Simon Commander and Fabrizio Coricelli MonitoringandEvaluatingSocialProgramsinDevelopingCountries: A HandbookforPolicymakers,Managers, andResearchers Joseph Valadez and Michael Bamberger AgroindustrialInvestmentandOperations James G. Brown with Deloitte & Touche LaborMdarkets in an Era ofAdjustment Edited by Susan Horton, Ravi Kanbur, and Dipak Mazumdar Vol. 1-IssuesPapers; Vol. 2-CaseStudies DoesPrivatizationDeliver?Highlightsfrom a WorldBank Conference Edited by Ahmed Galal and Mary Shirley 7heAdaptiveEconomy:AdjustmentPolicies inSmall,Low-IncomeCountries TonyKillick FinancialRegulation: Changing theRules ofthe Game Edited by Dimitri Vittas PublicEnterpriseReform: 7heLessonsofExperience Mary Shirley andJohn Nel is (Also available in French and Spanish) PrivatizationandControlofState-OwnedEnterprises Edited by Ravi Ramamurti and Raymond Vernon FinanceattheFrontier.Debt CapacityandtheRoleofCredit in thePrivateEconomy J. D. Von Pischke EDI DEVELOPMENT STUDIES Preventing Bank Crises Lessonsfrom Recent Global Bank Failures Proceedings of a conference co-sponsored by the Federal Reserve Bank of Chicago and the Economic Development Institute of the World Bank Edited by Gerard Caprio, Jr. William C. Hunter George G. Kaufman Danny M. Leipziger TheWorldBank Washington, D. C. Copyright 01998 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 HStreet,N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing September 1998 The Economic Development Institute (EDI) was established by the World Bank in 1955 to train officials concerned with development planning, policymaking, investment analysis, and project implementation in member developing countries. At present the substance of the ED1's work emphasizes macroeconomic and sectoral economic policy analysis. Through a variety of courses, seminars, and workshops, most of which are given overseas in cooperation with local institutions, the EDI seeks to sharpen analytical skills used in policy analysis and to broaden understanding of the experience of individual countries with economic development. Although the EDI's publications are designed to support its training activities, many are of interest to a much broader audience. EDI materials, including any findings, interpretations, and conclusions, are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U. S. A. The backlist of publications by the World Bank is shown in the annual Index ofPublications, which is available from the Office of the Publisher. Library of Congress Cataloging-in-Publication Data Preventing bank crises: lessons from recent global bank failures proceedings of a conference sponsored by the Federal Reserve Bank of Chicago and the Economic Development Institute of the World Bank / edited by Gerard Caprino, Jr.... [et al.]. p. cm.-(EDI development studies, ISSN 1020-105) Includes bibliographical references and index. ISBN 0-8213-4202-9 1. Bank failures-Congresses. 2. Bank failures-Case studies -Congresses. I. Caprio, Gerard. II. Federal Reserve Bank of Chicago. III. Economic Development Institute (Washington, D.C.) IV. Series. HG1521.P74 1998 332.1-dc21 98-5911 CIP Contents Abbreviations and Acronyms vi Foreword vii Acknowlecdgments viii Introducticn ix PART I. AVOIDING BANKING CRISES 1. Preventing Banking Crises 3 Jeffrey A. Frankel 2. Regulat:ory Efforts to Prevent Banking Crises 13 Michael H. Moskow 3. The Role of the International Monetary Fund 27 Karin Lissakers PART II. BANK FAILURES IN LATIN AMERICAN COUNTRIES 4. The Argentine Banking Crisis: Observations and Lessons 35 Danny PM. Leipziger 5. Lessons Learned from the Chilean Experience 43 Jorge Marshall 6. Lessons from Recent Global Bank Failures: The Case of Brazil 53 Paul L. Bydalek iii iv Contients PART III. BANK FAILURES IN TRANSITION ECONOMIES 7. Restructuring Distressed Banks in Transition Economies: Lessons from Central Europe and Ukraine 69 Michael Borish and Fernando Montes-Negret 8. Preventing Banking Crises by Supporting the Truth 127 Ivan Remsik PART IV BANK FAILURES IN INDUSTRIAL COUNTRIES 9. Lessons from Bank Failures in the United States 133 James R. Barth and Robert E. Litan 10. The Banking Crisis in Japan 173 Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 11. Bank Failures in Scandinavia 195 Sigbjorn Atle Berg 12. Banking Disasters: Causes and Preventative Measures, Lessons Derived from the U.S. Experience 209 Richard J. Herring 13. The Case for International Banking Standards 237 Morris Goldstein PART V COMMONALITIES, MISTAKES, AND LESSONS 14. Global Banking Crises: Commonalities, Mistakes, and Lessons 249 Douglas D. Evanoff 15. Deposit Insurance 255 Gillian G. Holway Garcia 16. Understanding and Preventing Bank Crises 269 Edward J. Kane 17. Banking Crises in Perspective: Two Causes and One Cure 279 Geoffrey P. Miller 18. Bank Crises: Commonalities, Mistakes, and Lessons Viewed from a South Asian Standpoint 287 John Williamson Contents v PART VI. PREVENTING FUTURE BANKING CRISES: WHAT CAN AND SHOULD BE DONE? 19. Suggestions for Improvements 295 Joseph Bisignano 20. Building a Better and Safer Banking System in Latini America and the Caribbean 303 E. Gerald Corrigan 21. Bank Restructuring Revisited 325 Andrew, Sheng 22. What C'an and Should Be Done to Prevent Future Financial Crises? 333 Edwin M. Truman PART VII. CONFERENCE OVERVIEW 23. Summary 341 George C. Kaufman 24. Concluding Thoughts 345 Gerard Caprio, Jr. Contributo:rs 349 Conference Agenda 361 Conference Participants 365 Index 371 Abbreviations and Acronyms BIS Bank for International Settlements BNA Banco de la Naci6n de Argentina CCPC Cooperative Credit Purchasing Company Japan) CEE Central and Eastern Europe CMEA Council for Mutual Economic Assistance DIC Deposit Insurance Corporation (Japan) DIS Deposit insurance system EU European Union FDICIA Federal Deposit Insurance Corporation Improvement Act FSA Financial Supervisory Agency Japan) FSU Former Soviet Union G-7 Group of Seven G-10 Group of Ten GDP Gross domestic product IMF International Monetary Fund S&L Savings and loan institution SOCB State-owned commercial bank SOE State-owned enterprise vi Foreword The papers in this volume were prepared for the conference "Preventing Bank Crises: Lessons from Recent Global Bank Failures," held in Chicago in June 1997. The conference was organized jointly by the Federal Reserve Bank of Chicago and the Economic Development Institute of the World Bank. Among the 120 participants from 36 countries were senior govern- ment officials from central banks and finance ministries, heads of bank supervision bodies, regulators, bank executives, academics, representatives of the financial press, and experts from the World Bank and the Interna- tional Monetary Fund. During the past 15 years, bank failures and insolvencies have emerged all over the globe, both in industrial and in developing countries. East Asian economies currently dominate the news as their financial sectors demon- strate severe weaknesses, and the International Monetary Fund, the World Bank, and governments intervene to provide financial support. Not only have the crises been widespread, but their costs have been extremely high, thereby creating an urgent need to encourage governments and bank regu- lators to establish and strictly enforce oversight measures and not to delay in redressing banking sector weaknesses. The conference exposed participants to the underlying causes of bank failures as well as to the approaches for dealing with them effectively. This volume focuses on lessons learned internationally and on ways to avoid major banking crises. It therefore addresses one of the key issues currently facing international financial markets and sheds additional light on the topic. Vinod Thomas, Director Economic Development Institute vii Acknowledgments This volume was a joint effort of the Federal Reserve Bank of Chicago and the Economic Development Institute of the World Bank, and many people were involved in its preparation at both institutions. The authors thank Shirley Harris, who helped organize and coordinate both the conference and this volume, and Rita M. Molloy, who provided valuable editorial as- sistance. The authors also thank Mary Elizabeth Ward for coordinating the preparation of the manuscript. She was ably assisted by Chip Vance. John Didier provided oversight and guidance, and Belle Lamdany helped pre- pare the volume for publication. Alice S. Dowsett did a superb job of edit- ing the entire volume. viii Introduction As the organizers of the conference on which this volume is based, we could not have predicted that the 1997 East Asian financial crisis would follow our conference within the space of weeks. What we did know, however, was that bankers and policymakers had not sufficiently internalized the lessons of past bank crises to reduce materially the chances of further crises. The conference on "Preventing Bank Crises: Lessons from Recent Glo- bal Bank Failures" was a joint effort of the Economic Development Insti- tute of the World Bank and the Federal Reserve Bank of Chicago. It took place near Chicago on June 11-13, 1997, and was noteworthy both because of the quality of the speakers and participants (see the conference agenda and the list of participants at the end of this volume) and because it brought together a broad range of experience from industrial, developing, and tran- sition economies. Therefore this volume allows us to compare experiences as we search for common causes, evaluate resolution processes, and seek to prevent such crises in the future. Surprisingly, a seminar at the World Bank in 1983 on the consequences of banking system failure in developing countries elicited a common reac- tion: interesting issues, but aside from a few countries in Latin America, why should this topic be of general concern to development experts? Many were quick to point out that losses were "merely" transfers from one group of society to another, and thus beneath the concern of any worthy macroeconomists. Now, 14 years and about 100 episodes later, those who shrug off banking system problems are in the minority, especially after the East Asian crises of 1997-98. The last one-and-a-half decades have been unprecedented in financial history in the commonality of systemic banking problems, whether co- vert, in the form of financial distress in which net worth in the banking ix x Introduction system is negative, or overt, in the form of runs on the banking system and usually on the country's currency as well. From the richest countries- Japan, the Scandinavian countries, the United States-to the poorest in Sub-Saharan Africa, from the fastest growing countries in East Asia to the new transition economies, few countries have been spared some form of banking crisis. Beyond their sheer numbers, these events have been extraordinarily expensive in terms of their fiscal costs alone. When losses reach the point where the depositors fear for the safety of their funds, the functioning of the banking system is suspect, the payments system may be compromised, and the real sector ceases to find credit, governments step in. Although the U.S. savings and loan crisis of the 1980s received worldwide attention and cost U.S. taxpayers an estimated US$150 billion to US$180 billion, this event was small (3 percent) in relation to gross domestic product compared with many other crises. Indeed, according to official data, by this measure it would not make the top 25 U.S. crises, and perhaps not even the top 50 given available anecdotal information. Current crises in Japan, the Repub- lic of Korea, Indonesia, and Thailand will probably produce losses several times the size of those of the United States. Such large costs imply enormous transfers. Whereas the bill for crises can be readily spread out over time in industrial countries' capital mar- kets, these markets are generally shallow in the developing world, imply- ing a reliance on international markets that are notoriously fickle. Thus, when a large bill looms, government officials tend to delay its recognition, even though case after case-several of which are reviewed in this vol- ume-has demonstrated that this supervisory forbearance allows the losses to increase as bankers continue their money losing operations and gamble for resurrection. When large fiscal bills have to be repaid, they can destabi- lize the best laid macroeconomic programs, and in extreme cases force au- thorities to pay by printing money. More tellingly, large-scale bank insol- vency means that a significant misallocation of resources has occurred, often in the form of vacant or half-finished buildings in many crisis countries and white elephant investments in others. This misallocation leads to slower growth and poorer distribution. As banking crises are clearly an important public policy issue, how to prevent them, or at least minimize their cost, is an important concern of officials, advisers, academics, and indeed, all those concerned with economic welfare. Although many reviews of banking crises have taken place in the last two years, this volume distinguishes itself by the breadth of cases covered and the mix of perspectives offered, namely, those of Introduction xi regulators, crisis managers, policymakers, and academics, some of whom held multiple roles in addition to trying merely to understand the na- ture of the problem. The cases covered here-in Japan, the United States, Eastern Europe, Latin America, and Scandinavia-are all reviewed by experienced experts from these regions. Topics discussed in this volume include the following: * Government guarantees of bank liabilities * Maiket discipline to supplement regulatory discipline * Accuracy and truth in accounting and reporting to make private markets more efficient and enhance the responsiveness and account- ability of bankers and regulators * Transparency and disclosure of bank and regulatory agency activities * Independence of bank regulators from political influence and greater accountability to the public * Adequacy of privately provided capital to absorb bank losses and adverse exogenous shocks * Privatization of publicly owned banks - Foreign ownership of banks to augment domestic private capital and intensify competition i Legal systems, particularly with respect to bankruptcy laws - Bank infrastructure, including the training of both bankers and bank regu!Lators. There appears to be a broad consensus that we know how to resolve most ongoing banking problems, but that there are serious political prob- lems in implementing the necessary solutions, particularly when public agreement is required to use public (taxpayer) funds to resolve insolven- cies and validate both explicit and implicit government guarantees to pro- tect depositors from losses. Governments and regulators are under politi- cal pressure to delay taking actions that they know are correct, but would probably be painful and unpopular. By reducing the threat of runs on banks, the safety net permits regulators to forbear taking these actions. Evidence from around the world clearly demonstrates that many regulators have become poor and unfaithful agents for their principals (the taxpayers) and have, albeit unintentionally, exacerbated the frequency of banking crises. This agency problem has been as costly as the banks' moral hazard prob- lem. Thus a failure to police reckless drivers puts other drivers at risk. Like- wise, highway departments have learned that putting guardrails on moun- tain passes encourages some drivers to increase their speed and increase, rather than clecrease, the likelihood and severity of accidents. xii Introduction The issue of moral hazard is at the root of many of the recent financial crises in East Asia, as banks avoid due diligence in the belief that govern- ments will cover their mistakes. Foreign lenders compound this problem by substituting implicit sovereign guarantees for their own financial due diligence and failing to discriminate among borrowers. Supervisory fail- ures compound failures of corporate governance, and systematically weak segments of the financial sector-ranging from short-term finance compa- nies in Thailand to merchant banks in the Republic of Korea-precipitate systemwide bank failures. Financial liberalization is unaccompanied by proper supervision and regulation. In a number of countries procedures for failure resolution are lacking, allowing one sphere of the financial sector's weaknesses to jeopardize the entire banking system. These lessons are seen repeatedly in the cases presented in this volume. The disparity between enacting sound laws and regulations and implement- ing those regulations is also seen. Weak enforcement is at the root of almost all the crises examined. The need to move to best practice is therefore compelling. Many of the episodes reviewed here followed episodes of financial re- form, and saying that the reforms caused the crises is tempting. This would be the wrong conclusion. First, it assumes the absence of problems before the reform, which is clearly untrue. A more probable situation is that the prereform controls helped conceal the problem, and indeed, were respon- sible for the incentives that sowed the seeds of crisis. Second, and extremely important for authorities considering financial reforms, is the observation that a common element of many reform programs was the early imple- mentation of "stroke of the pen" reforms, while difficult institutional de- velopment, such as creating sound financial supervision, was delayed. The cases examined highlight the importance of correct financial sector incen- tives, regulatory structures, and supervision infrastructure. Although participants were reminded of the role that bankers' disaster myopia plays-the tendency of humans to assign unrealistically low prob- abilities to events that are more distant in time-this myopia also affects foreign investors, resident depositors, and taxpayers. Consequently, au- thorities need not only allow some participants to be exposed to risk, but need to educate and remind various groups of that fact. Good economic times erode not just lending standards, but the political will to pursue sound financial regulation. By highlighting the causes and consequences of crises and what it takes to prevent them, this volume should help keep these issues at the forefront of concern. During the conference panels of renowned experts drawn from academia, banking, and regulatory agencies reviewed the causes of the Introduction xiii crises and the lessons to be drawn, creating a mixture of individual cases with cross-country lessons. Officials in countries currently in the midst of a crisis or those fearing the commencement of one in their own economy will profiL from the material presented here. Notwithstanding the diverse proximate causes and the particular ways in which crises play out in dif- ferent financial circumstances, most feature not only troublesome macro- economic environments, but also perverse incentive problems in bank- ing. Thus beyond the standard plea for better economic management, the authors of this volume posit that achieving incentive-compatible finan- cial regulation, that is, a system that induces participants to behave in a safe and sound manner, is a prime concern for governments, and ulti- mately for electorates. This volume contains observations by knowledgeable observers of and actors in financial crises, as well as novel recommendations for improving banking sector resiliency and recuperative abilities. Given the events of 1997, these clearly warrant close examination. Part I. Avoiding Banking Crises 1 Preventing Banking Crises Jeffrey A. Frankel As a menmber of the administration, I spend a great deal of time talking about the current exceptionally good economic performance of the United States: the economy is booming; the unemployment rate has declined Lo a level unseen since the 1973 oil shock, while inflation has remained tame; investment is high; exports are strong; consumers are confident; and we are well down the road to the first balanced federal budget since 1969. Unfortunately, the United States' strong performance coincides with weakenecd economic performance by some of its international part- ners. This is particularly striking in Asia, where Japan is still recover- ing from a severe recession in the early 1990s. Even some other Asian economies whose breakneck pace was undisturbed by the Group of Seven recessions of the early 1990s, like the Republic of Korea, Singapore, and Thailand, slowed down in 1996, engendering an at- tack of economic angst. Thus since 1990 the United States appears to have traded places with East Asia. Then the Japanese and other Asian economies looked unstoppable, as if Asia had discovered some secret for investment, growth, and economic performance that the United States lacked. This led economists to examine the many structural dif- ferences across the Pacific and to focus on the benefits of the Asian system compared with the U.S. system. The comparison extended to banking and finance. Considering how and why views have changed since then is instructive. 3 4 Preventing Banking Crises Shifting Financial Models Capitalism is broad enough to encompass competing models, including competing models of financial systems. The U.S. model-shared with the United Kingdom, and so sometimes referred to as the Anglo-Saxon model-emphasizes arms-length market relationships. For example, it re- lies heavily on securities markets. To be sure, banks play an important role, but even bank loans tend to be made on arms-length terms. Certainly the government has little to say about where bank credit is allocated. By contrast, the system that has developed in some Asian countries has fol- lowed more in the footsteps of the Japanese model. Without denying the important differences among these countries, the Asian systems tradition- ally seem to place greater reliance on bank loans than on securities mar- kets, exhibit high debt-equity ratios, have closer relationships between banks and the companies that borrow from them, allow extensive corpo- rate cross-shareholding, and feature greater guidance from the govern- ment in their credit allocation decisions. A common feature has been the imposition of compulsory financing of certain activities on banks, in com- bination with repression of the rest of the financial system through, for instance, taxes on securities transactions. This is in some sense the classi- cal Asian financial model. It probably best characterizes Japan some 30 to 40 years ago, and is a more up-to-date characterization of some other Asian countries (for references see Frankel 1995). Germany and other continental European countries have their own sys- tems, but their emphasis on banking relationships versus securities mar- kets resembles the Japanese model more closely than the Anglo-American model. Of course, this simple dual classification of the world's financial systems omits a lot. To take one example, Germany's universal banking system, which has spread to the rest of Europe and to Canada, could be viewed as the antithesis of the U.S.'s Glass-Steagall law (which segregates banking from securities), yet in this regard Japan more closely resembles the United States in that financial institutions are legally segregated by function. (This is no coincidence. Article 65 became law in 1948 under the influence of the American occupation.) Some 5 to 10 years ago, economists were wondering if the Japanese system might not be superior to the Anglo-Saxon one. Their research took both theoretical and empirical forms (Frankel 1993; Hoshi, Kashyap, and Sharfstein 1990a,b,c). The theoretical models assumed asymmetric infor- mation between borrowers and lenders. The idea is that from the view- point of a firm seeking to finance an investment project, typical investors in the securities markets are strangers who have no way of knowing whether Jeffrey A. Frankel 5 the firm's project will have as high a return as its managers claim it will. Such investors will demand a premium to compensate them. Researchers confirmed that firms were better able to finance their investment projects intemally than when they had to go to the securities markets and convince strangers of the worthiness of their projects. Relationship banking was thought to be a possible way around the asymmetric information prob- lems that impeded capital markets. Investigators said that when a firm suffered a temporary setback, the short-sighted American financial system would cut it off from new funds, while Japanese banks had longer time horizons, and would give the borrower the resources to see it through. The question ultimately was an empirical one. The Japanese system seemed to work extremely well. It produced miracle rates of investment and correspondingly high rates of growth. Observers said that it had helped provide Japan with low cost capital, thereby giving Japanese industry an advantage over American competitors. Versions of this type of financial system in other Asian countries seemed to be working well also, in con- trast to the recurrent crises in Latin America.' Now, however, the much vaunted Japanese financial system is look- ing tarnished. The attribute of the system that previously appeared to be a virtue-banks' willingness to go on lending to firms in distress- now turns out to have led to serious problems. Borrowers who should have beeni cut off were not, with the result that further billions were lost. The public has had to pay twice, once in the form of slowed eco- nomic growth as the result of the prolonged overhang of bad loans (and other aspects of the burst bubble), and then again as taxpayers when the government ends up footing the bill. Every country encounters bumps in the road, and concluding too much from a single episode is unwise. However, several aspects of the Asian model have been called into question recently. One of them is the close relationsh.ip between banks and borrowers, and another is administrative guidance irom the government. East Asian financial systems appear to have been less able to withstand economic .hocks than the U.S. system, even though the latter is far from 1. The Republic of Korea, for example, achieved remarkable growth rates. In a 1993 paper (Frankel 1993a), I allowed that "one should hesitate before condemning Korean 'finarcial repression,' given how successful the development process has been over the last thirty years" (p.96). But I did add: "Nevertheless, it may be time to move on to a new stage." Now, four years later, I am prepared to conclude that it is indeed time for Korea to move on. 6 Preventing Banking Crises perfect. Asian systems worked well as long as economies grew fast and steadily. Problems arose when economies slowed or faltered. They include mounting bad loans, overextended property markets, and some scandals. When governments responded by papering over problems rather than addressing them squarely so as to put them in the past, this did not help the situation. By contrast, the U.S. financial system seems to have with- stood shocks more readily, for example, the 1990-91 recession, despite the earlier costly procrastination regarding the savings and loan problem. Increasing the International Focus on Banking Stability In the last few years, the intemational focus on banking crises has increased. As with many other international economic phenomena, increased vulner- ability to financial crises can be traced to two overarching trends that have swept the world: liberalization and globalization. Governments everywhere have embraced market liberalization as the path to faster economic growth. This interest in liberalization stems from a wide consensus that the more statist models of development have failed. Governments worldwide have realized that to grow they must rely on markets. Goldsmith (1969), McKinnon (1973), and Shaw (1973) pointed out the drawbacks of financial repression, which both discourages the accumulation of savings and interferes with their effi- cient allocation. Since then, economists have come to understand that a liberalized, privately focused financial system is a key element of a successful growth strategy. However, as countries liberalize their fi- nancial systems, unconstrained financial institutions have more scope for making potentially costly mistakes. The second overarching trend is globalization. Globalization increases the international effects of domestic financial crises. Private capital in- flows have grown in importance in the 1990s, and many of these flows are intermediated through domestic banking systems. This increases the systems' vulnerability to international shocks, and crises can be trans- mitted from one country to the next. Some numbers illustrate the increase in private capital flows. In the eight years following the onset of the debt crisis in 1982, net private capital flows into developing countries averaged only US$21 billion a year. Since 1991, however, total private capital inflows have climbed to an average of US$146 billion per year. Portfolio capital flows have grown even faster, from US$6 billion to US$54 billion a year, more than a third of total flows. They tapered off somewhat after the peso crisis in early 1995, but hit a new high in 1996. Jeffrey A. Frankel 7 Volatile private portfolio flows can interact with liberalized banking systems to increase the likelihood of foreign exchange pressures. Domestic monetary policies must respond to these pressures, especially when a coun- try is attempting to manage or peg its exchange rate, but to do this, monetary authorit]es must be able to raise interest rates temporarily. If the domestic banking system is already weakened by asset quality problems, raising inter- est rates will be more costly, and they will hesitate. This hesitation can in- crease speculative pressures against the exchange rate, leading to a full-fledged speculative attack. A weak domestic banking system was one element that made it clifficult for Mexico to defend its peso in 1994. Recent Banking Crises A number of prominent banking crises have occurred in recent years. Perhaps the first major one brought about by the current wave of lib- eralization was the U.S. savings and loan crisis in the 1980s, where a liberalized regulatory regime was grafted onto financial institutions with weak capital positions. Severe problems also hit banks in some Nordic countries. Policymakers had liber,alized the rules for these banks, but supervision had failed to keep up. When macroeconomic volatility hit these countries, banks became deeply insolvent, and state takeovers ensued. State capital guarantees and capital injections ran as high as 8.2 percent of gross domestic product (GDP) for Finland (IMF 1994, p. 75). Clean-up costs following banking problems in Spain were even larger. In dollar terms, the banking problems that have plagued Japan since 1990 may represent the biggest financial crisis in recent history. The banking practices that looked so attractive in the 1980s proved to de- pend on a continuously growing economy. They also depended on a stock market bubble, which burst in 1990, sending stock prices plung- ing more than 50 percent in two and a half years. By 1996, nonper- forming loans in Japan had reached 3.3 percent of total loans as offi- cially reported by the major banks (IMF 1996, p. 83). Some signs indicate that Japan's financial problems have lessened recently, partly because banks have begun to get their own houses in order. Nonetheless, for the last seven years a weak financial system has exerted a drag on the Japanese economy. Two years ago we were reminded of the importance of banking sys- tems in developing countries as well. After the peso crisis hit Mexico in December 1994, a weak banking system became a serious constraint to 8 Preventing Banking Crises Mexican financial policy. Problem loans that had mounted in the three years preceding the crisis had already weakened the banking system. Nonperforming loans rose from 4 percent of total loans in 1991 to 8 percent in mid-1994 (Goldstein 1997, p. 7). Mexico also highlighted the issue of contagion. The peso crisis spread in the form of the famous tequila effect to other Latin American countries-especially Argentina and Brazil-and even translated into foreign exchange pressures on some Asian emerging mar- kets-including Indonesia, the Philippines, and Thailand-as well as some industrial countries with weak economic fundamentals. Banking problems have also become apparent in other countries recently. To name just a few, Korea is experiencing pressures on its banks in the aftermath of Hanbo's bankruptcy; Russia and other transition economies are being plagued by poorly managed banks, with incestuous relationships between bank owners and bank borrowers; and the Czech Republic and Thailand are also facing pressures. The cost of banking crises has been severe in many countries. Estimates indicate that since 1980 the costs of resolving banking crises in developing and transition economies have reached almost US$250 billion. According to Goldstein (1997, p. 4), 67 banking systems have encountered crises since 1980, and 52 of these have been in developing countries. Policy Responses The prevalence of banking and financial problems calls for policymakers to develop coordinated international responses. The right answer is certainly not to retreat from globalization or from liberalization, because these have both brought many economic benefits. I wonder if even the Greiders who warn of the dangers of the globalized marketplace truly want such a retreat. Telling banks not to make bad loans is of limited practical use. Banks exist in part to make risky investments, and expecting them to have no bad loans is unrealistic. In an efficient, well-functioning financial system, banks should even fail occasionally. What a financial system should be able to provide is the ability to function efficiently, to support adequate levels of investment smoothly, and to withstand adverse economic shocks. Elements of a financial system that appear to make it more resilient include transparency, sound accounting practices, strong capital ad- equacy, and rules-based supervision. While acknowledging such fail- ures as the savings and loan crisis, and without overselling the U.S. financial system, I would suggest that these are key elements of the American approach. The United States has made further moves in re- cent years to improve rules-based supervision and capital adequacy, Jeffrey A. Frankel 9 for example, "prompt corrective action," though this rule has yet to be tested cluring a downturn. Japan has also moved in this direction. The fact that different Japa- nese banks now pay different borrowing spreads provides evidence that arms-length market transactions are more prevalent in Japan than they used to be. Tokyo's "Big Bang" financial reforms will continue this trend. International efforts to control banking risks go back to the Basle initia- tives of the 1970s. The original Basle Concordat laid down procedures for international banking supervision, to address issues related to the Herstatt failure in 1974. More recently, Basle has provided a forum for bank regula- tors to coordinate on their capital requirements for international banks (IMF 1996, p. 141). This helps level the playing field between banks in different industrial countries, while also increasing the stability of these banks. The accord on capital to be allocated for credit risk has been fol- lowed by an accord on market risks, interest rate risks, and foreign ex- change risks. The Basle initiatives generally were intended to cover only the participating industrial countries, and did not cover banks in devel- oping or transition economies. As recent events demonstrate, emerging markets could benefit from more stringent rules. Such rules could also improve international financial stability. Goldstein (1997) has called for the development of an international bank- ing standard that lays down principles for bank regulation. Emerging mar- ket countries that adopted the standard would reap the domestic benefits of a more stable and efficient financial system and would also benefit from internatio nal investors' increased confidence in their markets. Since the Halifax summit of the Group of Seven countries in 1995, the United States has encouraged efforts to improve financial stability in emerg- ing markets more or less along these lines. In the past year the industrial country members of the Basle Committee on Banking Supervision, which includes the United States, have worked with developing countries to put in place a frarmework for financial stability in the form of a set of core principles of effective banking supervision. Representatives of G-10 and non-G-10 econo- mies also o:rganized a working party that issued a report in April 1997 detail- ing key elerments of a robust financial system. The principles included a proper legal framework, adequate accounting principles, a strong payments and settlement system, high quality and timely financial disclosure, effective risk management and internal controls, and capital sufficient for the risks taken (Working Party on Financial Stability in Emerging Market Economies 1997). The report also laid out an international strategy to promote financial stabil- ity, a strategy that G-10 ministers and governors have endorsed and that re- ceived a further push at the Denver Summit in mid-1997. 10 Preventing Banking Crises In mid-1997 the Group of Thirty previewed another proposal that is based on the idea that globally active financial institutions should get together and develop standards for running their businesses safely. This international self-regulation could complement current efforts to coordinate national regulatory regimes. Also relevant is a recent re- port from a task force of the Institute for International Finance. Such private sector initiatives, not to mention important work at the Inter- national Monetary Fund, the Bank for International Settlements, and other international organizations, are welcome. Conclusion Some policymakers have suggested that Mexico was the first financial cri- sis of the 21st century If the current initiatives bear fruit, however, we can hope to limit the frequency and magnitude of the financial crises of the coming century. With increasing international economic interdependence and the prosperity that comes with it, ensuring that the 21st century sees a minimum of such disruptions is important. The United States and its inter- national partners are working to strengthen the stability of domestic finan- cial systems to contribute to continued international financial stability. References Frankel, Jeffrey. 1993. "The Evolving Japanese Financial System and the Cost of Capi- tal." In Ingo Walter and Takato Hiraki, eds., Restructuring Japan's Financial Mar- kets. New York: Irwin Press and New York University. _ . 1995. "Recent Changes in the Financial Systems of Asian and Pacific Coun- tries." In Kuniho Sawamoto, Zenta Nakajima, and Hiroo Taguchi, eds., Finan- cial Stability in a Changing Environment. MacMillan Press. Goldsmith, Raymond. 1969. Financial Structure and Development. New Haven, Con- necticut: Yale University Press. Goldstein, Morris. 1997. The Casefor an International Banking Standard. Washington, D. C.: Institute for International Economics. Hoshi, Takeo, Anil Kashyap, and David Sharfstein. 1990a. "Bank Monitoring and Investment: Evidence from the Changing Structure of Japanese Corporate Banking Relationship." In R. G. Hubbard, ed., Asymmetric Information, Corpo- rate Finance, and Investment. Chicago: University of Chicago Press. - 1990b. "Corporate Structure, Liquidity, and Investment: Evidence from Japa- nese Panel Data." Quarterly Journal of Economics (September). .1990c. "The Role of Banks in Reducing the Costs of Financial Distress in Japan." Journal of Financial Economics 27 (September): 67-88. JeffreyA.Frankrel 11 IMF (International Monetary Fund). 1994. International CapitalMarkets: Developments, Prospects, and Policy Issues. Washington, D. C. - 1996. International Capital Markets: Developments, Prospects, and Policy Issues. Washington, D. C. McKinnon, Ronald. 1973. Money and Capital in Economic Development. Washington, D. C.: The Brookings Institution. Shaw, Edward. 1973. Financial Deepening in Economic Development. New York: Ox- ford tJniversity Press. Working Party on Financial Stability in Emerging Market Economies. 1997. "Finan- cial Stability in Emerging Market Economies: A Strategy for the Formulation, Adoption, and Implementation of Sound Principles and Practices to Strengthen Financial Systems." Bank for International Settlements, Basle, Switzerland. 2 Regulatory Efforts to Prevent Banking Crises Michael H. Moskow I am pleased to have this opportunity to discuss, from the regulator's per- spective, ways to prevent banking crises. I believe this conference serves as an excellent forum for sharing information available on the causes and con- sequences of past crises. Why repeat avoidable mistakes of the past? It also allows for a discussion of alternative methods to resolve banking crises once they occur. For example, how have the Latin American countries ad- dressed this challenge? Does it differ from the approach taken in the United States? Are there advantages of one approach over another? Finally, this forum permits sharing ideas as to how best to structure the industry and its regulatory oversight to prevent future crises. All three elements of this discussiorn-a historical description of past crises, a critique of alternative means of resolving crises, and an evaluation of methods to prevent future crises-are important and I believe can prove valuable in efforts to im- prove the stability of the industry in the future. This chapter stresses various ways to prevent banking crises, empha- sizing the r ole of regulation. While prompt resolution of any crisis is neces- sary, and is an important function of any public regulatory agency, it is not its major responsibility. I believe we spend an inordinate amount of time positioning ourselves to be able to pick up the pieces when and if a crisis occurs. Once we get to the point where we are cleaning up a crisis, the system has failed. My objective, therefore, is to emphasize preventive ac- tion. I do not believe that banking crises are inevitable. The more time spent on prevention, the less the need for crisis resolution. 13 14 Regulatory Efforts to Prevent Banking Crses In my remarks I will (a) examine why banking supervision and evalua- tion is necessary in a market-oriented financial system and discuss the ob- jective of this supervision and of evaluation; (b) discuss the causes of fi- nancial crises; (c) discuss steps that I believe could be taken to prevent these crises, emphasizing the key components required for an efficient bank- ing system (that is, the infrastructure) and the principles that I believe should drive bank regulation; (d) comment briefly on the organization of bank supervision and the role for international coordination of banking super- vision, evaluation, and reporting requirements. The Need for Supervision and Performance Evaluation Let me quickly review why we need bank supervision and evaluation in market economies. I emphasize that this is for private market economies. If the state is in control of banks' assets or asset allocation decisions, his- tory has shown that resources will not be efficiently allocated. In other words, credit allocation in response to market forces differs significantly from credit allocation in response to political forces. This chapter deals only with regulation in private banking markets. I will restrict my comments to address regulation in private banking markets. Banking systems serve a vital intermediary function in a market economy Banks collect savings from individuals and businesses and make loans to other individuals and enterprises that need longer-term funds to finance investments in plant and equipment and shorter term financing for inventories and working capital. In making loans, banks undertake a credit analysis of the borrower and assure themselves that the borrower has sufficient capability to repay the loan. In this way, banks allocate a scarce resource, financial capital, to its most productive uses. The importance of a strong banking sector to a country's economic growth and development is well known. Efficient financial systems help countries to grow, partly by mobilizing additional financial resources, and partly by allocating those resources to the best uses (see, for example, Evanoff and Israilevich 1990; Jayaratne and Strahan 1996; King and Levine 1993; Schumpeter 1969). Banks are the primary mechanisms for the trans- mission of monetary policy and they play an important role in determin- ing the supply of money in the economy. They also typically form the backbone of the payments system. Why is regulatory oversight needed? Arguing for government in- tervention or regulation in markets that are perfect and efficient is dif- ficult. In the absence of distorting factors, the decisions of individual Michael H. Moskow 15 agents in the economy should promote the public interest in general. That is, agents acting in their own self-interest and weighing the costs and benefits of their decisions would make the same decisions as a benevo:lent, omnipotent social planner. However, instances may occur where the costs and benefits to an individual agent associated with a particular action may diverge from the costs and benefits to society. This is where regulation has a part to play to ensure that the actions individual agents take reflect the costs and benefits to the public in general. Such characteristics are typically associated with banking. Relative to other types of firms, banks have low capital to asset and cash to asset ratios and are highly leveraged, making them more prone to severe liquidity problems. As bank assets are typically somewhat opaque, inves- tors may not have as much information on the condition of banks as does bank management. This asymmetric information between the banks and investors-a form of capital market imperfection-raises the possibility that during stress, as they are unable to determine whether banks are healthy, funds may be withdrawn from a number of banks, thereby creating liquid- ity problems (Diamond and Dybvig 1983). Banks within a system are closely intertwined via interbank borrowing, interbank balances, and payment clear- ing activi-ty. Therefore the potential for spillover effects is typically thought to be worse in banking than in other industries. The failure of one bank may cause depositors at another bank to suspect that their bank could also be the victim of financial distress. The classic regulator fear is that such behav- ior may feed on itself and the failure of one important bank may trigger runs on other banks. In addition, decisions made at one bank may take into account the risk implications for its own portfolio, but not the spillover ef- fects on other banks. Thus entire payments systems may be adversely af- fected instead of individual transactions between banks. The fear is that the whole process can multiply until it results in a full-fledged banking panic that has an adverse impact on economic activity. As banks are so important to the functioning of the economy, to the extent that the externalities described exist, banking supervision and evalu- ation are equally important. However, regulations aimed at resolving the problems generated by market failure can become the root cause of indus- try problerms. Numerous countries have demonstrated that moral hazard problems, r egulatory forbearance, and incentive distortions resulting from the mispricing of the safety net are the underlying causes of banking in- dustry problems. The United States, for example, is currently paying dearly for poorly structured regulations and inadequate supervision and evalua- tion of its savings and loan associations (for a summary of events see Barth 16 Regulatory Efforts to Prevent Banking Crises 1991; Barth, Bartholomew, and Bradley 1990; Benston and Kaufman 1997). Estimates indicate that the failure of hundreds of these savings and loan associations may have cost American taxpayers anywhere from US$175 billion to US$225 billion. Inadequate supervision can also have less direct, though no less painful, costs. Institutions that extend poor quality loans in speculative attempts to earn higher returns will take on higher risks. Gen- erally, such loans do not allocate capital to its best use. The rapid run-up in American real estate prices in the 1980s is a recent case in point. Fueled by imprudent bank lending, the rapid increase in real estate values was one of the primary reasons for the so-called credit crunch in the United States. It also contributed to the deterioration in the financial condition of many large U.S. commercial banks and to the U.S. economic recession of the early 1990s. So how should supervision and regulation be used as preventive medi- cine against financial crises? I will address this question, first, by examining the causes of financial crises, and then by discussing the regulatory prin- ciples and required industry infrastructure necessary to prevent such crises. The Causes of Financial Crises The causes of financial crises can generally be divided into two categories: those induced by macroeconomic factors and those caused by poor microeconomic infrastructures. The two are interdependent. By far the major cause of financial crisis is an unstable economy. This results in deteriorating asset quality, asset price bubbles, and wide swings in asset prices and exchange rates that strain the fundamental business of banking and can lead to systemwide problems. A feedback process may also occur. During upswings banks tend to become "excessively exuber- ant" and exacerbate the business cycle by driving up asset prices beyond those characterized by prudent lending. In the United States, in the ab- sence of severe macroeconomic shocks, the banking industry has been im- pressively stable, almost void of crises. Although macroeconomic shocks may initiate most banking problems, they can be made significantly worse by microeconomic structural prob- lems. These include corporate governance problems; distorted incentive systems generated by poorly structured regulatory arrangements; and poor management practices, including inside lending, fraud, and general inept- ness induced by poor internal controls and information infrastructures. These shortcomings allow what may be relatively minor problems to grow into major problems. Michael H. Moskow 17 The resolution of these problems not only enhances social welfare, but it also generates benefits for most elements of society. Better market or su- pervisoiy oversight prevents the failure of poorly managed banks from spilling over and leading to failures at healthy banks as a result of banking panics or runs. Rigorous market and self-evaluations allow bank manag- ers to spot potential problem areas and to take action to address these be- fore additional problems develop. When all banks carry out such evalua- tions anci disclose them in a fonn that preserves proprietary information, bank managers then have standards of performance by which they can better manage their firms. Similarly, from the perspective of bank stock- holders and investors, market and supervisory evaluations provide needed information on the quality of the bank's management and the riskiness of its investments, and gives information useful in comparing these invest- ments to alternative investments. Means for Addressing the Industry's Problems So how do we address these potential causes of banking crises? What is needed to avoid financial crises? I believe the solutions fall into three inter- dependent categories: = Sound macroeconomic policy * Minimum infrastructure requirements * Bank regulatory principles. Sound Macroeconomic Policy The need for sound macroeconomic policy is self-evident. Some would argue that in preventing financial crises, creating the environment for stable growth is the single most important solution policymakers can bring to the table. It is also typically the major responsibility of central banks around the world. Yet macroeconomic instability continues to be the major cause of financial crises. Infrastructure Requirements Whether the regulator is the market or the public sector regulator, ad- equate information is necessary if it is to succeed. Thus some fundamen- tal infrastructure requirements are called for if banking is to be a stable, 18 Regulatory Efforts to Preivent Banking Crises functioning industry. The following are among the key components that underpin efficient banking systems: * A system of laws and rules for corporate governance and property rights, including bankruptcy laws, and laws that describe the rights of creditors in seizing or disposing of borrowers' assets * A uniform set of transparent accounting standards, statements, and supporting schedules and reports * A facility that provides for external bank auditors and examiners * A set of rules for public disclosure of nonproprietary finan- cial information. It should be noted that most of these components fall outside the direct control of banks and bank supervisors, because they are frequently the re- sponsibility of another sector of the government. Even so, they are vital to the work of bank supervisors and to the bank performance evaluation pro- cess. Without these elements in place the banking system's efficiency is certain to be impaired. Thus in the case of developing countries and transi- tion economies, these items must be properly structured before the bank- ing system is privatized. Note that for industrial countries, having some, but not all, of the components in place is a recipe for trouble. The accounting system is perhaps the component that is most basic to the efficiency of financial markets. The rules all firms must follow in preparing their financial statements must be clearly specified. These statements-balance sheets, income statements, and various schedules- communicate vital information about the enterprise to creditors, inves- tors, commercial counterparts, and regulators alike. A set of consistent, structured, and well-defined standards for finan- cial reporting forms part of the core of the resource allocation process. Banks need dependable financial statements from borrowers to be able to perform adequate credit analysis. Uniform accounting standards allow banks to compare borrowers' relative merits. Similarly, bank supervisors, investors, and managers need dependable bank financial statements to make informed judgments about banks' financial health and performance. For a bank, a vitally important component of an accounting system is a consistent set of rules relating to valuation of assets. Assets need to be valued in a bank's books at their true worth, particularly where this is less that the price paid for them. Without such rules, the bank accounting sys- tem is of limited value. Dependable financial information is necessary not only to gauge the health of individual institutions, but also to make comparisons between Michael H. Moskow 19 different banks. In this way bank supervisors can identify those banks in need of attention. Similarly, bank investors can identify those banks most deserving of their capital. To ensure that users of financial statements can be confident that the accounts are prepared according to the proper prin- ciples, some form of external check is necessary. In market economies, this added comfort is typically secured by the services of an external auditor. Closely related to the standardization of accounting principles is the need for public disclosure. The question is not whether financial statements be made available to the public, but how often they should be provided and the appropriate amount of information that should be given. As stated earlier, if markets possess both the relevant information and the capability to discipline banks adequately, additional regulation has a limited role. If markets possess the disciplining capabilities but incomplete information, regulations should provide the market with adequate disclo- sure about firms' characteristics. The general level of public disclosure has risen as financial markets have demanded more and better information. This is particularly true in banking, where there is a need for additional and better information on hidden reserves, provisions for loan losses, and nonperforming loans. The benefits of disclosure are one of the lessons that emerge from the derivatives debacles of the last few years. If regulatory structures could ensure that firms had to disclose both their ex ante ratio- nale for their hedging positions as well as the ex post performance, many derivatives' positions would have been unwound much sooner (for ex- ample, the Gibson, P&G, and Dharmala cases), thereby preventing the large losses that occurred. Adequate levels of disclosure allow market forces to create strong incentives for banks' management, boards of directors, and regulators to behave prudently. Disclos ure is not, however, a cure-all. Not everyone agrees that deposi- tors can interpret disclosures appropriately. The conventional notion that more disclosure is better ignores the fact that some of this information might be useless to the market; that it might make the task of extracting useful information more difficult for market participants; and that it might have a significant effect on the efficient operation of firms, for example, because of the forced disclosure of strategic and proprietary information by firms. These effects must be considered in the design of the appropriate level (and content) of disclosure requirements for firms. Having said that, I believe that more disclosure is generally preferable to less disclosure as long as it does not produce negative external effects. Disclosure and market disci- pline represent powerful complements to the direct regulation of institu- tions, and must be considered an important part of the regulator's arsenal. 20 Regulatory Efforts to Prevent Banking Crises (See Moser and Venkataraman 1996 for a discussion of the appropriate role of disclosure in financial services.) A publication recently released by the Bank for International Settle- ments (1997) notes that disclosure complements effective bank supervi- sion. The Bank's Committee on Banking Supervision has set up a sub- group to study issues related to disclosure and to provide guidance to the banking industry. In the United States the Securities and Exchange Commission's and the Financial Accounting Standards Board's propos- als on disclosure and accounting for derivatives take important steps in the provision of information to the market. Pinciples Driving Bank Regulation As noted earlier, regulatory interference with the marketplace is warranted under certain circumstances. Obviously, an industry's regulatory frame- work can have a major impact on the efficiency and direction in which the industry evolves. For example, in the United States limitations on geo- graphic expansion have led to a relatively fragmented banking industry with a higher bank per capita ratio than in any other industrial country. One can cite numerous other examples where regulation resulted in firms innovating to exploit a regulatory loophole rather than innovating in a di- rection that is socially optimal. Therefore, we should be extremely careful in developing bank regulations. Policymakers should take three overall principles into account when devising bank regulations, namely: * Regulation should be goal-oriented. * Regulation should accomplish its stated goals efficiently. * Regulation should evolve as technology and market structure change and should not discourage these developments. Let us consider each of these principles in turn. First, regulation should be goal-oriented, not process-oriented. We should start by asking the fol- lowing questions: Why are we regulating? What fundamental public policy objective do we want to accomplish? Why is this an essential principle for regulation? We ask these questions because the fundamental goals of regu- lation typically do not change, even though the industry, technology, and market structure may change radically. Thus, not being wedded to approaches used in the past is important. They are just attempts to achieve the goal, not the goal itself. While this may seem obvious, it is remarkable how often this basic principle is ignored. As an observer of the regulatory process (both from the inside Michael H. Moskow 21 and from the outside) for more than 25 years, I would say that most regulatory bodies take whatever regulatory framework is currently be- ing applied as given. Regulatory innovation usually takes the form of looking for better ways to apply the current framework. Rarely, in my experience, do regulatory agencies take as their primary concern the fundamental goal. Thus the new regulatory proposals being introduced around the world are encouraging. Second, regulatory goals should be accomplished in the most efficient way possible. A regulatory approach is efficient if it accomplishes the de- sired goal with the least amount of collateral damage to the industry's ac- tivity. That is, one wants to use the least intrusive approach that achieves the goals. Efficient regulation often either relies on market mechanisms or has the property of incentive compatibility. Note that the infrastructure discussed earlier is required before we can rely on these market-driven mechanisms. Let us review these alternatives for efficient regulation. Ma.ximum reliance on market mechanisms. The first question must al- ways be: Is government regulation even necessary? Are market in- centives sufficient and available information adequate to ensure that market participants, acting in their own self-interest, effectively achieve the regulatory goals? If the answer is yes, then the market itself is the best dynamic regulator and no further government ac- tion is needed. If the answer is no, then we must look for the barriers to self-regulation by market forces. Perhaps government action can remove these barriers, in which case that would be the direction to look for the optimal regulatory strategy. - lIf the main barrier to market self-regulation is lack of informa- tion, the best role of government regulation may simply be to pr-ovide information. Presumably a bank's credit rating, and thus its cost of funds, would be adversely affected if the information were unfavorable. This may be sufficient to induce prudent be- havior without additional regulatory intervention. - If the main barrier to market self-regulation is market incomplete- ness, the government's role would be to encourage the develop- ment of additional markets. - If the main barrier to self-regulation is the existence of barriers to free entry into the industry, the focus of regulation should be to rernove these barriers, given that whenever possible, regulation should encourage, not restrict, market competition. Ironically, these barriers are often created by the government in the first place. 22 Regulatory Efforts to Prevent Banking Crises Incentive-compatible regulation. To clarify the concept of incentive compatibility, let us contrast the command approach with the in- centive compatible approach. Under the command approach the government simply states what actions are desired of the regulated firm, what actions are permitted, and what actions are prohibited. Under the incentive compatible approach the regulator seeks to align firms' incentives with social goals. In other words, this ap- proach makes it in the firms' own self-interest to achieve the regu- latory objectives efficiently. In this context note that the word de- regulation, which is often used to describe the process of replacing inefficient regulation with more efficient approaches, is somewhat misleading. The regulatory goals have not changed, so in that sense deregulation is not taking place. The change involves "smart regu- lating," that is, achieving the same regulatory goals in a better way. The third principle is that regulation should evolve as technology, mar- ket structure, and the industry change and should not discourage these developments. While this principle seems obvious, it is rarely applied. Rather, we often see regulatory approaches in place long after techno- logical changes have rendered them obsolete. If possible, regulation should be self-evolving. If regulatory change must pass through cumbersome national or international political bod- ies, then changes may be difficult to implement. Having a structure that is designed to evolve as the industry evolves is better. This is a two-way street: innovations within the industry can affect the viability and appro- priateness of certain regulatory schemes, and similarly, regulations them- selves can stifle or encourage innovation. Let us consider some applications of these regulatory principles tied to proposed bank regulatory structures. In the process I want to empha- size the incentive compatible nature of these proposals. The moral haz- ard problems created by a mispriced safety net have engendered much debate and a number of regulatory proposals aimed at resolving these problems. In the United States, now that banking conditions are relatively good, many proponents are arguing for implementing additional deposit insurance reform. (Earlier reforms were implemented in the early 1990s, when various means to ensure prompt corrective action during banking crises were legislatively introduced. See Benston and Kaufman 1997.) How do some of these proposed reforms fit into the infrastructure requirements and regulatory principles discussed earlier? Let us consider three reform proposals: (a) decreasing the safety net significantly and increasing the role of disclosure, (b) introducing the Michael H. Moskow 23 narrow bank concept, and (c) altering the capital structure to expand the role of subordinated debt. Proposals to decrease the safety net and increase the role of disclosure and market discipline are based on the contention that the systemic impli- cations cf banking failures are relatively limited and, if given adequate information, the private sector could adequately oversee the activities of the banking sector. This approach is currently being evaluated in New Zealand i'see Calomiris 1997). The Reserve Bank of New Zealand decided that from. mid-1996, banks in New Zealand would be required to make detailed disclosure statements, including information about credit ratings, guarantees, impaired assets, material exposure, and capital adequacy. In return, the Reserve Bank removed a number of previous regulatory struc- tures. Every bank branch displays a two-page summary of these disclo- sures to allow depositors to assess the institution's creditworthiness. The Reserve Bank is not responsible for bailing out depositors. The presump- tion is that with full disclosure, depositors are capable of assessing the risk inherent in using a particular institution and no longer need government protection. This approach obviously requires a well-developed infrastruc- ture with a free flow of information. Tfhe narrow bank proposal would limit insurance coverage to a narrow class of deposits that would be covered by extremely safe and liquid as- sets. The insurance fund would not cover activities outside this narrow class of deposits. The market would discipline all other activities (see chapter 17 by Geolfrey Miller in this volume). This proposal satisfies the regula- tory principles laid out earlier, but depends critically on the completeness of the infrastructure and the government's credibility to avoid expanding the safety net beyond the stated limits. To my knowledge no country has implemented this proposal to date. The third proposal involves increasing the role of subordinated debt in the capital structure of banks to decrease the moral hazard problem, in- crease market discipline, and provide for an improved failure resolution process. A command regulation approach to solving the moral hazard prob- lem would have regulators mandate a maximum level of risk for any in- sured bank. This approach has the typical problems associated with com- mand regulation. First, it is difficult for regulators to measure accurately the risk associated with a bank's loan portfolio. Second, any credible effort to measure this risk accurately is likely to be extremely intrusive. Finally, a one-size-fits-all restriction on bank risk may actually prevent capital from flowing to valuable investment projects. Using the favorable characteris- tics of subordinated debt may resolve these problems. Proponents argue that debt holders serve as a superior buffer against income variations for 24 Regulatory Efforts to Pren)ent Banking Crises both depositors and the insurance fund, because in contrast to equity hold- ers, they do not have the potential to realize up-side gains from having banks hold riskier portfolios. In addition, instead of having uninsured de- positors run the bank when asset quality is questioned-thereby disrupt- ing the bank resolution process-debt holders cannot run, but only walk away from the bank as their issues come due. Thus an orderly resolution process could take place. This proposal is consistent with the regulatory principles discussed earlier and requires a sufficient infrastructure, includ- ing mature capital markets, to allow for the required market discipline. Countries without this infrastructure may want to use some form of prompt corrective action to substitute for the disciplining role of subordinated debt. Argentina has recently adopted a somewhat similar approach, one that requires debt issues to submit banks to the forces of the marketplace. It has also been proposed in the United States as an extension to a reform pro- posal that emphasizes market discipline recently released by the Bankers Roundtable (1997). (For a comparison of alternative regulatory schemes with one that relies more heavily on the use of subordinated debt in the capital structure see Evanoff 1993.) Each of these proposals meet the principles outlined earlier and war- rant additional consideration. Organization of Bank Supervision and the Role of International Coordination Let me conclude my remarks with a few observations concerning market discipline and international coordination among banking regulators around the world. The organization of bank supervision differs from country to country. In England (until very recently) and Italy, for example, this is a central bank function. Other countries such as Canada and Switzerland have bank supervisory authorities that are separate from and independent of the cen- tral bank. Still others like Germany, Japan, and the United States have mixed systems, where the central bank shares bank supervisory responsibility with other government agencies. Thus no agreed upon best system exists. How- ever, in countries beginning the transition to a market economy, the func- tion should probably be part of the central bank's responsibilities. In these economies it is easier to ensure the supervisor's independence and the chance for informational synergy is higher when the two functions are housed together, and it is important that the two functions work closely together to resolve crises. Michael H. Moskow 25 As with the central bank, the supervisor must be free from undue politi- cal pressure or control. The bank supervisor should have enough authority to supervise the banks, but these powers should not lead the supervisor into managing the banks. This task belongs to banks' managers and directors. The Basle Committee on Banking Supervision has produced a set of principles to guide bank supervisors in regulating and interacting with the branches of foreign banks operating in the domestic economy. Its report on the topic (Bank for International Settlements 1997) will undoubtedly be- come required reading for all bank supervisors. It is imperative that banks and banking supervisors around the world abide by sound principles if the global banking system is to remain healthy. Conclusion Financial and economic liberalization, combined with the globalization of markets, have clearly changed the contours of the world's financial system. Interest rate and exchange rate volatility has increased; competi- tion among financial institutions, both bank and nonbank, has intensi- fied; and new financial products are continuously being developed. In the face of these changes, banks worldwide have had to develop new markets and services to maintain their market shares and to meet the growing needs of customers. The changing financial environment has also presented important su- pervisory chlallenges, most notably, adapting laws and regulations to chang- ing market realities, improving the infrastructure undergirding the finan- cial system, and recognizing the need for international coordination of supervisory and regulatory efforts. It is therefore important that discus- sions such a1s these we are having at this conference continue in order to ensure stability of the global financial system. References Bank for International Settlements. 1997. "Core Principles for Effective Banking Supervision." Basle Committee on Banking Supervision, Basle, Switzerland. Bankers Roundtable, The. 1997. "Deposit Insurance Reform in the Public Interest, Partnering for Financial Services Modernization." Report of the Subcommittee and Working Group on Deposit Insurance Reform Retail Issues and Deposit Insurance Committee. Washington, D. C. Barth, James R. 1991. The Great Savings and Loan Debacle. Washington, D. C.: Univer- sity Press of America. 26 Regulatory Efforts to Prevent Banking Crises Barth, James R., Philip F. Bartholomew, and Michael G. Bradley. 1990. "Determi- nants of Thrift Institution Resolution Costs." Journal of Finance 45(3): 731-54. Benston, George J., and George G. Kaufman. 1997. 'FDICIA after Five Years." The Journal of Economic Perspectives 11(3): 139. Calomiris, Charles W. 1997. "Designing the Post-Modern Bank Safety Net: Lessons from Developed and Developing Economies." The Bankers' Roundtable Pro- gram for Reforming Federal Deposit Insurance, May 23, American Enterprise Institute, Washington, D. C. Diamond, Douglas W., and Philip H. Dybvig. 1983. "Bank Runs, Deposit Insur- ance, and Liquidity." Journal of Political Economy 91(3): 401-19. Evanoff, Douglas D. 1993. "Preferred Sources of Market Discipline." Yale Journal on Regulation 10(2): 347-67. Evanoff, Douglas D., and Philip R. Israilevich. 1990. "Regional Regulatory Effects on Bank Efficiency." Regional Economic Issues Working Paper Series, WP-1990/ 4. Federal Reserve Bank of Chicago, Chicago. Jayaratne, Jith, and Philip E. Strahan. 1996. "The Finance-Growth Nexus: Evidence from Bank Branch Deregulation." The Quarterly Journal of Economics 111(3): 639. King, Robert, and Ross Levine. 1993. 'Finance and Growth: Schumpeter Might Be Right." Quarterly Journal of Economics CVIII (1993b): 717-38. Moser, James T., and Subu Venkataraman. 1996. "The Economics of Disclosure Re- quirements for Derivatives." Chicago FedLetter October(110): 1-3. Schumpeter, Joseph. 1969. The Theonr of Economic Development. Oxford: Oxford Uni- versity Press. 3 The RLole of the International Monetary Fund Karin Lissakers Why should the International Monetary Fund (IMF) have a role in avoid- ing banking crises at all? Why should an institution designed to promote international monetary stability and to help its member governments avoid or cope with balance of payments crises be concerned with banking regu- lation and banking crises? Chapter 2 gave at least part of the answer when it noted that there are three keys to preventing bank crises: having a sound macroeconomic policy, possessing an effective financial infrastructure, and following guiding regulatory principles. The Fund's role in the first area is obvious. Advising governments on macroeconomic policy, on monetary and fiscal policy, is the IMF's bread and butter, what the Fund does most day to day. The Fund's involvement in the other two areas-financial in- frastructure and guiding regulatory principles-is somewhat more surpris- ing and a far more recent undertaking. Group of Seven to Modernize International Architecture The Fund's new mission in the banking area is an outgrowth of a broad international policy debate. During the Group of Seven (G-7) heads of state econo:mic summit in Naples in 1994, President Clinton challenged his Canadian, European, and Japanese partners to begin to "modernize the international architecture" for the 21st century. This marked the be- ginning of a systematic, international effort led by the United States, 27 28 The Role of the International Monetary Fund particularly the Department of the Treasury and the Federal Reserve Board of Governors, first, to identify threats to international economic and financial stability that required cooperative international solutions; and second, to ensure that the institutional framework to develop and implement such cooperative solutions exists. The globalization of financial services has been a particular focus of the G-7's efforts during the past two years. The G-7 deliberations are part of an ever widening national and international dialogue that includes the Group of Ten, the Bank for International Settlements with its enlarged member- ship, the Basle Committee on Banking Supervision, and new regional cen- tral bank and supervisory group forums that have been formed in the Americas and more recently in Asia. At the G-7 summit President Clinton hosted in Denver in July 1997, the leaders announced the following: * Steps toward the establishment of a multilateral network of super- vision appropriate to today's global markets and global institutions * Progress toward a framework of strong supervisory principles for the major globally active financial institutions * New steps to improve transparency * Steps to reduce risk in payment and settlement systems * Endorsement of a concerted international strategy to help emerg- ing economies strengthen their financial systems, including a new, universally applicable set of core principles for effective banking supervision. These initiatives will help reduce the risks and costs of future crises. As part of this exercise, major industrial countries have asked, and the full 182 country membership has agreed, that the IMF should augment its traditional macroeconomic advisory and balance of payments sup- port role with more concerted activity in the banking and bank regula- tory area. It will work closely with the World Bank, which is also build- ing up its capacity in this field. The IMF's activities in the banking area could be summarized as "evangelical," "prophesying," and "salvaging." The IMF will undertake to spread the gospel to member governments of sound regulatory and supervisory principles that bank supervisors and other regulatory groups have endorsed. The Fund has already begun to collaborate with various expert groups in drafting policy guidelines for IMF staff to incorporate in their annual macroeconomic consultations with member governments. The Fund will also try to alert member authorities to weaknesses or risks to financial stability; to issue warnings of emerging banking problems; Karin Lissakers 29 and if disaster strikes, to help governments manage crises in banking systems, which it can do by incorporating banking rehabilitation in Fund- supported stabilization programs, providing technical assistance, and helping mobilize other sources of expertise and financing as the IMF did recently for Mexico and Venezuela. How Does Preventing Banking Crises Fit with Traditional IMF Activities? The Fund's new banking role is a more natural fit with its traditional mac- roeconoraic activity than one may think. The G-7 turned to the IMF be- cause it offers several features no other international institution does. It has a unique mandate for universal surveillance: it is the only institution that goes in and "looks at the books" in 182 countries once a year. Its an- nual detailed economic consultation mandated under Article IV of the IMF's Articles of Agreement involves a close dialogue with monetary and finance officials at both the technical and policymaking levels. These consultations already cover key fiscal and monetary policy issues, including some regu- latory matters. Adding banking and financial supervisory policy to the Fund's surveillance exercises will require some beefing up of its expertise in this area, but nevertheless, it seems a natural extension of existing sur- veillance activities. Indeed, one could argue that it is a necessary exten- sion, because left unaddressed, weaknesses in a country's banking system can trigger or compound the fiscal and monetary problems the Fund seeks to prevent through its surveillance. The IMF has a second channel to member countries through its condi- tional lending, that is, Fund financing tied to corrective policy measures. While in the past the Fund has not consistently recognized the impor- tance of incorporating banking issues into its macroeconomic adjustment programs, it now has a better understanding of the links between the banking regulatory system and balance of payments and macroeconomic instability, and that those linkages become more potent in liberalized, internationally exposed financial markets. Banking and macroeconomic problems can feed off each other in what Myrdal often referred to as "cir- cular causat:ion with cumulative effects." Failure to take into account the impact on the banking system of nec- essary macroeconomic policy adjustment can cause IMF-supported ad- justment programs to fail. Fund-supported macroeconomic adjustment programs to eliminate unsustainable current account deficits typically include monetary and fiscal tightening, and often involve exchange rate 30 The Role of the International Monetary Fund adjustment. If higher interest rates and currency depreciation mandated under the program create distress in the banking system and the gov- ernment responds with emergency provision of liquidity to distressed banks through central bank credit, this can undermine price stabiliza- tion. A fiscal bailout of the banking system can blow a huge hole in the government budget, as demonstrated by both industrial and develop- ing country banking crises, that leaves both the banking system and the Fund's stabilization plan in a shambles. However, Fund conditionality through a well-designed adjustment program that also addresses bank- ing problems can be an effective means of permanently improving the performance of member countries' financial sectors. Emerging Markets Pose a Special Challenge Emerging markets, including the former communist states, pose spe- cial challenges. These markets are characterized by extremely short ma- turities on financial liabilities, so interest rate shocks are transmitted almost immediately through to borrowers. Heavy dependence on for- eign borrowing to finance high levels of investment loads exchange rate risk on banks or their customers-risks that because of hubris or lim- ited market options may be largely unhedged. A political commitment by the government to hold the exchange rate firm will, of course, exac- erbate the tendency to take unhedged exchange rate positions. Com- pounding the problem is the fact that banks in these markets may have poor asset quality to begin with because of insider dealing and weak supervision. Low public confidence in banks and in the authorities can quickly trigger bank runs and capital flight. A lack of transparency in these markets feeds the worst fears of both domestic and foreign de- positors and investors. Consequently, banks in emerging markets are both more likely to be subjected to macroeconomic shocks-internal and exogenous-and less able to absorb them. The first task for the Fund and its partners is to develop regulatory and policy guidelines tailored to these emerging market conditions. As already mentioned, the IMF and the World Bank are working closely with the Group of Ten, the Bank for International Settlements, and the Basle group to do just that. Once these guidelines are agreed, the IMF will begin to proselytize. Strengthening banking systems in the emerging markets is not just a mat- ter of fine-tuning regulation and beefing up supervision. Poor supervision and lack of transparency in the financial sector often reflect more endemic weakness in the legal order generally and a poor quality of governance. Karin Lissakers 31 All emerging markets, by definition, are undergoing rapid change. The transition from a government-directed, relationship-based economy to a market-driven, rules-based system is difficult. In Asia, for example, deregulation may be extensive on paper, but governments remain reluc- tant really to let go, and economic actors may operate on the assumption that they will not, and therefore take excessive risks. In former commu- nist states, Russia, for example, the old legal order has been destroyed, but a new order is not yet fully formed. Thus nonpayment of obligations is endemic and contracts are unenforceable. Many African countries have weak administrative capacity and corruption is rampant in both the pub- lic and private sectors. The international establishment's attitude toward governance issues has undergone a profound change, from benign tolerance to recognition of gov- ernance as an economic factor with strong macroeconomic and micro- economic effects, including on the financial sector. Both the IMF and the World Bank are actively pressing member governments to root out corrup- tion, improve the management of public funds, and increase the transpar- ency and accountability of public management generally The IMF has inter- rupted financial support for governments that fail to respond. Fund adjustment programs are also putting greater emphasis on the legal infra- structure, including property law, contract enforcement, and payments settle- ment. Such measures are obviously critically important in transforming former centrally planned systems into successful market-based economies. The Fund is beginning to apply the lessons learned in these transition coun- tries to its work in other emerging markets and in developing countries gen- erally, paying particular attention to banking supervision and regulation. The IMF's biggest challenge will be to integrate financial market and regulatory expertise with its macroeconomic work on a day-to-day basis. This will require both some retraining and bringing in new people. How- ever, the Fund's management and board are fully committed to this en- deavor. Its importance will only grow as the IMF becomes more deeply engaged in capital account liberalization. Part II. Bank Failures in Latin American Countries 4 The Argentine Banking Crisis: Observations and Lessons Danny M. Leipziger The Argentine banking crisis, which followed closely on the heels of the Mexican crisis, was caused largely by factors that were external to Argen- tina. That is not to say that the system had no weaknesses, and certainly the Convertibility Plan and the quasi-Currency Board added additional risks, but Argentina's economy was not mismanaged. Argentina seems to be of those rare cases of a country whose mac- roeconornic fundamentals were basically sound, yet which suffered a banking crisis. It had a current economic deficit that was in the range of 3 to 4 percent of gross domestic product. It had stable political and economic structures. It had relatively high reserves. Its debt structure was long term. Argentina had also completed the bulk of its reforms, for example, in terms of privatization, and was therefore unlike Mexico. It did have one major weakness, the Currency Board itself, insofar as it had limited lender of last resort capacity. Seen differently, however, this limitation gave the peso its credibility. The Crisis In any event, the crisis in Argentina was extremely swift. Between Decem- ber 1994 and the first quarter of 1995 the banking system lost about 16 per- cent of deposits, and the Central Bank lost about US$5 billion in reserves, one-third its total. Table 4.1 depicts the chronology of events in Argentina. 35 36 Tie Argentine Banking Crisis: Observations and Lessons Table 4.1. The December 1994-April 1995 Argentine Banking Crisis Event Outcomes Initial shock * Domestic minicrisis following failure of bond trading house shakes confidence and causes banks to cut lines to these mayoristas. * Tequila effect shakes confidence in Latin America and investors re-evaluate Argentine exposure. * Stock and bond markets suffer large losses. * Banks call in loans extended to dealers and provincial banks, now largely insolvent because of earlier mnisman- agement. Aftershock * Sensing increased risks to convertibility, deposit with- drawals begin that amount to US$2 billion in two weeks. * Liquidity crisis forces banks to cut credit lines. * Central Bank persuades top five banks to provide US$250 million safety net. * Central Bank establishes second net via reserve require- ment reduction of top 25 banks, yielding US$790 million. Continuing crisis * Deposits fall further during January-March 1995, reach- ing 16 percent reduction or US$8 billion. * Interbank interest rates skyrocket. * Dollarization increases. - Central Bank extends extraordinary liquidity assistance above limits of bank capital and for longer than 30 days, totaling US$1.7 billion rediscounts and US$300 million repos. - Some banks fail. Freefall stops * International package (International Monetary Fund, World Bank, Inter-American Development Bank) plus domestic and international bond issues restore confi- dence. * Strong commitment to convertibility maintained, although reserve level falls by US$5 billion, close to minimum possible. X Deposit insurance (limited, privately financed) announced. * Dual bank restructuring funds to private provincial banks and to restructure private banks established with aid of multilateral banks. * Fiscal strengthening plans announced. * Bank consolidation as 28 cooperative and 5 wholesale banks close. Results * Provincial banks moribund, 15 in process of privatization or closure or are acquired. * Top 10 private banks increase market share as deposits begin to return. Source: Caprio and others (1996). Danny M. Leipziger 37 The initial crisis started in a particular segment of the banking system, the wholesale banks, many of which were overextended. Jittery investors began to withdraw from Argentina. Falling stock and bond prices created difficulties for some banks, liquidity was affected, resulting in enormously high interbank rates. Provincial banks lost their sources of distress financ- ing and many became technically insolvent. The Central Bank reacted quickly and tried to establish an effective safety net, basically drawing on the stronger banks and relying on the mandated extremely high reserve requirements. The Central Bank tried to recycle some of the excess liquidity, inevitably from stronger to weaker commercial banks, but legal and political issues limited its ability to act on provincial banks. The crisis reached tremendous proportions, with a reduction of depos- its of US$8 billion in the course of three months and a flight to quality, for instance, people moving accounts not only into U.S. dollars, but also to foreign l:ranch banks. The Central Bank again responded with some ex- traordinary liquidity assistance, which allowed it to give liquidity to banks for longe r than 30 days and in excess of their net worth. This required emer- gency legislation to amend the Central Bank's charter. Nevertheless, some banks did fail and a crisis of confidence ensued. You can imagine being in Argentina, looking at the newspaper every day, and comparing the stock of international reserves and the money supply, and when they got very close, under the Convertibility Plan it was clear what could happen, as the money supply was fully backed by reserves. The free fall was finally stopped by a change in these expecta- tions, a firm policy announcement by Domingo Cavallo, the minister of finance, and external assistance. First, the government announced fiscal austerity measures, unusual in a crisis, but important for Argentina, and issued some international bonds to restore confidence and provide ad- ditional liquidity. Second, it did finally obtain an international package from the International Monetary Fund, the World Bank, and the Inter- American Development Bank. Third, the government announced a de- posit insurance scheme, which was extremely limited, but was meant to allay the fears of smaller depositors, and it did let some banks fail. (The choice facing Minister Cavallo at that juncture was either to let banks fail or to lose the Convertibility Plan itself.) A major factor that influenced this crisis was that the Argentine bank- ing system had a tremendous proportion of its assets in public banks, roughly half of the system's assets. Within that public sector, assets were split approximately 50-50 between the Banco de la Naci6n de Argentina (BNA), the large federal public bank, and a whole host of provincial and municipal banks. Initially, weaknesses in the provincial and municipal 38 The Argentine Banking Crisis: Observations and Lessons banks, as well as some wholesale traders, exacerbated the difficulties in the banking system, resulting in a contagious loss of confidence. The economic fundamentals in Argentina were quite good at the time, and banking supervision was reasonably good. Capital requirements were higher than elsewhere, at 11.5 percent, 8 percent in Tier One capital. Nev- ertheless, during this crisis the wholesale banks lost 70 percent of their deposits, the provincials lost 40 percent, and the commercial banks lost 30 percent, so the system suffered a tremendous shock. The remarkable thing is that in the end the system was able to rebound, with some bank- ing failures and some losses, but basically within a year deposits were up to their original levels and the system survived, albeit with a smaller num- ber of banks (table 4.2). Vulnerability to Crisis The following five aspects of the Argentine banking system made it sus- ceptible to banking crises: * The provincial banks were extremely weak. Estimates indicated that they had nonperforming portfolios of about 40 percent before the crisis. * The BNA, although reasonably well-managed, was basically "too large to fail," a dangerous state of affairs. Admittedly, the BNA was a useful instrument for the government during the crisis, because the government was able to use it to recycle liquidity. Nevertheless, in the aftermath of the crisis the BNA itself restructured many of its loans, and no one knows exactly either its condition or the true cost of that restructuring. Table 4.2. Consolidation of the Argentine Financial Sector, 1990-97 (number of institutions) Month/Year Institution 12/90 12/94 12/95 12/96 03/97 Government owned 36 33 31 20 20 Private sector, domestic 57 66 56 64 61 Private sector, foreign 31 31 30 28 27 Cooperative 45 38 10 8 7 Other financial companies 51 37 30 26 26 Total financial system 220 205 157 146 141 Source: Banco Central de la Republica Argentina (1996). Danny M. Leipziger 39 • The questionable lending activities by some of the wholesale banks and quasi-banks also contributed. Many of these institutions took in high-rate deposits and lent money to weak provincial banks who were saddled with their own nonperforming loans, many to the prov- inces that were also their owners. This was the segment of the bank- ing sector that was not easy for the superintendency to supervise because of Argentina's federal structure, and it was clearly in trouble. * The phenomenon of veteran depositors was yet another contribut- ing factor. Everyone in Argentina has the equivalent of an MBA in finance based on the history of the last 30 years, so that people read risk signals very quickly, and these produce rapid asset shifts. * T he quasi-Currency Board places significant restrictions on the lender of last resort capacity of the Central Bank. That capacity is not zero, because a portion of the currency can be backed by bonds, so it has a limited capacity to act as lender of last resort, but it is quite circum- scribed and the Central Bank cannot add to its resources in a crisis unless it can borrow from other central banks or the Bank for Inter- national Settlements. (The former is difficult in a regional crisis and the latter proved to be largely unavailable.) Resilierncy in Crisis A number of other factors contributed to Argentina's resiliency. One, as already mentioned, was the high capital requirements, well above the Basle norms. Reserve requirements were required to be at 32 percent of deposits, which was extremely high, but related to the Currency Board concept. Another important factor was that supervision had improved tremendously in Argentina in the previous few years, with new hirings in the superinten- dency, many more qualified people, and an active training relationship with the New York Federal Reserve Bank. To this some would add that the lack of deposit insurance was something in Argentina's favor: people had to be much more alert about where they put their funds. Crisis management is clearly critically irnportant. The Central Bank's quick response took the form of conditional liquidity and active encouragement of mergers and acquisitions. The economic team of Minister of Finance Cavallo, Central Bank Governor Rogue Fernandez, and the executive branch were able to work swiftly with the National Assembly to pass key legislative changes at the time of the crisis, first on the fiscal side to increase revenues and cut expenditures, but also to provide greater flexibility for the Central Bank to deal with suspended banks, basically by segregating their assets 40 The Argentine Banking Crisis: Observations and Lessons and being able to restructure and deal with troubled banks without getting caught up in the courts. The idea was to avoid putting a failed bank through the legal process, because that would take 10 years and not lead anywhere. The practical policy alternative was to give additional powers to the Central Bank to enable it to segregate parts of the balance sheet, place a certain pro- portion of the assets and liabilities of banks with other banks, and allow losses to be taken in the other areas. Policy Lessons Let us take a look at the lessons of the Argentine crisis and extract policy observations for other developing or transition countries. First, the gov- ernment should keep a watchful eye on public sector banks, be they the provincial or municipal banks in Argentina, the state banks in Brazil, or the banks in Mexico, which in the end can carry implicit national or sub-national guarantees. None of Argentina's provincial governors wanted to see their banks fail, as they would be unable to explain to their constituents that the state government did not stand behind those deposits. Nevertheless, a number of provincial banks in Argentina did close, and many cooperatives were merged with the help of World Bank and Inter-American Development Bank loans.' Second, the general consensus is that governments should let banks fail if that can be dealt with in a normnal commercial fashion. Public policymakers tend to try to prevent bank failures, but in the end weaknesses tend to accu- mulate, and if a standard failure resolution process is not available, the out- come is structural weaknesses that can be detrimental when a crisis, such as the tequila effect, materializes. Systemic failure then becomes a possibility. Third is the question of how to judge portfolio quality in a more dy- namic sense. Portfolios that appear perfectly healthy one day can look quiite unhealthy the next day depending on the external circumstances. In tran- sition economies the situation may depend on what is happening with enterprise privatization, and in other developing countries on what is go- ing on in the external environment, or so-called contagion.2 This was the case in Argentina. Looking at the quality of the portfolio under various assumptions would therefore be helpful. 1. During 1996 at least nine provincial banks were privatized (that is, deals com- pleted), following intervention in 1995. Many cooperatives were either acquired or merged. 2. Portfolio quality in East Asia, for example, has tended to be highly correlated with real estate prices. Danny M. Leipziger 41 Fourth, having the appropriate regulation and legislation in place, and giving the Central Bank enough power so that banks know that it means business, is key. An extremely important factor in Argentina (in contrast to Mexico) was that bank owners knew they would lose their equity. This forced many banks to look for acquisition partners and other ways to re- capitalize, but they only did so with the implicit threat of the Central Bank hanging over them. Bank closures have some positive demonstration ef- fects, and that was a pretty clear lesson for the ensuing spate of mergers and acquisitions.3 Of course, the Central Bank, in its eagerness to see merg- ers and acquisitions rather than outright closures, may sanction the con- solidation of weak banks, either creating larger weak banks or increasing the risks of the stronger banks. Interestingly, the largest, strongest banks in Argentina refused to absorb many troubled banks in the aftermath of the crisis, despite inducements to do so. Finally, the fiscal side is important as an indicator of how serious the governmnent is in its actions. The entire rescue package for Argentina would not have succeeded had the government not been able to con- vince (a) the international community, that is, the International Mon- etary Fund and capital markets, that it was serious in its commitment to the convertibility plan and this required fiscal control; and (b) the multilateral development banks, namely, the World Bank and the Inter- American Development Bank, that it was serious about privatizing or closing moribund provincial banks and promoting the closure or acqui- sition of unviable commercial banks. In the end, therefore, successful crisis management depended on strong policy leadership. References Banco Central de la Republica Argentina. 1996. "Estados Contables de las Entidades Financieras." Buenos Aires. Caprio, Gerard, Jr., Michael Dooley, Danny Leipziger, and Carl Walsh. 1996. "The Lender of Last Resort Function Under a Currency Board: The Case of Argen- tina." Open Economies Review 7:195-220. 3. Between January 1995 and September 1995 eight banks were liquidated. 5 Lessons Learnedfrom the Chilean Experience Jorge Marshall This chapter discusses the main lessons learned from the Chilean banking crisis of the 1980s, the basic principles of the regulatory framework that have been applied since the crisis, and banking regulation and supervision issues that still need to be dealt with. Lessons Learned from the Crisis The Chilean banking crisis took place in 1982-84. Its estimated cost ranges from 30 to 40 percent of GDP. The ingredients of the crisis were a combination of erroneous mac- roeconormic policies and weak regulation. In the years before the crisis, the exchange rate was fixed, domestic inflation was running above in- ternational levels, and wages were indexed. This resulted in a signifi- cant misalignment of key prices, in particular, of domestic asset prices, the exchange rate, and wages. The inconsistent macroeconomic policies also led to rapid credit expan- sion and large capital inflows into a banking sector that had been recently deregulated and whose supervision was weak. Nonfinancial conglomer- ates owned the banks, which lent to the companies that owned them, and connected lending was widespread throughout the Chilean banking sec- tor. This led to situations in which the banks increased their capital base using money lent to banks by their owners. This type of connected lending has played an important role in most banking crises in Latin America. The only difference is that in the 1980s most of these operations were domestic, 43 44 Lessons Learnedfrom the Chilean Experience and in the 1990s some international counterpart was involved. In the later cases, where a conglomerate owns both a domestic and an off-shore bank, identification of the corporate structure and regulation are more difficult. Although deposits were not insured according to the law, previous bailouts of banks by the government resulted in full insurance of de- positors. This led to the belief that deposit insurance was generalized. This implicit deposit insurance and the lack of transparency about banks' financial conditions exercised a considerable role in depositors' compla- cent behavior in regard to the risks that the banks were taking. This ex- acerbated moral hazard problems in the banks, thereby increasing risk taking in credit and investment. For example, some banks went bankrupt when they borrowed abroad to finance credit to the nontradables sector, like real estate. Even when a balance sheet shows that the values of assets and liabilities are equal, the underlying values at "normal" prices may not be equal. The difference is usually excessive risk taken by the bank. If there is a change in the interest rate or exchange rate to normal values, the correction in the value of credit is much larger in the nontradables sector. Also, as prices change, the net value of the firm or the market value of collat- eral of banks' loans also change, with an additional negative effect on adverse selection and moral hazard problems. Principles of Prudential Regulation and Stability After the 1982-84 debt crisis in Latin America, new ideas about prudential regulation spread to several countries. Chile was one country that learned the lessons of the crisis rapidly and approved a completely new approach to banking regulation and supervision. Several other Latin American coun- tries have also adopted the same approach toward prudential regulation. The first element of the new standards was capital adequacy. The Gen- eral Banking Law enacted after the 1982-84 crisis did not require banks to follow the capital adequacy guidelines outlined by the Basle regulations, but a banking bill passed in the Congress during 1997 does contain such standards. The new capital requirement will take the Basle Committee guidelines as a minimum, and will incorporate incentives for banks to have stringent capital-asset ratios. Most Latin American countries have adopted standards for capital-asset ratio superior to the Basle recommendation of 8 percent based on the notion that banks in emerging markets face higher risks than banks in industrial countries. The minimum ratio in Brazil is 8 percent, in Argentina it is 11.5 Jorge Marshall 45 percent, and in the Chilean legislation it is 8 percent, but the law also en- courages a ratio of 10 percent by offering more expeditious treatment in the supervision process for international operations. Debate on what is the opti- mal capital-asset ratio for Latin American banks continues. Thne second component of the new standards was a limited deposit in- surance scheme, which was introduced in the 1980s. More recently other countries, such as Argentina, have also introduced such insurance schemes, although with different characteristics. The case of Argentina is interesting in the importance that policymakers have given to the liquidity standards of banks. Today, total liquid assets over liabilities in Argentinean banks amount to 11 to 20 percent, and if one takes into account the new system of international repurchasing agree- ment, liquid assets are close to 30 percent. The emphasis on liquidity is due to the currency board arrangement, which is linked to the free con- vertibility regime of the Argentinean peso at a fixed exchange rate. The more restricted the deposit insurance and the maneuverability of the lender of last resort, the larger the role of banks' liquidity standards. In these circumstances, any loss in confidence by depositors that may lead to liquidity problems is first addressed using liquidity held by the banks, rather than by a liquidity facility, as may be the case in countries that use their cenitral bank's lender of last resort facility. More recently, the Argentinean government created a private deposit insurance fund and a contingent international liquidity with private foreign institutions. The liquidity requirement played an important role in the Argentinean bank- ing sector after the contagious effect of the Mexican crisis. While liquid- ity is an adequate tool for dealing with contagion problems, capital and strong regulatory tools are the only solution for the kind of distorted be- havior that leads to losses in the value of bank assets. Other tools of prudential regulation are limits on connected lending and single borrowers; loan assessment and provisioning (this is an early warning system for banks); and standards on market risk exposures, espe- cially for foreign currency risk, that establish that net foreign denominated assets may not exceed in absolute value 20 percent of capital and reserves. Information generation and disclosure are other components of reforms of the regulatory framework in Chile and in other countries. The use of private rislc-rating systems and the increasing role of institutional inves- tors, such as pension funds, are complementary to the disclosure policy. Also, directors of public companies are considered responsible for the main boards' decisions. These are all steps Chile took after the banking crisis. Subsequently, Argentina has been moving fast in the same direction. 46 Lessons Learnedfrom the Chzlean Experience Limiting the scope of banks' activities may restrict risk taking. In Chile the 1986 Banking Law allowed banks to conduct a series of new businesses in addition to traditional banking. At that time banks were not allowed to keep shares in their portfolios except for completing the repayment of a nonperforming loan or for investing in their own subsidiaries. Capital as- signed to subsidiaries must be deducted from the capital of the bank. The 1986 law authorized banks to conduct several financial businesses through subsidiaries such as leasing, factoring, financial advisory, mutual funds, closed-end funds, and securities brokerage. In Chile the 1986 Banking Law established a prompt corrective action mechanism for solving insolvency in financial institutions. First, when a bank is detected to have solvency problems (measured through relevant indicators established by law), its director has to call the shareholders to vote for or against preventive capitalization. If it does not work, a creditor agreement mechanism may operate as a private solution device to bank- ruptcy. The agreement may consist of the capitalization of debt, debt for- giveness, and extension of debt maturity. The agreement will not affect the deposits guaranteed by the Central Bank (demand deposits). Only if the creditor agreement is finally rejected by the superintendency or the credi- tors themselves, the bank gets into a situation of forced liquidation. These criteria occurred during a period when macroeconomic stabil- ity started to be followed and when structural and institutional reforms were conducted, such as tax incentives to stimulate investment and sav- ings, debt-equity swaps to reduce overindebtedness, trade reform to open the economy more, and changes in the institutional framework for cor- porations to enhance corporate governance. Most Latin American countries are now introducing these principles of prudential regulation. The performance of the Chilean banking sector dem- onstrates the benefits of learning the lessons and applying the right pre- ventive measures. Table 5.1 presents an overview of the Chilean banking sector. It indicates that during 1985-97 the performance of the banking sys- tem has become more sound as a result of the new regulatory principles, macroeconomic stability, and the development of an adequate institutional framework. Provisions are the allowances for the estimated loan losses, which are based on a system that categorizes loan portfolios. These provi- sions have been decreasing over time. The leverage ratio is the debt-equity ratio. Under the 1986 Banking Law, which was recently amended, a bank's obligations to third parties (including deposits) could not exceed 20 times the bank's capital and reserves. However, to get the highest rating the le- verage ratio needed to be less than 17. This ratio has averaged around 11 or Jorge Marshall 47 12 in the recent years, with some fluctuation. The Chilean legislation has now introduced the Basle regulations for capital adequacy Nonperforming loans have been declining from about 3.5 percent of all loans in the mid- 1980s to less than 1 percent in recent years. Credit growth has been sub- stantial; however, given the levels of bank penetration into new markets and clients and the economy's high growth rate, credit growth is within safe bounds. Finally, banks' profitability has been high for international standards, although it has been diminishing in recent years. Overall Chile's banking sector is performing well. Financial stability faces no important risks or significant dangers. This behavior is also re- lated tc the economy's achievements, which are higher growth rates and lower inflation than in the past. Bank Regulation Bank regulation is a dynamic issue. Technology changes, new com- petitive forces arise, and new financial products replace old prod- ucts. This requires an evolving approach to banking regulation. The following paragraphs discuss three important issues in bank regula- tion in Latin America: international banking, disclosure and market discipline, and capital market development. Table 5.1. Bank Performance in Chile, 1985-96 Provisions Nonperforming Credit Profit (percentage Leverage loans (percentage (percentage (percentage Year of loans) (ratio) of loans) of growth) of capital) 1985 8.1 11.8 3.5 21.1 19.6 1986 7.3 9.2 3.5 18.2 13.5 1987 5.5 9.8 2.7 26.6 20.0 1988 5.0 9.9 2.0 22.5 28.4 1989 4.3 10.6 1.9 33.5 26.3 1990 4.7 9.8 2.1 18.4 22.9 1991 4.5 10.2 1.8 22.4 16.8 1992 3.2 11.1 1.2 34.6 17.6 1993 2.7 11.2 0.8 29.4 22.1 1994 1.6 11.1 1.0 14.3 20.1 1995 1.4 12.1 0.9 27.6 20.0 1996 1.3 11.8 1.0 18.8 18.3 Source: Central Bank of Chie data. 48 Lessons Learniedfrom the Chilean Experienice International Banking International banking introduces new issues into the regulation of banks. Most of them are already in practice in industrial countries, but few are practiced in emerging markets. The whole topic of international banking has changed in such markets, from the receipt of lending that originated in industrial countries to a number of new situations, including invest- ment in subsidiaries and branches abroad and cross-border banking. It is important to distinguish between two stages in the process of bank internationalization. The first stage consists of the participation of cross-border lending into Chile and the physical installation of foreign banks in Chile. That process has already taken place since the 1970s, with beneficial effects on competition, technology, and improvements in risk controls. Moreover, the country benefits from fewer problems of connected lending, good supervision from abroad, and greater stability of financial flows. A second stage of development occurs when domestic banks start to go international through cross-border operations or investment in sub- sidiaries and branches abroad. Many Latin American banks now have a regional strategy that cov- ers their operations in several countries. Others may want to lend from one country to another. Regulations must cover these new types of operations, but given their novelty, policymakers must accumu- late experience and proceed cautiously. Chile, for example, has re- cently authorized banks to finance international trade contracts to third countries and to do commercial lending abroad. Another new challenge is the supervision of branches or subsidiaries of domestic banks in other Latin American countries. International banking introduces new dimensions in the management of risk: country and exchange rate risk. In addition, international bank- ing gives rise to increased competition, which may mean more entries into and exits from the domestic banking sector. These new risks demon- strate the limitations of uniform capital-asset ratio criteria and pave the way for debate of more comprehensive, detailed, and firm-specific meth- ods for measuring and monitoring risk. Parallel and off-shore banks introduce other types of issues into in- ternational banking. These are domestic banks owned by a financial con- glomerate or banks whose headquarters are located in countries that do not have strong regulatory frameworks or good information gathering and disclosure systems. Similar problems may arise when a domestic bank or the owners of the bank own a branch or a subsidiary abroad not Jorge Marshall 49 subject to good supervision from the country of origin. This type of par- allel banking has recently been present in Ecuador, Peru, and Venezu- ela, and was the case in Chile in the 1980s. Another recent trend is the burgeoning growth of regional financial networks or organizations. Currently owners of Chilean, Mexican, and Spanish banks are buying banks in other Latin American countries. The international dimension of banking represents an important op- portunity for introducing more efficient financial services, technical inno- vations, and sounder competition; however, it is also a source of new risks that the regulatory framework needs to take into account. Under the tradi- tional framework Chile had branches of foreign banks with industrial coun- try supervisors doing most of the work. Now Chile needs to do the same work locally. This topic is the most important part of the new banking bill. Perhiaps the main concern in international banking is that a low risk country, with the lowest financing costs of any foreign country in the region, but still more expensive than international funds from indus- trial economies, funds other countries where the risk is higher. Chile does not have an advantage of low cost funds when compared to indus- trial economies, so the question is, what is the advantage of a Latin American country with respect to its neighbors? The cost of funds in any Latin American country is higher than the cost of funds in a world financial center. However, the advantage lies in having better informa- tion for companies closely linked to the domestic economy. Monitoring borrowers' behavior, producing and gathering information, and design- ing financial products may be more efficient when closer to the bor- rower, but the regulatory framework needs to distinguish between these real advantages and pure country risk intermediation. Disclosure and Market Discipline Another issue that will become important in emerging markets is the in- troduction of disclosure standards and market discipline criteria in the regu- latory process. Everyone agrees that this is essential, and most Latin Ameri- can countries are moving in this direction by asking for more information on banks' policies and performance. Moreover, banks are satisfying this requirement. However, market discipline is linked to the quality of the in- stitutions th.at read and process the information, which include financial news media, financial analysts, investment advisers, and the institutions participating in the financial system. In emerging markets these actors are less developed than in industrial economies. Under these conditions, the 50 Lessons Learnedfrom the Chilean Experience supply of more information will have an effect over time as the market increases its interest in banks' financial condition. The movement toward requiring increased disclosure of information is apparent, as well as the concern about the quality of that information. In this connection Chile has an efficient accounting system for banks and pub- licly traded corporations, and Argentina is also moving in this direction. Complementary to the disclosure of information, Argentina set up a system in which banks have to issue subordinated bonds as a way to in- crease the discipline obtained from the market. This recently launched ini- tiative will provide a device to monitor the financial condition of each bank. In Chile, private rating agencies and the official system of bank rating are playing an important role. The experience with subordinated bonds may also provide a market assessment of banks' risk and constitute an early warning indicator. While a private rating classification industry is good, a market assessment is better. The assumption behind these initiatives is that market capacity to as- sume a new role in the supervision process is available. In Chile this mar- ket capacity has been developed by the pension funds, which are active institutional investors. The pension funds have a demand for information and the professional capacity to process it. Insurance companies play a similar role. Thus the demand for information is associated with the devel- opment of a long-term capital market. If a well-developed, long-term capi- tal market exists, there will be a demand for information, and disclosure will lead to market discipline. The development of market capacity to help supervise Latin America banks is in an intermediate stage of development. Some countries are more advanced, but the institutional framework in which markets op- erate needs to be improved. This is also important given that market discipline encourages the introduction of new technologies and prod- ucts into the financial system. Capital Market Development The final issue is the relationship between the relative decline of commer- cial banking and the expansion of other financial institutions in the capital markets. This is a process that started more than a decade ago in the indus- trial countries and is more recent in emerging markets, although Chile has experienced this disintermediation phenomenon for almost a decade. The changes in the financial market structure raise several issues for banking regulation, leaving aside the effects on monetary policy. First, there Jorge Marshall 51 are operations that both banks and nonbank financial institutions can carry out, but in a different regulatory setting. The consequences are usually that banks start to increase their scope of activities, which implies that they face new risks that need to be analyzed and supervised. Second, banks and nonbank financial institutions are frequently con- nected by a holding company or a conglomerate. Thus Chile is likely to see an increase in the creation and expansion of financial conglomerates. This requires guidelines and supervision. The existence of financial conglomer- ates without strict regulation increases the moral hazard and adverse se- lection problems in the behavior of banks and related parties. This feature of financial conglomerates is prominent in most Latin American countries, including C hile, Colombia, Mexico, and Peru. The general rule should be consolidate d supervision of financial conglomerates, with global standards of capital adequacy, according to a conglomerate's risks. Another topic in capital market development is licensing criteria. Cur- rently most countries are experiencing a reduction in the number of banks and a consolidation in their structure. Some countries will probably see an increase in the number of licenses granted and in banking competition. In the 1990s in Chile the number of banks has fallen from 40 to 33 as a result of no new entrants, exits, and mergers. In the future, however, as a result of the applicat;ion of the new law, Chile will have more explicit entry condi- tions, and hence more competition. 6 Lessonsfrom Recent Global Bank Failures: The Caise of Brazil Paul L. Bydalek This chapter is written from the perspective of a leading Brazilian rating agency with easy access to the executive management of Brazil's principal banks. These insights are different from those of regulators or consultants. Bankers provide rating agencies with confidential insights into their op- erations and describe their competitive strategies, plans, and problems. By inference, the rating agency learns about the foibles of their competition. Dozens of due diligence sessions give rating agencies intimate knowledge about institutions and the system. Brazil's transition in three years from hyperinflation to predictable sta- bility put pressure on the financial system. Action by the monetary au- thorities and about 50 bank failures and other mergers provide extraordi- nary raw material for analyses and hypotheses. Observing these events gives us abundant suggestions on how to predict banker behavior, and possibly how to prevent failures. The observations that follow are based exclusively on empirical observation of the Brazilian "laboratory." What becomes evident is that Brazil's experiences, while apparently unique, are similar to those of most other countries, only taking place more rapidly Classic central bank monitoring can no longer work. Banks are too com- plex; transactions are now multidimensional and transnational; and mar- ket circumstances are too dynamic for government agencies to follow, un- derstand, and discipline. The regulator's role has changed to providing a level playing field, to preventing catastrophic disruptions, and to feeding 53 54 Lessonsfrom Recent Global Bank Failures: The Case of Brazil the marketplace with sufficient data to permit effective regulation and natu- ral selection by the market. Brazil is different from the typical Latin American country in a num- ber of ways, and one must try not to compare Brazil with Argentina and Mexico. By most economic measures, Brazil accounts for about half of Latin America. Exports and imports each represent less than 10 percent of GDP, thereby permitting Brazil the luxury of little dependence on outsiders, except for financing. There is also a vast cultural difference between Brazil and the rest of Latin America, partly because of Brazil's Portuguese roots, and partly because of its size and complexity. A grass- roots democracy in place since the mid-1980s with more than a dozen significant political parties means the need for endless political nego- tiation to accomplish any reforms. Change is slow, but lasting. Laws governing economic activity and the financial markets are many and changing, and need continuing interpretation, which reflect the country's Portuguese heritage. The business model, however, is that of the United States, where most leaders are educated. The letter of the law, not the spirit, is the rule. Exploiting loopholes spells success and accolades. Enforcement of most laws is weak, with impunity common for the privi- leged. White-collar crime is rarely disciplined. Contrary to many countries, the Central Bank does not publish data on individual banks, and gathering homogenous data over several years is troublesome. Horizontal comparisons over years require re- classification of accounts and careful selection of data. All banks must publish audited financial statements with June 30 and December 31 data and an accompanying auditor's opinion and explanatory notes in widely circulated newspapers. Collecting and collating data from the bottom up is almost the only method of monitoring the system available to analysts outside the Central Bank. Until mid-1994 intense inflation provided banks with most of their revenues and enabled them to enjoy extraordinary profits. Inflation in 1993 was more than 2 percent per working day. Currently inflation is running at less than 8 percent per year. In 1993 float income repre- sented more than 50 percent of revenues for most branch banks, but float is now less than 2 percent. In 1993 GDP data showed that finan- cial activity represented 16 percent of all activity. In 1995 this figure fell to 7 percent, is much lower today, and is falling. These events forced a contraction in the size of the financial system. Tables 6.1 and 6.2 show changes in the system during recent years and form the basis for the rest of this chapter. Paul L. Bydalek 55 Table 6.1. Number of Brazilian Banks, 1980-96 Year Number of banks a 1980 116 1]981 116 1982 119 1983 118 1984 116 1985 112 1986 110 1387 108 1988 111 1989 184 1990 219 1991 216 1992 233 1993 245 1994 246 1995 241 1996 234 a. The definition of what constitutes a bank is unknown. Source: Federacao Brasileira de Bancos' - Febraban. Table 6.2. Nvmber of Licenses Granted to Brazilian Financial Institutions by Type of Institution, June 1994 and April 1997 Type of institution June 1994 April 1997 Commercial banks 34 38 Multiple banks 212 189 Development banks 6 6 Investment banks 17 23 Savings banks 2 2 Cooperative societies 853 984 Finance companies 42 46 Securities brokers 244 205 Exchange brokers 43 39 Securities dealers 371 266 Investment societies 4 2 Leasing companies 67 74 Mortgage companies 24 20 Savings and loan companies 2 2 Source: Banco Central do Brasil. 56 Lessonsfrom Recent Global Bank Failures: The Case of Brazil The 1982 Mexican crisis had little effect on Brazil's local banking sys- tem, reflecting the country's self-sufficiency. In the mid-1980s three major banks closed. At the time, rumors about problems at Bamerindus, Economico, and Nacional were circulating. Only in late 1994, with stabili- zation, did the number of banks begin to fall. Before the 1988 constitutional reform the number of banks was roughly constant, with virtually no new licenses granted. The new constitution of 1988 permitted anyone with reasonable credentials and less than US$5 mil- lion in capital to open a new bank. Many brokers became bankers, and given the prevalent intense inflation and instability, the primary activity of banks and brokers was trading in government securities and interest arbitrage. As frequently occurs in Brazil, data sometimes conflict, and data from the Central Bank inexplicably differs from that shown in table 6.1. Part of the difference resides with the definition of banks. Brazil has both pri- vately owned banks and government owned banks. Each can be divided into other categories, particularly multiple (or universal), commercial, savings, and investment banks. As table 6.2 shows, during 1980-96, on a net basis, 23 multiple banks closed, while 4 commercial banks and 6 in- vestment banks opened. Brokerage operations suffered the greatest attri- tion, falling by 148 to 510. Brazilian securities are now traded in New York, international securities firms are now operating in Brazil, and larger Brazilian banks are spreading into brokerage. Since mid-1994, about 50 banks have closed, most with intervention by the authorities. Others have merged. Simultaneously, new banks have been opening, some sponsored by private Brazilian capital, but most with inter- national backing. Although the 1988 constitution freezes foreign participa- tion in the system, if national interests are well served, the president can grant special licenses to international banks. President Cardoso has autho- rized Rabo, Banque National de Paris, Morgan Stanley, CS First Boston, Merrill Lynch, and Ford, among others, to open wholly owned operations and granted joint venture approvals to others. The disappearance of inflation and the return to traditional bank- ing caused many disruptions and consolidations within the local sys- tem. The arrival of international competition will narrow margins fur- ther. Banks without a well-defined strategy and niche will disappear during the next few years. During 1985 through mid-1997 the Central Bank had 11 presidents man- aging monetary policy for Brazil. Thus during the period of intense infla- tion and numerous new bank openings, the turnover of Central Bank presi- dents was high. In 1989, when 73 new banking licenses were granted, the Paul L. Bydalek 57 Central Bank had two presidents, each of whom served for six months. Continuity of management is a valid business principle that was ignored during these turbulent years. It is foolish to expect a political appointee, whose career might be short-lived, who is injected into a complex manage- rial situation and subject to myriad pressures from politicians and busi- ness people, to introduce long-term discipline and lasting change. Only when the second and third echelons of government are convinced and re- ceive strong leadership will lasting change actually occur. Five years ago the Banco do Brasil, Banespa, Bamerindus, Economico, and Nacional were among Brazil's most prominent banks, but all have suffered in the intervening years. Since 1995 the government controlled Banco do Brasil has conducted itself much like a private bank, avoiding political involvement and loans. Cleaning up the prior abuse caused the write-off of billions of dollars in loans and the need for a capital injection of US$8 billion in early 1996. The Central Bank intervened in Banespa, owvned by the state of Sao Paulo, in December 1994, and no published financial statements have been available since then. Losses exceed US$25 billion, with ownership now being transferred to the fed- eral government for privatization. The government seized Bamerindus, Econ6mico, and Nacional and passed them on to private investors. In addition, the government used more than US$20 billion for bailouts for seven private banks under a special program. During 1992-96, as seen in tables 6.3 and 6.4, private banks at least doubled their equity, basically by retained earnings. Citibank was the largest international bank. The performance of Excel was phenom- enal, with equity expanding fivefold following its acquisition of the defunct Econ6mico together with minority investors. Since 1994 Banespa and other government owned banks have been withering, losing market share, facing up to enormous bad asset problems, with most destined to become defunct. Note that in the mid-1980s a banking crisis forced the closing of several institutions. Bancos Bamerindus, Economico, and Nacional were cited as fragile and as targets for a takeover or intervention. Politically connected owners bought support and a second life. A decade later the inevitable occurred and all three closed. Even before the onset of stability in July 1994, loans were expanding as dollar funding into Brazil began, with the largest increase occurring dur- ing the second half of 1994. Deposits plateaued, with major growth occur- ring off balance sheet as money market funds. These funds, always man- aged by banks, currently total about US$120 billion. All government owned 58 Lessons from Recent Global Bank Failures: The Case of Brazil Table 6.3. Largest Brazilian Banks by Equity, December 1992 Bank US$ millions Banco do Brasil 6,178 Bradesco 2,508 Itaui 1,818 Banespa 1,212 Real (group) 711 Unibanco 603 Bamerindus 602 Mercantil de Sao Paulo 484 Econ6mico 437 Nacional 418 Excel 48 Total 15,019 Total less Banco do Brasil 8,841 Source: Atlantic Rating. banks and many mid-sized commercial banks have encountered diffi- culty in changing with the macroeconomic situation. Many bank owners would like to sell, but buyers are few and selective. As time passes, these owners must decide how to abandon banking, retrieving their capital and redeploying it elsewhere. Postponement of this decision will cause future bank failures. Table 6.4. Largest Brazilian Banks by Equity, December 1996 Bank US$ millions Growth since 1992 (percent) Bradesco 5,384 114.7 Banco do Brasil 5,380 -12.9 Itau 3,868 112.8 Unibanco 2,085 245.5 Real (group) 1,543 116.9 Bamerindus 1,270 111.0 Mercantil de Sao Paulo 985 103.7 BCN 937 145.3 Safra 715 150.9 Citibank 569 142.1 Excel Economico 510 962.5 Total 23,246 54.8 Total less Banco do Brasil 17,866 102.1 Source: Atlantic Rating. Paul L. Bydalek 59 System Wleaknesses System weaknesses include the following: protected activity until 1994, loose enforcement, impunity, a strong lobby, and no credit culture. Until 1989 virtually no new banks could be formed, with the Central Bank being extremely selective about licenses sold, even auctioned, or awarded. The constitutional reform in 1989 permitted virtually anyone to start a bank. Until 1994, banks faced few challenges beyond inflation, and most reported handsome profits during these years. Inflation suspended classical banking. Lending in local currency was short term, 30 days, to match certificate of deposit funding. Inflation camouflaged problems; de- linquencies would wither and disappear. Banking was money brokering. With few problems, system monitoring was lax and enforcement of regu- lations loose. Until 1986 no legislation defined crimes against the financial system. A government official recently reported (O Globo newspaper, May 18,1997) that during 1986-95 the Central Bank identified 682 crimes against the financial system sufficient for reporting to the justice system for pros- ecution. Only three crimes resulted in condemnation. Impunity exists today for the bankers involved in bank failures since 1994. While several investigations by the justice system are under way, only one banker is under house arrest pending completion of investigations. In general, white-collar crime goes unpunished. Another characteristic of Brazilian banking is that the Central Bank maintains a tight and complex net of regulations over the system, but bank- ers are encouraged to find and exploit loopholes. Bankers are expected to fulfill the letter of the law, not the spirit. Loopholes are windows of oppor- tunity. Creativity is rewarded with profits. The bankers' agility is far greater than that of the regulators. Bankers also enjoy an unusually strong lobby in Congress and with the executive branch of government. The banking associations are effective in obtaining favorable legislation for activities. Nonetheless, banks pay higher income taxes than other commercial and industrial activities. Inflation prevented the development of a credit culture. Financial pro- jections were deemed nearly worthless, borrowers' financial statements were considered to be unreliable, and funding was short term. Cash flow lending did not exist. As stability arrived in mid-1994, banks expanded loan assets, usually to the consumer or to the middle market. Money man- agers became lenders. Many subsequent bank failures were caused by fool- ish lending by inexperienced bankers. 60 Lessons from Recent Global Bank Fa0lures: The Case of Brazil System Strengths System strengths can be summed up as follows: bankers tested by 2 percent per day inflation; technology denial until 1991, followed by catch-up activ- ity; a society that embraces change; a cashless society with all payments in banks; few loans, high Basle ratios; no dollarization ever; a crawling peg exchange rate; audited statements twice a year; and grass-roots change. The Brazilian commercial and financial system (universal banks) is largely patterned after the U.S. model. A major difference is that banks with multiple licenses can engage in virtually any capital market, commer- cial banking, investment banking, credit card issuance, stock exchange bro- kerage, or leasing activity. Most banks are now acquiring life and casualty insurance subsidiaries and moving into pension fund management. The banks manage all money market mutual funds. The country is sufficiently large and diversified so that connected lending inside Brazil is uncommon. While abuse does occur, enforcement is efficient. The intense inflation of the 1980s and early 1990s, combined with churn- ing politics, tested all private entrepreneurs. Most survived by cutting their dependence on the government and cultivating self-sufficiency. Those bank- ers who are prospering today are generally the same professionals that op- erated during those turbulent times and are proven, competent managers. Until 1991 Brazil barred imports of software and hardware. With the relaxation of barriers, state-of-the-art technology was imported, always for the front office and sometimes for the back office. Substantial gains from float motivated high investment in technology to control money flows. The technology of some institutions is now superior to that of their northern hemisphere counterparts. Possibly because of this prior denial of imported technology, the culture embraces change. Cellular telephones, the Intemet, and home banking are everywhere. The absence of obsolete equipment to discard has meant that local companies and individuals crave the most modern technology and are almost oblivious to its cost. Individuals and companies make payments for most account receiv- ables and payables through banks. Checks are rarely sent by mail for ac- count settlement. Salaries and collections are credited by banks. Banks also collect rents, tuition, utility bills, and taxes as well as other predictable pay- ments. Salaries of low-income workers are credited to bank accounts and accessed by ATM machines. People use checks to pay for very small pur- chases. Inflation created a cashless society. Until the advent of stability, loan assets were a small proportion of total bank assets, and consequently capitalization ratios were lofty. Loans are now climbing and represent about 50 percent of the system's assets. Paul L. Bydalek 61 Correspondingly, capitalization ratios are falling. Nonetheless, Bank for International Settlements ratios for some banks exceed 20 percent, and for most banks exceed 12 percent. The loan delinquencies of early 1995 and the various subsequent failures intimidated bankers, who now undertake growth of loan assets more carefully. Most Latin American economies permit transactions within their bor- ders in local currency or in dollars. Even bank deposits are in both curren- cies. Brazil has always avoided dollarization of in-country transactions. With few exceptions, transactions must be in local currency. Since the in- troduction of the real in mid-1994, it has been the consistent measure of value, with the dollar used as a remote reference. Brazilians measure value in their own currency, and most people ignore the dollar exchange rate. The crawling peg exchange rate dissipates pressure for devaluation. Currently the currency is devaluing against the dollar at a rate close to changes in the domestic wholesale price index and price indexes of in- ternational trading partners. Authorities insist that despite building pres- sure, there will be no sudden change in the exchange rate to avoid re- kindling an inflationary psychology. Banks publish complete financial statements twice a year along with opinions by outside auditors, thereby enhancing credibility. Unfortunately, the Central B3ank does not provide individual data on any bank to the mar- ketplace, as is common elsewhere. Change is occurring within all segments of Brazil, from the ground up, and not by decree. Society is negotiating change, with lobbies active every- where. While this seemingly endless discussion results in slow forward movement, change is lasting. The various bankers' and industry associa- tions constantly discuss with the authorities how to improve the system and to prevent new failures. Dialogue, negotiation, and a bottom-up de- mocracy have replaced the dictatorial years of the past. Management Strengthening Proposals for improving management call for the following: lower turn- over of senior staff and greater stability, freedom from political pressures, clear mission and immediate objectives, power to attain most objectives, and accountability. Administration of the monetary system is a complex undertaking that requires unusual talents. For many years a hiring freeze on new govern- ment staff has been in place, so the actual number of staff is falling at a time when needs are greater than in the past. The high turnover rate of Central 62 Lessons from Recent Global Bank Failures: The Case of Brazil Bank directors is the primary issue. The handful of directors and their sup- port staff set policy and liaise with other parts of the government and the public. The remuneration paid to Central Bank directors is small compared to their earning potential in the marketplace and insufficient to compen- sate for the harassment from politicians and the public when they announce unpopular decisions. Besides, as their track record shows, a few months in office is a guarantee for later financial success. Managerial stability by pro- fessionals must be structurally encouraged. The centralization of economic powers in the Central Bank and other executive and congressional departments encourages lobbying. Every municipality, state, agency, and economic association knows that intense pressure on the few decisionmakers can dilute new executive acts. The sheer number of executive orders issued offers unusual potential for lob- byists. Central Bank policymakers should be exempt from political pres- sures, possibly by fixed mandates, and by not serving at the mere plea- sure of the executive branch. Also needed is a clear, long-term mission with concise objectives that transcends individuals and directorates. Removing the personality fac- tor from decisionmaking would permit the monetary machine to operate more efficiently. Today, the personality of the finance minister and the president of the Central Bank often determine policy and its method of execution. As people change, policies change. Brazil's Central Bank both monitors monetary aggregates, fine-tuning macroeconomics, and supervises institutions. Political pressures spill over into the supervisory function. A split-off of the supervisory function into an autonomous unit will not automatically solve the political issue. Banks with strong patronage, all government banks, and some commercial banks with political connections will still be protected from discipline. The au- thorities should have dealt with Bancos Bamerindus, Econ6mico, and Nacional severely in the mid-1980s, but because of strong political and re- gional pressures, each escaped to explode later on a much grander scale. Central Bank inspectors could not enforce discipline, thus problems fes- tered and grew. Without power to enforce regulations for all institutions, those with political protection are free to continue abusing the system. Af- ter an institution reaches the point of too big to fail, the government loses control totally Later, the owners of failed institutions are rarely held re- sponsible for their professional negligence, and retire from banking wealthy, leaving society to bear the onus of their behavior. As government enforcement is lax and white-collar crime occurs with impunity, possibly the marketplace could address these cultural failings. If Paul L. Bydalek 63 sources of funding for institutions-large depositors, debenture holders, and banks financing international trade-were to hold score cards for each bank, these sophisticated investors could monitor and induce change. As small problems occur within an institution, for instance, as the volume of renegotiated loans without cash inflow increases, the marketplace should somehow be notified. Subtle differences in delinquencies among institu- tions during a business slowdown would permit an entry on the score card. Investors should also monitor management actions to cure incipient prob- lems. Variations between institutions would affect the cost of funding, pe- nalizing the less professional organization and squeezing it back into con- formity. The informational flow to the marketplace must be constant and thorough, permitting early identification of changes. Sudden discovery of problems when they are already pronounced could damage confidence in that institu tion, causing continuity problems. An increased flow of facts to the marketplace will permit efficient regulation via liability pricing to banks. Reduced Political Pressure Political pressure on the system needs to be reduced through long-term, staggered raandates; clear quantitative limits imposed on the system; au- tomatic, reactive discipline if limits are breached; public censure, fines, and suspension for breachers; and early public disclosure of problems. Professionalizing and depoliticizing management of the monetary system can cure the high turnover of top executives. The system should look for professionals serving long-term, staggered mandates who are protected from political caprice. The need for constant interpretation of the multitude of regulations imposed in trying to close the windows of opportunity found by agile pri- vate bankers provides unusually great opportunities for abuse. Numerous and ever changing regulations, coupled with lax enforcement, cause bank- ers to test their limits, to push opportunities to the point of resistance of monetary authorities, who respond with subjective judgments and cen- sures. Some limits, if breached, should have automatic and predictable con- sequences, for instance, nonperforming loans that exceed predetermined limits should automatically impede the expansion of total loans or bring about a tightening of loan-to-equity ratios, and should not require negotia- tion between authorities and the bank. If automatic corrective action does not follow small breaches of limits, then banks arLd bankers should suffer veiled public censure, thereby per- mitting the miiarketplace to sniff out improprieties. Continued breaches 64 Lessonsfrom Recent Global Bank Failures: The Case of Brazil should result in changes in management, overt public censure, fines, and suspension of selected activities. As the justice system is slow, the Central Bank should be empowered to discipline breaches of Central Bank regula- tions, with denial of access to the system and other internal measures. Gradual and continual disclosure to the market of improper conduct, with early veiled criticisms later transformed into loud action, in a highly pre- dictable format, would pressure bankers. The efficiency of gradual market pressure is far greater than the quiet and invisible action that supposedly results from traditional Central Bank discipline. Deposit Insurance A deposit insurance scheme with a small deductible would put pressure on deposit authorities and cause depositors to be vigilant and not passive. Until 1994 Brazil had no deposit insurance, but with the many failures it was introduced. The new system is laudably managed by the banks them- selves, with mandatory monthly contributions calculated on deposits, in- suring 100 percent of all deposits up to about US$20,000. The government covers shortfalls at the insurance authority. The knowledge that their de- posits are insured reduces the need for vigilance by small depositors. A coinsurance feature that requires even small depositors to suffer a slight loss would be healthy. With the general public oblivious to bank problems, smaller depositors operate in confidence. The three large failed banks cited earlier suffered little loss in savings deposits during the months preceding their closing, even with newspapers aggressively calling attention to their problems. Some migration of even small deposits from weaker to stronger institutions would cause bank management to be more responsive to the need for quality stewardship and worry the deposit insuring authority. Pressure from the market by small and sophisticated depositors and by peers should focus on institutions with deviant performance. The deposit insurance authority should pressure both the Central Bank and the prob- lem institution to take prompt remedial action before losses occur. Transparency Transparency would be increased by semi-annual and quarterly up- dates, specially trained public auditors, standardized scope of audits set by authorities, personal liability placed on auditors and bank direc- tors, standardized form of disclosure, concise and forward looking pro- visioning regulations, restructuring reported when problematical, and encouragement of rating agencies. Paul L. Bydalek 65 The marketplace discipline needs information to operate. A continu- ing flow of abundant, reliable data from each institution will permit the market to compare institutions, rewarding and penalizing as appropri- ate. Currently, all financial institutions must publish half-year financial statements, with accompanying explanatory notes and an auditor's opin- ion. Only banks with shares listed on stock exchanges must publish quar- terly data. The quality of this reporting varies. A few confident and solid banks offer quarterly consolidated data audited by international audi- tors. Most offer unconsolidated data, with data about on-shore and off- shore subsidiaries reported as investments. They offer virtually no data about the size or quality of these investments. The complications of auditing a bank demand specially trained audi- tors, yet banks can use any of hundreds of auditors, many of whom lack the requis:ite specialized qualifications. While the audits fulfill the mini- mum standlards set by the authorities, the scope of audits and the transpar- ency of reporting vary widely. The scope of audits and the financial disclo- sure to the market should be standardized, with the common denominator substantially above current levels. A large, solid, and liquid institution should abide by the same criteria as a small, struggling institution, with both providing abundant, standardized disclosure. Quarterly, detailed, consolidated financial statements with opinions by auditors would permit efficient market regulation. The banks would bear the cost of the increased disclosure and auditing requirements. Standarclized disclosure will facilitate comparison to peers. Sophisti- cated investors will be empowered to price funding objectively and accu- rately. Any deterioration of basic ratios, such as the risk-asset-based capital ratio, would have immediate impact on funding costs. Coupled with disclosure should be concise quantitative benchmarks. Bank management and directors and sophisticated investors can measure data against these benchmarks. Outsiders can make realistic judgments about performance. Subjectivity for provisioning for doubtful assets should be reduced, with basic rules for recognizing deteriorating assets clear, with little interpretation required. Cash inflow from a loan should be a basic quality criterion. Doubtful loans performing only on an accrual basis (no cash received) should be booked in financial statements clearly and sepa- rately. In addition, restructured loans, where management is concerned about ultimate payment, must be specially recognized on the balance sheet. Loans performing normally should be separated from those with signs of loss. The wealth of data made available by banks could overwhelm deposi- tors and investors, thereby camouflaging problems. Large institutional in- vestors have the manpower to track complex data. To help less sophisticated 66 Lessonsfrom Recent Global Bank Fazhlires: The Case of Brazil investors, the authorities should encourage the presence of rating agencies. A rating agency specializing in banks soon acquires an in-depth understand- ing of the game playing bankers can indulge in. Peers call attention to their relative strengths and successful strategies, giving the insightful agency hints about where to search for problems elsewhere. Agencies meet with top ex- ecutives of banks several times a year, and are able to query management about the confidential circumstances surrounding decisions. The agency with this confidential information, and after visits to peers to obtain other interpretations of market events, enjoys a unique perspective on the sys- tem. Conclusions expressed as ratings reinforce judgments made by sophis- ticated investors and are benchmarks for smaller investors. Summary Chemotherapy is required, but not life-threatening surgery The adage "an ounce of prevention is worth a pound of cure" is appropriate in this instance. Many of the problems of the Brazilian system could have been avoided with better disclosure by authorities and by banks. If enforcement by the authorities is lax or impossible, then let the market undertake part of this discipline. With continuous, standardized, and reliable disclosure, the market will reward and penalize. Nonstandardized and unreliable disclo- sure is the responsibility of the authorities. The market will never detect deceit, at least in the early stages. Clever accounting practices and smooth talking management will delude the market until problems become too significant to hide. When an efficient market cannot function, the problem becomes one that needs a Central Bank solution. Full transparency all the time should be a requirement. Treating small problems as they develop and appear reduces the danger of catastrophes later. When bank manage- ment is under the scrutiny of the market and peers, poor performance will be challenged early. Complex banking and modem informational technol- ogy change the historic regulatory role of the Central Bank. Part III. Bank Failures in Transition Economies 7 Restructuring Distressed Banks in Transition Economies: Lessons from Central Europe and Ukraine Michael Borish and Fernando Montes-Negret The collapse of central planning in Central and Eastern Europe (CEE) and the former Soviet Union (FSU) and the opening of their formerly socialist economies to the West presented enterprises and banks in these countries with an inescapable imperative: the need to undergo radical restructuring- by means of privatization, divestiture, liquidation, and/or reorganization. This had a number of consequences for the countries' banking sectors. Macroeconomic Transition from Socialism and Implications for Banking Sectors One outcome was that state-owned enterprises (SOEs) were suddenly and simultaneously faced with imported consumer goods of higher quality from the West, sharp reductions in state subsidies, the disappearance of tradi- tional Council for Mutual Economic Assistance (CMEA) markets for out- put, and interruptions in the supply of critically needed energy imports for their top-heavy industrial sectors. In the worst case scenarios, econo- mies experiernced dramatic cuts in output because of the collapse of mar- kets. These cuts were characterized by an initial period of inventory sell- offs and barter trade that ultimately culminated in arrears on most liability accounts (for example, wages, benefits, taxes, social security, bank debt, trade debt) as inventories and output subsequently diminished. 69 70 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine Faced with insufficient working and investment capital, many SOEs turned to banks to obtain credits that allowed them to escape hard bud- get constraints temporarily and to defer needed restructuring or liqui- dation. In some cases, hyperinflation mitigated the balance sheet effect, rewarding enterprise debtors for taking on obligations in currencies that were depreciating on a daily basis. However, the effect on the loan port- folios of state owned commercial banks' (SOCBs) was rapid deteriora- tion. SOCBs had already inherited the burden of risky portfolios from the former monobank system. I The second stage of lending to SOEs led to a significant growth in nonperforming loans-in many cases unrec- ognized until new accounting standards and prudential regulations were in force-and a contraction of liquidity as monetary policy became re- strictive to rein in the sometimes devastating and always damaging ef- fects of inflation.2 In CEE countries that were not part of the FSU, the banks' deteriorating position resulted in an increasing, but unquantified, liability for the state, as implicit deposit insurance and vested SOE in- terests looking to banks for financing precluded the closure of large SOCBs. In FSU countries, hyperinflation eliminated asset-liability val- ues, but continued patronage of loss-making enterprises, initial major refinancings from central banks, and the run-up of arrears failed to pro- vide the needed resources and confidence to put transactions on an ef- ficient market basis. In practice, governments pursued a range of approaches to deal with these problems. This chapter assesses banking sector restructuring (or the lack thereof) under distressed conditions in four countries: the Czech Re- public,3 Hungary, Poland, and Ukraine. All four of these countries have taken different approaches, although broad differentiations are apparent between 1. These portfolios were risky in three general ways: (a) heavy concentration of exposure to SOEs with large arrears to SOCBs, other SOEs, workers, and even to the state (taxes, pensions); (b) geographic risk, given the fragmented and specialized way in which the monobank system was split up; and (c) product risk (financing of indus- trial and arms exports to and construction projects in countries that sometimes proved to be significant credit risks). 2. Annualized inflation rates were as high as 4,735 percent in Ukraine in 1993 and 586 percent in Poland in 1990. Peak inflation rates were much more moderate in Hun- gary, 35 percent in 1991, and in the Czech-Slovak Federal Republic, 57 percent in 1991. 3. At the time of transformation, the Czech Republic was actually part of the Czech- Slovak Federal Republic. On January 1, 1993, the republic was officially disbanded and two independent countries were established: the Czech Republic and the Slovak Repub- lic. This paper refers to the Czech Republic, but events prior to 1993 refer to the Czech- Slovak Federal Republic or to the earlier period of central planning of Czechoslovakia. Michael Borish and Fernando Montes-Negret 71 most FSU and CEE countries, which are highlighted in table 7.1. However, even wiftin these categories, major differences exist. Ukraine differs widely from Estonia, and more recently Latvia, in its approaches and results, even though all three belong to the community of FSU countries. Likewise, ap- proaches taken by the Czech Republic, Hungary, and Poland differed in the early 1990s, although they are now converging in some ways as the coun- tries address structural, institutional, and policy weaknesses. The Socialist Incentive Structure At the outset of transformation, the CEE and FSU economies were highly distorted. The level of distortion varied, and was probably much less in Hungary, Poland, and the former Yugoslavia as reflected in levels of trade with Westeem markets in the 1980s and the higher proportion of private farming and land ownership compared with other CMEA countries.4'5 How- ever, as a general rule, the sectoral distribution of economic output was highly focused on industry. While agriculture was important for food se- curity and as input into the agro-industrial sector, services were virtually nonexistent except in the form of transportation of and warehousing for inventories.6 (Common to all transition countries has been the rapid emer- gence of the private sector in services, which has made up for this vacuum without ha ving to compete with vested industrial and state farm interests from the socialist period [see Borish and Noel 19961.) Market-based con- sumer criteria in the domestic economy were subordinated to central plan- ning prerogatives, although some exposure to global standards and qual- ity requirements was evident, insofar as trade with non-CMEA markets existed. Market-based production and costing systems were absent because of constraints imposed on commercial competition. These characteristics reflected the organization, operations, and incentive structures of centrally planned economies whose primary focus was on production and special- ization based on predetermined physical targets and plans, which were driven by Co.mmunist Party criteria and disregarded financial performance. 4. For instance, in 1989, exports of goods and services from Hungary and Poland to the European Community approximated US$3.2 billion and US$4.6 billion in value, respectively Since the transformation, trade with the European Union has typically come to account for two-thirds of total exports. These values are now generally about three times the export values achieved in 1989-90. 5. In Poland, small farm plots remained private during the communist era. Like- wise, the former Yugoslavia permitted private ownership of small farm plots and housing. 6. Yugoslavia had an active tourist trade in the 1980s, as did some of the Black Sea countries. Nevertheless, these were more of an exception than a rule throughout the CMEA. 72 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine Table 7.1. Initial Effects and Approaches to Banking Reform in Formerly Socialist Counitries Category Central and Eastern Europe Former Soviet Union Inflation Less problematic in Hungary and Erased real balance sheet Czech Republic; shock therapy in values Poland Privatization Gradual in most cases; rapid but Changed ownership struc- partial in the Czech Republic; recent ture, but not governance acceleration in Hungary Entry Proliferation in Poland by 1992, Few new banks net of mono- slowdown thereafter; increase in bank spinoffs Czech Republic and Hungary Liquidations Implicit deposit insurance; small Banks allowed to fail in some bank failures only, and these have cases; sometimes depositors been few assumed costs, in other cases banks continue to operate even after revocation of their licenses Recapitalization Single (Poland), and multiple Infrequent (Czech, Hungary) recapitalizations, usually via fiscal methods (bonds) Enabling Reasonably open competition; Local monopolies in some environment foreign investment materialized regional markets; weak court (mainly in Hungary); some pro- systems, bankruptcy, and tection of specialized banks (sav- collateral laws ings, housing, agriculture) Regulation and Improving licensing and super- Weak legal framework and supervision vision through significant technical licensing standards; insuf- assistance in Poland; weaker per- ficient supervisory enforce- formance in Czech Republic and ment Hungary; limited resolution capacity Banking skills Limited but improving quickly with Weak or absent, but irnproving foreign investment (Hungary), technical assistance (twinning in Poland); contracts with West European banks Motivation Accession to the European Union Quick profitability; specula- tive period with significant 'gray" economy a. Recapitalization in Poland refers to the state commerical banks, but not to the speciahzed banks, namely BGZ (agriculture), PKO BP (local currency savings and housing), Bank Handlowy (foreign commerical trade), PKO SA (foreign currency savings), PBR (the Polish Development Bank), and BRE (export finance). Michael Borish and Fernando Montes-Negret 73 Few CMEA countries deviated from the general tenets of this model, al- though the extent of centralization, of exposure to non-CMEA markets, and even of private ownership varied. Consistent with such a model, socialist economies lacked banks in the market sense. Banks were institutions that served as accounting control and cash disbursement vehicles in support of larger economic planning and investment requirements. The role of state owned banks-which until the late 1980s were usually part of the monobank system-was twofold: (a) to disburse funds or transfer payments passively through the noncash circuit to other SOEs upon instruction from line ministries, and (b) to pro- vide basic pension entitlements and other savings services to enterprises and their employees. Disbursements to SOEs were made without regard to creditworthiness or riskiness. For liquidity, state enterprises reimbursed banks upon instruction from government ministries, but only after enter- prise allocations were made for production and the provision of benefits and social services to employees and their communities. When SOEs lacked resources to reimburse banks, the government balanced the account. When the goverrnent failed to replenish bank resources from the national trea- sury, banks accumulated and rolled over large stocks of loans that would have been technically in default under market conditions. Likewise, asso- ciated interest income from such loans accrued and was capitalized, add- ing to the size of the overall refinancing. This mode of operation distorted the management and pricing of monetary and fiscal resources, failed to instill financial discipline in banks and SOEs with regard to resource scar- city, ultimately led to a collapse of socialist economies, and made the in- terim transition more difficult. Box 7.1 highlights key characteristics of the overall collapse, recognizing that these effects differed across countries and were less damaging in many Central European countries that had begun to open up to the West than in other socialist economies that had not. Reduced Liquidity and the Run- Up of Arrears Macroeconomic decline (table 7.2) manifested itself in a general deterioration of loan quality, particularly in Poland and the FSU countries. These problems, all brought to liglht by initial reforms, represented a combination of external factors (inflation rates, foreign exchange rates and reserves, trade liberalization, removal of subsidies, collapse of the CMEA) that were sufficient by themselves to erase portfolio values, as occurred in the FSU countries. Hyperinflation wiped out real household savings (in Ukraine; in Poland, households held more foreign cur- rency), and household distrust or liinted liquidity kept savings out of the bank- ing system (FSIJ, Poland), causing a disintermediation effect. The Czech Republic Table 7.2. Macroeconomic Indicators of Decline in Transitional Countries, Selected Years Czech Reputblic Hungary Poland Ukraine Indicator 1989 1991--92 1989 1991-92 1989 1991-92 1992 1993 Average inflation rates (%) 1.4 56.6 17.0 35.0 245.6 70.3 - 4,734.9 Average exchange rate' 15.1 29.5 59.1 74.7 0.1 1.1 749.0 25,000.0 Unemployment (%) 0.0 3.4 0.0 10.4 0.0 11.5 0.3 0.4 Foreign exchange reserves (year end) b - 2.8 2.2 5.0 2.1 3.2 2.7 2.0 Merchandise trade (US$1 billions) - 32.2 26.3 24.9 28.3 28.9 23.2 28.1 M2/GDP (%) - 75.7 41.1 46.9 81.1 31.8 - - Fiscal deficits/GDP - 0.4 1.3 5.4 6.0 6.7 24.2 11.7 -Not available. a. Local currency units to the U.S. dollar. b. Expressed in terms of months of imports of goods and nonfactor services. Source: International Labour Office, International Monetary Fund, and World Bank data. Michael Borish and Fernando Montes-Negret 75 Box 7.1. Characteristics of the Collapse of Socialism Regional Trade Domestic, regional, and CMEA trade relations collapsed, resulting in drastic interrup- tions in production and distribution. This was particularly the case with energy, where critical supplies of power were reduced because of countries' inability to provide hard currency in exchange. For Ukraine, this led to barter trade, which was less efficient and reliable. For the Czech Republic, Hungary, and Poland, merchandise trade (exports and imports) was more stable, and eventually increased because of the shift in trade patterns away frcm the centrally planned economnies to those of Western Europe. Output, Employment, and Incomes Domestic output, formal employment, and real incomes all declined sharply. Allowing for weaknesses in statistics, GDP fell nearly 15 percent in real terms from 1989-92 in non-FSU countries, and even more in the FSU region. In most transitional countries, industrial employment has fallen by at least one-third, and unemployment rates still range in tie 10 to 20 percent range. (Official rates may overstate unlemploymentbecause of the growth of the informal sector in transition countries. However, significant under- employm,nt among those officially employed is often not captured.) The Czech Repub- lic remain; a noteworthy exception, with unemployment rates in the 3 to 4 percent range, although this may be due to the absence of deep industrial restructuring. Ukraine's offi- cial unemployment rate had not exceeded 0.4 percent by 1994 despite economic col- lapse, reflecting anachronistic statistical indicators. Pricing and Production Energy prices increased sharply because of reduced subsidies, which resulted in explic- itly higher production costs and output prices. Inflation Rates and Purchasing Power Inflation ra tes increased dramatically in Poland and Ukraine, while in the Czech Repub- lic and Hungary inflation rates were more moderate. In the FSU (Ukraine included), this resulted in :reduced purchasing power and reduced values of real fixed income. In Po- land, fiscal deficits reached nearly 7 percent of GDP from 1991-92, largely because of social assistance payments. Source: World Bank data. experienced greater stability because of a disciplined macroeconomic framework and the public's confidence in irnplicit deposit insurance. Hungary was similarly less "shocked" because of its gradual opening to Western markets that dated back to the 1960s. However, in all four countries, production declines and limited trade with hard currency markets led to a decrease in foreign exchange resources. Mean- while, domestic currencies lost value, particularly in ruble-based economies. The zloty likewise lost value in Poland. This prompted a chain of events in which cash payments diminished, lack of creditworthiness and competitiveness became more apparent, and the introduction of hard budget constraints became a necessity for renewed stability and improved resource allocation at firm levels. 76 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine Governments imposed hard budget constraints on banks and enterprises in varying degrees and at varying times for a number of reasons. From a policy standpoint, the objective was to introduce a measure of disciplined financial management under market conditions to restore macroeconomic balance and to reverse structural weaknesses so that companies (and coun- tries) could compete. From a fiscal standpoint, it was a necessity because of the shrinking tax base. Profits taxes assessed earlier were no longer gen- erating sufficient revenues, so governments had to cut subsidies and sup- port (see Barbone and Marchetti 1994). Another contributing factor in CEE countries was the need to make benefits payments (pensions, social secu- rity). These were particularly high in Hungary and Poland, and were re- flected in fiscal deficits. FSU countries including Ukraine were unable to honor these commitments: inflation wiped out real savings and fiscal re- sources were far scarcer than in CEE countries. In Ukraine, scarce fiscal resources were instead used for subsidies and directed credits for agricul- ture and industry in an attempt to maintain production and jobs. As budget constraints were eventually imposed, enterprises increased the use of barter in their dealings, and ran up arrears on their debts (ar- rears were particularly pronounced on trade debt to domestic and for- eign suppliers, tax and social security payments to the government, wage payments to employees, and both principal and interest to banks). These problems were particularly severe in Poland and Ukraine, although prob- lems emerged in the Czech Republic and Hungary as well. In Poland and Ukraine this led to an interruption of production and distribution flows, as the quality of receivables diminished, cash was short, price increases were built in to products to cover for some anticipated delays and losses, and orders were taken to utilize capacity and keep people partly em- ployed instead of reducing excess capacity and overheads. (Fewer dis- ruptions occurred in the Czech Republic and Hungary, because of greater investment and remittance flows into these economies.) In countries where disruptions were the greatest, a broad deterioration in the market for bank services resulted, and this prompted a tightening of lending conditions. In the case of Ukraine (and other FSU countries), these developments were particularly severe, affecting trade relations with FSU countries and prompting the need for a series of debt restructurings (debt restructuring with Russia and Turkmenistan were required to permit needed energy resources to flow into the country). Meanwhile, real household incomes and savings declined, leading to diminished confidence and flight from local to hard currencies. This slowed bank deposit mobilization, except in the Czech Republic, where confidence remained high despite negative Michael Borish and Fernando Montes-Negret 77 real rates paid on deposits. Table 7.3 summarizes some of the structural weaknesses that persisted during the early stages of transition. The Deterioration of Bank Portfolios While on a stock basis many of these banks appeared liquid, on a flow basis they were not. Excess liquidity on a stock basis was derived from the predominance of short-term assets (securities, loans) on their balance sheets, but eamings from these assets were often low (securities) or negative (fund- ing the cost of large portions of nonperforming loan portfolios), thereby reducing cash flow and constraining liquidity for ongoing operations. These financial weaknesses were symptomatic of larger operational flaws related to governance, management, and incentive structures typically found in state banks in CEE and FSU countries. This manifested itself in poor lend- ing decisions based on noncommercial criteria, which ultimately led to portfolio declines as economies collapsed and new, increasingly market- based, conditions were introduced. Against this backdrop, a second wave of bad loans materialized after the first wave inherited from the central planning era was recognized.7 This occurred for several reasons, namely, weak legislation and regula- tions, the absence of institutional capacity, and significant political pres- sure to loosen monetary and fiscal policy to restore production and re- duce spiraling unemployment. Neither banks nor enterprises were able to manage resources properly under such risky conditions, particularly as the environment was changing so quickly. The effect of hard budget constraints was to limit the quantity of funds available to banks to be able to grow out of their financial problems. Meanwhile, many of the loans made were poorly selected, further weakening the quality of their portfolios. These developments point to the importance of macroeconomic stability and the sequencing of reforms as transition countries were in- troducing an enabling environment for financial and private sector de- velopment. Also of considerable importance was the role of governance and management at the firm (bank) level, and the continued distortion of incentives despite nominal changes in laws and regulations. 7. While tthe inherited bad loans were recognized, the magnitude of the problem was not fully recognized. This was due to the lack of proper accounting and to delays in institutionalizing prudential regulations that properly classified loans and required banks to provision adequately for losses. These reforms occurred later, generally after 1991-92 in the most advanced CEE countries. 78 Restructuring Distressed Banks in Transition Economies: Central Europe and Uk-raine Table 7.3. Structural Factors Responsiblefor Poor Loan Quality Area influence Bank weaknesses Enterprise weaknesses Political * Central planning and control * Planning and control substi- preempted incentives for tuted for market in deter- active governance and mining economic needs, resource management distorted incentives Legal * Inadequate legal framework * Weak bankruptcy laws for loan recovery that favor debtors, provide * Weak court systems with no legal recourse to credi- untrained staff tors for run-up of arrears * Incomplete property registries * Property ownership rights and contractual obligations not clear * Weak collateral laws reduce lending Regulatory * Limited or no bank supervision * Weak enforcement of anti- * No risk management capacity monopoly provisions and protectionism, which limits competition * Unreliable accounting or no disclosure Financial * Poor or no credit risk assess- * Uncompetitive productiv- ment ity levels • Lack of portfolio diversifica- * Rising labor costs and over- tion staffing * Inadequate security or col- * Rising hard currency input lateral costs (energy) * Poor accounting standards * Outmoded technologies L Limited or nonexistent asset- * Poorly received products liability management * Thin consumer markets & Inadequate cash manage- ment e Weak strategic planning capabilities Operational * Inappropriate governance * Conflict of interest in terms structures and ownership of enterprise ownership, patterns management and gover- * Inexperienced bank manage- nance of banks ment * Inexperienced or inappro- * Lack of risk analysis and priate management (in a controls market context) * Weak incentives for better risk management Michael Borish and Fernando Montes-Negret 79 By 1991 estimates indicated that nonperforming loans amounted to up to 25 percent, possibly more, of GDP in CEE countries. Lower estimates reflected inaccurate accounting, unsuitable regulations con- cerning provisioning for losses and writing off of loans, and so on. Table 7.4 shows estimates of bad loan values for some of the largest banks. (In many cases, information was not provided to the public because of the size of the problem.) Many of these loans represented new lending flows in the early 1990s to large industrial companies that were overstaffed, uncompetitive, and unlikely to emerge as sound credit risks even with the introduction of some operational changes. As these new flows were in addition to what had been inherited from the earlier period (CEE) or had been eliminated because of hyperin- flation (FSU countries), it was evident by 1993-94 that initial reforms were insufficient to improve the quality of loan portfolios. Early Banking Sector Reforms and Institutional Responses A first round of reforms in transition countries involved a flurry of new legislation to break up monobank systems, reintroduce the concept of private property, and encourage the privatization of state assets. Nev- ertheless, as already noted, economic reality was that of spiraling de- cline. This put pressure on newly elected officials to respond quickly to the needs of those displaced by the old system before new laws and regulations were effectively implemented and market-based institutions were fully functioning. This resulted in continued state ownership of major banks (CEE countries), bank privatization without operational restructuring in the Ukraine, and direct (state) or indirect (national prop- erty funds, state pension funds) state ownership of major companies in all four coun!tries. In the three CEE countries examined, state banks rou- tinely accounted for about 60 percent of total assets and total loans, and about 70 percent or more of total deposits at the end of 1995. Ongoing state ownership practically ensured that the implementation of reforms would be slowed. In fairness to Poland, the slowdown in re- form was partly a reaction to problems associated with the rapid open- ing of the market to competition (low capital requirements, liberal licens- ing standards) prior to developing sufficient capacity for oversight. Likewise, in fairness to Hungary, all but one bank has been privatized since 1995 in a market that today is dominated by prime-rated institu- tions. By contrast, in the Czech Republic, interlocking directorates and structural weaknesses at the firm and bank levels have undermined overall Table 7.4. Estimates of Nonperforming Loanis in Transition Countries, 1991 and 1992 Czech Republic Hungarti Poland Ukraine Category 1991 1992 1991 1992 1991 1992 2991 1992 NPLs as a percentage of total bank loans 29 15-20 9 15-28 >33 25-60 - - NPLs as a percentage of c<, total bank assets 15 10 7 8 >13 13-30 NPLs as a percentage of GDP 21 14 6 9 >7 7-20 - - - Not available. NPLs Nonperformning loans. Note: Banks included in 1991 are KB, VUB, Konsolidacni, Investicni, and CSOB (Czech-Slovak Federal Republic); Budapest Bank, OTP, MlIB, MKB, K&H, and Postabank, based on total classified portfolio at end 1992 (Hungary); and seven SOCBs not privatized by the end of 1993 (Poland). Figures for 1992 represent figures and estimates of national sources and the International Monetary Fund for all banks. Figures are not available for Ukraine. Source: Dittus (1994); World Bank data. Michael Borish and Fernando Montes-Negret 81 competitiveness and raise serious questions about governance, manage- ment, and the depth of initial reforms. The Czech Republic is now in the process of dealing with many of these problems. In Ukraine, while "own- ership" changed, govemance and management practices did not (until very recently), and top managers of the state enterprises that had been majority shareholders of banks prior to share redistribution still made most major policy and personnel decisions.8 The following paragraphs discuss major reforms. Movement to Market-Oriented Systems One of the first banking sector reforms the transition countries implemented was to break up the monobank system into two tiers: a central bank re- sponsible for the conduct of monetary affairs and state owned commercial and specialized banks responsible for mobilizing deposits, lending, and carrying out other commercial banking activities. Hungary did this in 1987, followed by Poland in 1989 and Czechoslovakia in 1990. Ukraine moved in this direction as part of the FSU, and then did so as an independent coun- try in 1991. In general, the break-up of the monobank system was accom- panied by subsequent legislation that opened up the banking market to competitio:n. Key features included the following: * Clarifying the central bank's role and responsibilities in connection with monetary and banking matters. These responsibilities often involved determining monetary policy, safeguarding currency stability, man- aging clearinghouse and payments systems, licensing and super- vising commercial banks, and approving mergers and acquisitions. * Setting up the legal basisfor the establishment and existence of commercial banks. More often than not, this led to the initial creation of (a) larger, specialized banks with nationwide and global coverage focused on savings, housing finance, agriculture, and international trade; and (b) smaller, diversified commercial banks with limited geographic markets. * Establishing a regulatory framework for commercial banking. This in- cluded basic elements of legal and prudential regulatory systems and new accounting frameworks. 8. "Private" ownership of banks was established by reclassifying some of the banks' liabilities (for example, savings, deposits) as equity, triggered by a government stipulation in 1993 that the Ministry of Finance should control state enterprise equity. Meanwhile, as of the end of 1994, two large banks-Oschadny Bank (the savings bank) and Exim Bank (foreign trade)-remained state owned and continued to be tools of government policy, while also "commercializing" their operations. 82 Restructuring Distressed Banks in Transition Economies; Central Europe and Ukraine The process of breaking up the monobanks also created confusion with regard to the role of the second tier-commercial and specialized banks-largely because these reforms were not automatically or suffi- ciently accompanied by changes in incentive structures (legal and regu- latory frameworks, institutional development) or the financial discipline necessary for effective transformation to a market-based system. In some countries, the management of second-tier banks knew that the break- up of the monobank system was going to require them to operate ac- cording to commercial criteria. However, time and expertise were needed for the legal, regulatory, and institutional changes to take hold for a smooth transition. These developments and weaknesses prompted dif- ferent responses in terms of how governments, banks, and enterprises adapted themselves to the new commercial system, and how banking systems have evolved since the termination of the monobank system. Key developments were as follows: * Specifying the types and scope of activities permitted. Most CEE coun- tries have moved toward universal banking structures common to continental European Union (EU) systems. Directives from the EU have served as a basis for determining allowable activities in which banks and financial institutions may engage and the modalities of competition, supervision, diversification, and accounting. Ukraine has provided clarification of commercial banking and other finan- cial sector activities, although these efforts have not been shaped by EU directives. * Defining methods of regulatory enforcement. Laws generally specified the need for regulatory enforcement. Responsibility for enforcement has differed across countries. The central bank is prominent in most cases, although Hungary has differed in this respect. a Establishing basic regulatory guidelines. Transition countries, includ- ing CEE countries and Ukraine, have spelled out minimum capital and capital adequacy requirements, loan classification systems, lim- its on exposures (for instance, large, total large, insider lending, for- eign exchange), and other fundamentals of banking regulation, al- though actual enforcement remains uneven. * Establishing deposit insurancefunds. More recently, CEE countries have moved to establish explicit deposit insurance funds. While the privatization of state banks has been slow in CEE countries with the exception of Hungary since late 1995, the number of private banks has grown rapidly in both Central European and FSU countries. In the case Michael Borish and Fernando Montes-Negret 83 of the fo;rmer, this has largely involved the attraction of prime-rated for- eign banks, particularly in the Czech Republic and Hungary, and more recently in Poland after an initial surge through 1992. Smaller private do- mestic banks have also formed to meet local financing needs, particularly in Poland. In the case of the FSU countries, including Ukraine, privatization has largely involved "ownership transformation" without sufficient restruc- turing or foreign entry to be competitive. These countries adopted liberal licensing policies (low minimum capi- tal, flexible business plan requirements and management standards and skills) at thl'e beginning of banking reform. Moreover, they initially granted the same set of licensing requirements to foreign and domestic applicants. As a result, many new private banks were established in the first years of transition. Subsequently, the authorities replaced the initial liberal licens- ing policies with more restrictive policies, primarily in the form of increased minimum capital requirements and the introduction of differential require- ments for foreign applications. Rightly or wrongly, changes were motivated by concerns about (a) the increasing financial problems state banks experi- enced because of their exposure to deteriorating state enterprises; (b) the difficulties associated with bank supervision given the countries' limited supervisory capacity and a large number of small private banks; and (c) the belief (based partly on the problems with bank supervision) that bank fraud and failure were more likely to occur if the number of banks was too high, except in Ukraine, which until recently did little to rein in "priva- tized" banks, which are difficult to regulate. In CEE these licensing changes led to a slowdown in the number of new domestic private banks, an abey- ance of foreign investment in the sector (Poland), and some geographic and functional diversification of domestic banking activities. In some cases, the dominance of state banks has restricted the scope of expansion for private banks. For example, the size of private banks, measured by their shares in total banking sector resources on a stock basis, is small. At the end of 1995, private banks' average assets were US$420 million to US$470 million in the three CEE countries. Private banks accounted for nearly half of banking assets in Hungary at the time (more now), but only 30 to 40 percent in the Czech Republic and Poland. In otlher cases, the protection of state banks through uneven deposit insurance schemes (true today in Ukraine), protection via im- plicit guarantees (Czech Republic, Poland), and recapitalization (CEE countries) put private banks in a disadvantageous position. Average SOCB deposits were US$0.6 billion to US$3.9 billion in 1995, compared to an average of only US$135 million to US$190 million for private banks. Table 7.5. "Large" Banks' Assets, Capital, and Returnis, 1994 Tier one Capital Risk- Pre-tax capital Total assets as a adjusted profits Retuirn Return Couintry anid bankinig (US$ (US$ percentage capital (US$ on capital on assets subsection rmillionts) millions) of assets (percent) millionls) (percenit) (percent) Czech Republic 3,439 44,020 7.8 10.9 433 15.5 0.99 of which SOCB 3,175 38,826 8.2 11.2 420 16.2 1.07 of which private 264 5,194 5.7 8.3 13 3.9 0.40 Q Hungary 970 22,307 4.8 12.8 273 32.5 1.25 of which SOCB 523 16,013 3.9 11.9 131 29.1 0.85 of which private 447 6,294 7.1 15.1 142 36.3 2.25 Polanda 1,931 37,505 5.1 15.2 787 53.5 1.93 of which SOCB 1,386 30,371 4.8 13.2 355 25.6 1.21 of which private 545 7,134 7.6 17.0 432 93.9 6.04 Note: "Large" here means tier one capital of at least US$12 iLmillion. Figures for Ukraine were not available. a. Figures for Poland do not include BGZ, which would significantly reduce SOCB ratios. Private banks' profit indicators are probably overstated because of insufficient provisioning for loan losses, which is being corrected with increasing use of on-site inspections. Source: Annual reports; The Banker, September 1995; Bank (Poland), 1995 bank survey issue; World Bank data. Michael Borish and Fernando Montes-Negret 85 The three CEE governments also provided about US$15 billion to re- capitalize SOCBs, equivalent to 5 percent of same year GDP.9 Despite this protection, private banks have grown steadily in CEE mar- kets in the last five years, capturing many of the blue chip lending accounts and provi.ding fee-based advisory services. In addition to not having the burden of inherited nonperforming loans from the past, these banks are often more profitable, because they are better able to provide services in trade finance, corporate bond issues, custodial and trustee services, advi- sory services regarding international markets, and mergers and acquisi- tions. Only7 Ukraine has failed to accomplish this. Most CEE banks are stronger today than they were at the beginning of the transition. Capital ratios more accurately reflect solvency on a book- valued basis, as new accounting standards, stricter prudential regulations, and strengthened bank supervision have pressured banks to treat problem loans with more discipline.'" Whether enough discipline has been exer- cised in certain countries is now a topic of serious concern in the Czech Republic. However, as a general rule, these reforms contrast with the be- ginning of l:he transition, when state banks were not even aware of the existence and magnitude of problem loans. Table 7.5 highlights key capital and return nmeasures in 1994. Since then, Polish banks have shown improve- ment, particularly since late 1995, having benefited from earlier bank port- folio and operational restructuring, and bolstered by 6 to 7 percent real economic growth. Hungary has similarly shown improving performance, having benelited from privatization and additional foreign investment. 9. Total ncminal GDP for corresponding years in which recapitalizations occurred was US$315 billion (including the Slovak Republic as part of the Czech-Slovak Federal Republic figures for 1991 and 1993). Thus recapitalizations of about US$16 billion ap- proximate 5 percent of GDP. Interest charges have added to the cost, estimated to be comparatively low in Poland (0.6 percent), but up to 2.6 percent in the Czech-Slovak Federal Republic and 2.0 percent in Hungary at the end of 1993. Hungary's additional recapitalization in 1994 would add to this cost, as well as adding to intermediation costs for borrowers (see Dittus 1994; OECD 1993). 10. While capital adequacy ratios were only 6.3 percent at the end of 1994, risk- adjusted capital was 12.6 percent. In 1995 capital adequacy ratios were about 9.6 percent at year end, with private banks showing far higher ratios than SOCBs. Pri- vate banks, in particular, show a growing proportion of assets and capital and rising income figures and return ratios. However, private banks' returns may have been overstated in some countries. For instance, in Poland private domestic banks re- ported inordinately high returns on assets, which might reflect the overstatement of income and distort other ratios. Few of these banks were audited by international accounting firms in 1994. 86 Restrictuiring Distressed Banks in Transition Economes: Central Europe and Ukraine Deposit Ifnsuranice Trends Explicit deposit insurance is becoming increasingly prevalent in transition economies, particularly in those CEE countries that aspire to join the EU (Czech Republic, Hungary, Poland). As the EU standard for explicit insur- ance approximates per capita incomes (up to ECU 20,000), CEE countries have introduced measures along comparable lines: explicit insurance lim- its in Hungary are higher than per capita incomes, comparable in Poland, and lower in the Czech Republic. Perhaps more important during the tran- sition, the Czech Republic was able to sustain a high level of confidence among depositors in the banking system. This had more to do with a rea- sonably stable macroeconomic framework than with anything structural. The Czechs accomplished this despite depositors receiving negative real rates of interest-that were as high as -11.3 percent in 1993, stabilizing at -1.9 percent from 1995 through September 1996. Hungary and Poland pro- vided implicit insurance to depositors, although confidence appeared to be a bit lower there when looking at comparative intermediation rates (see table 7.2). This may have been more a function of firm- and household- specific liquidity needs, and less of an issue of actual banking capacity. Hungary and Poland, like the Czech Republic, paid negative real rates on deposits, although Hungary's real rates were less negative than the other countries' rates, and were positive in 1994. In the FSU, ruble-denominated savings lost value, and confidence will be more difficult to restore. Ukraine is still struggling to build a viable deposit base for its banking system. Banking Supervision and Accountting Standards Bank supervisory frameworks have improved in recent years, but they continue to experience weaknesses in trained personnel, information sys- tems, and overall risk management. Other key issues relate to the indepen- dence of banking supervision, which is often housed in the central bank, but frequently requires permission from the ministry of finance to enforce regulations. Nevertheless, capacity in Poland is clearly significantly greater than that which existed in the early 1990s, with improved information and analysis and a growing corps of trained personnel for off-site surveillance and on-site inspection. This has been reinforced by movement toward in- ternational accounting standards. All the CEE countries have introduced new accounting frameworks essential for regulation, governance, manage- ment, and overall development of a market economy. These are broadly consistent with international standards, and their introduction represents an institutional success, as the accounting and auditing professions did Michael Borish and Fernando Montes-Negret 87 not exist in the same manner just a few years ago. Domestic and interna- tional accounting firms have developed, supplemented, and reinforced local resources. Problems still remain regarding accuracy of information, con- solidation of statements, and levels of disclosure. However, improvements have also clearly been made. Key developments include the following: * Differing supervisory approaches have been introduced. Poland has shown the most comprehensive improvement, including the use of en-site inspections, detailed off-site surveillance and early warn- ing systems, and strict monitoring and reporting standards. The other three countries have less capacity and a far greater distance to go in enforcing Basle capital-based standards, let alone meeting standards assumed by the more pioneering risk-based approaches many of the more advanced economies are implementing. * Accounting standards have been introduced in CEE countries to provide increasing information and disclosure. These new standards are still be- ing developed, as are domestic accounting and auditing professions. Nevertheless, CEE countries have structured their frameworks to be ccnsistent with international standards. Often these have been based on specific legislation. However, institutional capacity is still comparatively weak in many cases, both in the origination of the inforrnation provided as well as in the analytical use of this infor- mation for regulatory, managerial, and investment purposes. Above all, crisis prevention, contingency planning, and crisis management need strengthening, and more complex risk management techniques will also need to be devised over time. This is likely to require con- tinued and significant levels of technical assistance. In Ukraine progress in introducing accounting reforms has been slow. Lack of an acceptable chart of accounts-a hurdle CEE countries overcame several years ago-remains a major drawback in the production of meaningful financial statements in many FSU countries. Governance and Managemient One of the more critical banking sector reforms has been the change in gov- ernance and management. Performance in this domain has varied. Poland has done well with the transformation of governance and management stan- dards at many of the state commercial banks in preparation for privatization; however, continued political and financial support for weaker specialized banks (for instance, BGZ in Poland) and resistance to accelerated privatization of "strategic" and politically powerful enterprise sectors has undermined 88 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine governance in these areas, even if management in some cases has improved. In Hungary, performance has improved significantly over the years, and has been given greater impetus since 1995 with the decision to privatize banks and impose hard budget constraints on loss-making enterprises. The Czech Republic's governance and management record is spotty, partly because of the linked ownership structure of the state (via the National Property Fund), banks, enterprises, and investrnent funds, as well as problems of informa- tion disclosure and accounting standards. Many of the recently proposed reforms to reinvigorate the economy reflect the economy's structural weak- ness, much of which emanates from lax governance requirements and man- agement standards. In Ukraine, governance still remains weak, and in some ways continues to resemble practices during the period of central planning. Key trends in the banking sector have included the following: Differing ownership patterns and governance practices. In the Czech Republic, the role of interlocking directorates and the prominence of banks-particularly given the state's continuing ownership role-raises questions about the allocation of resources and the degree of restructuring in the enterprise sector. Notwithstanding the structural change in the Czech Republic's labor market in the early 1990s (approximately 500,000 woman, usually mothers with children, or some 10 percent of the labor market, left the labor force once the authorities offered incentives for early retirement) and the fact that most heavy industry (where labor shedding was most needed) was located in the modern-day Slovak Republic, official unemployment rates of 3 to 4 percent were for a long time an aberration in the region. The government announced a series of proposed reforms in early 1997 intended to reduce the 8.6 per- cent current account deficit, which partly reflected weak com- petitiveness. In Poland, the Ministry of Finance introduced strengthened governance practices in 1993, pointing to the im- portance of privatization as a strategic objective. While privatization has not proceeded rapidly in Poland, state commer- cial banks (as opposed to some of the specialized banks) are con- sidered eminently more privatizable today than they were in 1993- 94. In Hungary, bank management was already strong, and the prominence and profitability of prime-rated banks established high governance standards. The shortcomings of state bank and state enterprise recapitalization in 1993-94 has given way to the most aggressive bank privatization program in the four countries. In fact, Hungary's privatization in banking is more complete than Michael Borish and Fernando Montes-Negret 89 in many EU countries. In Ukraine, governance is weak irrespec- tive of ownership, although current signs indicate that regula- tors are likely to be more active to contain systemic risks and to support macroeconomic improvements. Growing recognition of the needfor transparency and accountability. This is driven by bankers' needs to determine creditworthiness, regulators' needs to be able to rely on effective supervision, and capital markets' needs to make investment decisions. Increasingly, government policymakers similarly need better information to determine the effi- cienicy of the use of monetary and fiscal resources for budgetary and planning purposes, as well as for potential lender of last resort scenarios and deposit safety in the banking system. The last point has been criti- cal in tightening regulatory oversight and liquidating banks in the Czech Republic and Ukraine, accelerating privatization in Hungary, and pur- suing regular on-site inspections in Poland. Bank Resitructuring and Approaches to Sectoral Reform Initial refoim efforts have sought to achieve long-term growth objectives while mitigating some of the social costs associated with transition, ad- justment, and stabilization in the short term. In this context, the design of reform programs emphasized restructuring and privatizing banks and state enterprises. In practice, macroeconomic stabilization measures (hard budget constraints, fiscal balance, restrictive monetary policy) induced significant financial restructuring at the bank and enterprise levels in all four countries, with some operational restructuring resulting from less direct access to public resources. Most countries have at least partly priva- tized or liqu:idated most of their SOEs, often small and medium size en- terprises that lacked strategic value of any sort. Nevertheless, privatization in the bankinig system has generally lagged behind privatization in the enterprise sector, slowing progress toward a market economy. Hungary is the one notlable exception, because of its consistently sustained open- ness to private and foreign investment and its commitment since mid- 1995 to privatize the state banking sector fully. OriginaEly, reform programs anticipated that with appropriate assistance, state banks would play a leading role in enterprise restructuring efforts. Pro- gram designers expected banks to be able to conduct the needed analysis of their troubledborrowers, determine which enterprises should continue as going concerns and borrowers of the banks, provide financing and improved gover- nance for surviving enterprises, and initiate liquidation procedures against 90 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine those that were unlikely to be creditworthy in the future. l They perceived this role to be appropriate for banks because of their familiarity with individual enterprises. Policymakers also believed that bank-led restructuring could be more efficient in resolving debt disputes than centralized, nonbank (govern- ment agency) channels. However, because of their traditionally passive and noncommercial role, state banks lacked the institutional capacity and experience to restructure enterprises financially, physically, and operationally and to provide effective governance. 2 Ukrainian (and FSU in general) banks and enterprises were financially restructured by hyperinflation, but their operations and produc- tivity levels lag behind those found in CEE countries. Meanwhile, CEE coun- tries have taken a more gradual approach, restructuring their bad loans fi- nancially through recapitalization and operationally through banks' collection efforts, which have prompted some efficiencies, reorganizations, and changes in enterprise ownership, governance, and management. In some cases, such as the Czech Republic, governments used the "carve- out" approach to clean up bank balance sheets rapidly. In other cases, Po- land, for example, banks retained responsibility for recovery, even after recapitalization (see Montes-Negret and Papi 1997). In yet other cases, such as Hungary, recapitalization was intended to help banks and enterprises adapt to changing market conditions and grow out of their problems in a more disciplined and competitive way without inducing major social dis- location. However, in all cases restructuring needs persisted in banks (and enterprises) well after recapitalization. A sustained political commitment to and proper sequencing of reforms; a continuation of progress in creating an enabling environment; the in- troduction of suitable incentives for a wide range of stakeholders; the 11. Bank-led restructuring of the enterprise sector originally anticipated banks as- suming a comprehensive role. This included (a) conducting an analysis of problem debtors to determine the level of debt owed and how to have principal repaid and interest serviced; (b) restructuring the debt of potentially viable enterprises; (c) financ- ing the physical restructuring of potentially viable enterprises; (d) exercising corpo- rate governance over these enterprises; and (e) writing off the debts of, curtailing new credit to, and in some cases liquidating nonviable enterprises. Thus the decentralized bank-led approach assumed that banks would take the lead in the financial (debt), physical (property, plant, equipment, inventories), and operational (governance) re- structuring of viable enterprises, and accelerate the liquidation of nonviable enter- prises. This proved to be an overly ambitious conceptualization of banks' roles and capacity in general enterprise restructuring in transition countries. 12. This should not be surprising, as even in the most affluent economies, com- mercial banks are often poorly equipped to lead enterprise restructurings. Invest- ment banks and specialized consulting firms are often the turnaround experts in these markets. Michael Borish and Fernando Montes-Negrel 91 importance of "strategic" investment from, training by, and correspon- dent linkages with prime-rated institutions from market economies; and the proninent role of governance and management have emerged as criti- cal ingredients for successful reform. The following sections discuss spe- cific country experiences. Box 7.2 highlights the depth of illiquidity and insolvency in each country, which set the stage for restructuring.13 Czech (and Slovak) Republic: The Problem of Partial Privatization and Weak Governance The Czecla Republic has pursued a dual approach to privatization and private sector dev,elopment: voucher distribution for the vast majority of citizens who had previously been unable to accumulate much in the way of savings be- cause of the level of control and centralization that prevailed during the com- munist period, combined with efforts to attract investment from and joint ven- tures with Western companies. These efforts were backed by a commitment to macroeconomic stability to keep inflation and interest rates comparatively low, the currency stable (and strong), and fiscal accounts largely in balance. Until recently, results appeared to be positive at the macroeconomic level, although significant structural weaknesses have long been evident and culminated in deteriorating 1996 current account figures. In the banking sector, the Czech (and Slovak Federal) Republic opted to pursue an up-front carve-out of bad debts in the major banks as an integral part of the bank privatization program under the country's general mass privatization program. At the time, nonperforming loans were estimated to approximate 36 percent of total credit to nongovemment entities and 25 per- cent of GDP. As with challenges faced in the enterprise sector, the govern- ment believed that swift and mass privatization combined with a stable mac- roeconomic environmentwould be sufficient to work out structural problems. A carve-out of bad assets was intended to make the banks more amenable to privatization on a stock basis (more attractive value based on a clean bal- ance sheet)-as well as on a flow basis-by not burdening management with the overhang of bad debt from the central planning era, and giving banks the opportunity to become more profitable under new incentive structures. 13. This chapter refers to liquiditv and illiquidity on a cash or funds flow basis, not on the basis of current assets and liabilities on the balance sheet. Hence, liquidity problems or illiquidity reflect limited earnings flows to banks available for new lend- ing, investments, and operations. This may occur despite bank balance sheets holding substantial net current asset positions. 92 Restntcturing Distressed Banks in Transition Economies: Central Europe and Ukraine Box 7.2 Bank Portfolio Restructuring: Causes and Approaches Magnitude of the Problem In the Czech Republic, nonperforrning loans accounted for about 20 to 30 percent of total banking system loans. Hungary had a lower amount of nonperforming loans, but the amount increased quickly with the passage of new legislation in 1991-92, which led banks to recognize more of their bad loans.' Poland's bad loans were as much as 60 percent of total bank loans. In Ukraine, nonperforming loans are high and remain a problem today because of the absence of serious structural reform. The Problem of the State Sector Bad loans were highly concentrated in the state banking sector, which held 85 to 100 percent of nonperforming loans. Also, at least in the early stages, state enterprises repre- sented the bulk of nonperforming borrowers. This problem was particularly severe in Poland, where more than 60 percent of the seven SOCBs' loans were nonperforming, and where the loan portfolios of some of the larger specialized banks were in even worse shape. In the Czech Republic, where 30 percent or more of state owned banks' portfolios were nonperforming, individual state banks often carried much higher proportions of such loans than the average. In Hungary, state banks' loans were 64 percent nonperforming once new laws were adopted in 1991-92. In Ukraine, problem loans were in both state and "private" banks, the latter being generally controlled by SOE managers. Impact on Solvency and Liquidity The concentration of bad debts among a number of state owned banks meant that these banks faced severe problems of solvency and had low capital adequacy ratios. At the end of 1990, VUB (Czech-Slovak Federal Republic) reported a capital adequacy ratio of less than 2 percent. At the end of September 1992, 7 of the 14 largest Hungarian banks had negative capital adequacy ratios and 5 others had capital adequacy ratios that were below the 7.25 percent targeted by the end of 1992. These low capital adequacv ratios were more problematic (more so than would have been reported according to Basle standards) in that more rigid loan classification standards observed by market econo- mies were not yet in place. The introduction of these standards led to widespread provi- sioning and write-offs in CEE countries, prompting the need for recapitalizations to counter solvency problems (negative capital). Ukraine is now in the process of imple- menting comparable regulations. Only 14 banks currently meet relatively low minimum capital requirements, and most have iow or negative capital. Approaches CEE governments faced two main alternatives in dealing with the bad debt problem: (a) they could pursue a decentralized approach-as in Poland with commercial banks (but not specialized banks) and Hungary with commercial banks-whereby individual banks lead the financial restructuring process (or more modestly retain responsibility for loan recovery), with assistance from the state; or (b) they could follow a centralized approach- as in the Czech Republic, and to some extent in Hungary-by carving out bad debts and recapitalizing thebanks. (For an overview of the major options see Saunders and Sommariva 1993. Also see Borish, Long, and Noel 1995 for a review of country-specific restructuring programs in the banking and enterprise sectors.) Other options included debt cancellation or forgiveness; liquidation of insolvent state banks, with government assumption of banks' liabilities; and hyperinflationary elimnination of banks' liabilities. In Ukraine, hyperinfla- tion and liquidation have been used as tools, but serious restructuring still needs to occur. 1. Key factors were (a) the Act on Financial Institutions of 1991, which created three categories of nonperfoTning loans (substandard, doubtful, bad) and mandated provisioning for these loans (20, 50, 100 percent provisioning, respectively); (b) the Accounting Act of 1992, which allowed banks to create provisions out of pretax profits; and (c) the Bankruptcy Act of September 1991, which imposed bank- ruptcy proceedings on any firm that was in arrears by more than 90 days. Michael Borish and Fernando Montes-Negret 93 Such a carve-out was thus considered an effective tool in accelerating the privatization of state banks. Within this context, the Czech-Slovak Republic established the Konsolidacni Banka in March 1991 as a "loan hospital" or "bad bank" to clean up the balance sheets of the commercial banks and to work out nonperforming loans without burdening the other banks with the clean- up effort. The carve-out, recapitalization, and partial privatization were carried out in four steps as follows: * Carve-out. The government carved out K 110 billion (US$4 billion) in substandard and nonperforming assets from three of its major banks (in exchange for matching liabilities in the redistribution of credit and deposits of the savings banks), and placed them on the balance sheet of the newly created Konsolidacni Banka (which assumed more than 6,000 nonperforming loans granted prior to 1989 from the bal- ances of the banks participating in the carve-out). The K 110 billion taken over came from a total of K 180 billion (US$6 billion) of perma- nently revolving loans, the so-called TOZ credits, which were origi- nally given to enterprises at 6 percent interest and without a speci- fied maturity date. At the beginning of 1991 the authorities abolished all TOZ credits, and the credits taken over by Konsolidacni Banka were renegotiated at 13 percent interest (300 basis points above the discount rate) with an eight-year maturity. When the two republics separated in January 1993, Konsolidacna Banka Bratislava was cre- ated in the Slovak Republic to assume Sk 30 billion (US$1 billion) in nonperforming loans, while Kc 80 billion (US$2.8 billion) were allo- cated to the Czech Konsolidacni Banka. * New bondfinancing (first recapitalization). Also in 1991, the National Prop- erty Fimd issued K 50 billion (US$1.7 billion) in bonds. Of this bond issue, K 12 billion (US$400 million) was made available for capital (re- payable in currency), and K 38 billion (US$1.3 billion) was provided in bonds wvith a five-year maturity (repayable in the form of shares in priva- tized enterprises). Proceeds were transferred to the banks for the carve- out of bzad loans and to recapitalize the banks. Capital adequacy ratios for these banks consequently increased from 1.5 to 4.5 percent. * Inclusion in mass privatization program. The banks were immediately in- cluded in the mass privatization program. After the first wave of mass privatization, all the major banks of the system had been partially priva- tized, including Komercni Banka, Investicni Banka, Zivnostenska Banka, and Ceska Sporitelna. However, the largest shares of these banks re- mained with the state through the National Property Fund. 94 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine Assumption of additional nonperforming assets (second recapitalization). The carve-out and recapitalization were followed in 1993 with the assump- tion by the Czech and Slovak ministries of finance of K 95.5 billion (US$3.2 billion) of Obchodni Banka's (the Foreign Trade Bank's) as- sets and K 74.4 billion (US$2.5 billion) in liabilities in nonconvertible currencies from the central planning era. By the end of 1993 both gov- ernments had injected K 4.05 billion (US$135 million) into Obchodni Banka and transferred K 40 billion (US$1.3 billion) in bad or doubtful loans to separate collection units, thereby raising Obchodni Banka's capital adequacy ratio to 6.25 percent. Most observers believe that the carve-out helped the remaining banks clean up their balance sheets instead of bogging them down with problem loans that would have continued to be a burden for them. These measures were taken at the time to deal forthrightly with stocks of nonperforming loans during the Czech-Slovak Federal Republic period in advance of bank privatization. In mid-1997, Konsolidacni itself had about US$4 billion in assets, more than US$1 billion in reserves (reflecting the nature of its port- folio), and access to the Euromarkets. Meanwhile, since that time smaller banks have been merged and liquidated as problems in their portfolios have come to light. Nevertheless, given the still high levels of nonperforming loans cur- rently held in other banks' portfolios after the recapitalization, serious ques- tions remain about (a) the mixed ownership structures (banks, funds, en- terprises), interlocking directorates, and lack of transparency; (b) the impact these structural weaknesses have had on the governance and management of banks; and (c) the impact of these banking sector weaknesses on the enterprise sector, given the prominent role of banks and bank debt in the economy. As the Czech Republic had one of the highest levels of domestic credit provided by the banking sector to GDP in the world as recently as 1995 (and presumably in 1996), the quality of such lending is critically im- portant to the economy at large. 14 Problems remain with regard to ongoing high levels of nonperforming loans, as well as with SOCB governance, internal bank supervision, lev- els of information disclosure and accountability, and general risk that has manifested itself in the need for continuing high levels of provisions for loan losses. Some shortcomings are also apparent in the way that bank 14. Domestic credit provided by the banking sector was 93.4 percent of GDP in 1995, higher than in all other transition countries and higher than in most countries of the world (see World Bank 1997). Michael Borish and Fernando Montes-Negrel 95 privatization was handled, namely, the partial privatization and block- ing shares of banks that the National Property Fund has retained (through the fund, the government has kept 30 to 45 percent blocking shares in the four major banks). Corporate governance weaknesses remain as a result of cross-ownership between the state, investment privatization funds, banks, and large state enterprises. Investment privatization funds are among the largest shareholders in the banks, yet banks are among the largest shareholders in the biggest enterprises. Thus weaknesses with the approach appear to include inadequate governance and management in financially restructured banks caused by ownership structures, inadequate enforcement of prudential regulations or weaknesses in the framework, and lack of interest in some of the technical assistance that was previ- ously available to strengthen bank operations. Hungary: The Triumph of Structural Reform and Rapid Privatization After the acceleration of reforms in 1990, the Hungarian authorities became progressively aware of the weak financial situation of most Hungarian state banks. (As of the end of 1995, the government still held significant, and usu- ally majority, ownership in 10 banks, including the country's largest banks.) Above all, with the introduction of new banking and accounting standards combined with a strict bankruptcy law, it was apparent by 1993 that a sig- nificant portion of banks' loan portfolios were nonperforming-more than 40 percent of large state banks' loans were classified as such-and that the largest state banks were generally insolvent. Meanwhile, because of Hungary's long-standing commitment to an open environment that dated back to the late 1980s, foreign investment from prime-rated banks that main- tained high levels of competitiveness was significant, and in some cases si- phoned off the best companies in the Hungarian market. In 1993-94, when the fiscal deficit was beginning to balloon out of con- trol because of high social entitlements and a weakening economy, the gov- ernment opted to recapitalize its troubled state banks to restore solvency and make them more privatizable. It implemented a series of bank recapi- talization programs each year from 1991 through 1994, small in the early years, and more costly in 1993-94, when the magnitude of the problem was more fully recognized. The direct cost of Hungary's bank recapitaliza- tions was US$3.5 billion, with an additional estimated US$1.0 billion to US$1.5 billion in interest charges passed on to customers. These interest charges approximated 2 to 3 percent of year-to-year GDP from 1991 to 1994. The various recapitalizations took place as follows: 96 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine * 1991 consolidation agreement (US$0.1 billion). Government guar- antee for Ft 10.5 billion (US$100 million) for bad loans made be- fore 1987 during the monobank period, mainly to coal mines. This was accompanied by the drafting of new accounting rules and bankruptcy and banking laws. * 1992 loan consolidation scheme (US$1.1 billion). New accounting stan- dards led to a dramatic increase in recognition of nonperforming loans, prompting the 1992 loan consolidation scheme. Nonper- forming loans made to resident enterprises before September 1992 were swapped for 20-year, variable coupon government bonds at market rates. Swaps were made for 50 percent of loan values for loans made before 1992 and at 80 percent replacement for loans made during 1992. Fourteen banks and 60 savings cooperatives partici- pated in the scheme. In total, Ft 120.5 billion of bad debt was swapped for Ft 98.6 billion in government bonds. Bad debts were placed in the Credit Consolidation Fund managed by the Hungarian Invest- ment and Development Corporation, which got a license to operate in 1993. The debt swap involved 1,885 companies, of which 116 went bankrupt and 549 were liquidated. Three-quarters of the swapped bad debt involved 110 companies, of which about 100 went bank- rupt or were liquidated. * "13+1 " program (US$0.6 billion). This recapitalization involved swap- ping Ft 57 billion in bonds for 90 percent of the book value of bank loans to 13 large industrial enterprises and the state railway com- pany. The two state property funds were expected to turn these "stra- tegic" enterprises around based on a strategy of short-term loss con- tainment and financial restructuring. * 1993/1994 consolidation agreement (US$1.65 billion). In 1993, Parliament voted a bank recapitalization program to raise banks' capital adequacy ratios from -15 percent to zero by the end of 1993, and to 8 percent in 1994. Eight banks (and some savings cooperatives) received govern- ment bonds worth Ft 114.4 billion in 1993, of which MHB received about half (Ft 56 billion), K&H received about one-third (Ft 37 bil- lion), and Budapest Bank received Ft 6 billion. A second injection of Ft 50 billion in 1994 brought the total to Ft 165 billion and capital adequacy ratios to 4 percent. The Ministry of Finance ended up with 75 percent of voting shares until the end of 1995. In the first recapitalization effort, the 1991 consolidation agreement, the government issued a guarantee of 50 percent of Ft 10.5 billion (US$100 million) in doubtful loans that had been transferred to the banks in 1987. Michael Borish and Fernando Montes-Negret 97 This should have reduced the burden on the banks of bad loans to enter- prises, but it covered only 2 percent of the bad loans the banks were hold- ing at the end of 1990. The 1992 loan consolidation scheme was launched in 1992, but took effect in May 1993. The program involved the exchange of Ft 98.5 billion (US$1.1 billion at the time) in government bonds for loans wilkh a carrying value of Ft 120 billion (US$1.3 billion). The loans were carved out from the banks, with the goal of executing separate work- out arrangements. However, workout attempts achieved limited success, and the plan failed to remedy the problem of recurring credit quality de- terioration. In late 1993 the authorities introduced the "13+1" program, which involved the exchange of Ft 58 billion (US$620 million) in bonds for 90 percent of the book value of bank loans to 13 large industrial enter- prises and the railways. The Ministry of Industry and Trade selected these enterprises based on their strategic importance. The hope was to turn them around through loss reduction measures and financial reorganiza- tion without significant physical restructuring. The program achieved moderate success. The inadequacy of the first recapitalization schemes led the govern- ment to launch the 1993 bank consolidation program, intended to be a com- prehensive solution to bank problems. The government contracted Cr6dit Suisse First: Boston to evaluate the banks' portfolios. The evaluation re- vealed that (a) three of the five largest banks, all SOCBs, were technically insolvent; (b) a number of cooperatives were inadequately capitalized; (c) OTP's capital was deficient, a highly risky situation given its importance to depositors' confidence and the financing of the interbank market; and (d) ongoing SOCB operating losses would erase remaining capital in one year. Capital deficiency at the end of 1993 was estimated at Ft 139 billion (US$1.4 billion), the equivalent of 7 percent of nominal SOCB assets. This left little more than Ft 18 billion (US$179 million) in capital, or less than 1 percent of assets, showing that SOCBs had little net worth in the aggre- gate, even after four recapitalizations that had exceeded Ft 320 billion. Nonperforming loans were expected to continue to reduce SOCB capital, particularly at two of the largest banks (MHB, K&H), where estimates in- dicated that nonperforming loans were as high as 87 percent of total loans. Meanwhile, annual operating losses at 16 banks approximated Ft 20 bil- lion (about US$200 million), which would have eliminated SOCB capital in about one year if nonperforming loans and the Ft 139 billion capital deficiency were taken into account. Inmediate action to deal with solvency issues was warranted, given the risk of a liquidity crisis and high interest spreads. Estimates of the contribution of loan portfolio deterioration on the level of intermediation spreads suggest 98 Restructuring Distressed Banks in Transition Economies: Central Europe and U)kraine that banks raised interest rates to sound borrowers by nearly 400 basis points to offset the effect of nonperforming loans (see OECD 1993). This contributed to real margins of 7 to 8 percent through 1993, although these have come down steadily since 1994 as net margins tightened due to increased competition (real net margins were 6 percent in 1994, 5 percent in 1995, and 4 percent in 1996, and are very narrow today in the hotly contested blue chip sector) . This subse- quently resulted in reduced profits for SOCBs. Meanwhile, more competitive private and joint venture banks, with their stronger balance sheets, lower cost structures, and superior profitability, were able to offer better rates and be more precise in selecting creditworthy borrowers. The government recapitalized SOCBs to a zero percent capital adequacy ratio at the end of 1993. Two additional infusions followed in May and December 1994 to increase the capital adequacy ratios of the troubled banks to 8 percent. This 1993 consolidation program marked a shift in the government's strategy from the centralized approach more characteristic of the Czech Republic to a bank-led decentralization scheme more similar to that followed by Poland. To effect the capital infusion, the state issued 20-year bonds paying semi-annual interest at an annualized rate of 5 per- cent. The value of these bonds totaled Ft 165 billion (US$1.65 billion), about the same as all earlier recapitalizations combined (on a dollar basis). The banks were asked to evaluate their financial and operational posi- tions, prepare corrective action plans, develop special workout units, and improve key operational functions. However, these agreements were often ineffective, because they failed to (a) specify quantitative performance cri- teria and targets, leaving the extent of restructLring open to interpretation; (b) provide adequate incentives to owners, managers, and bank personnel for improved governance and performance; and (c) impose penalties for noncompliance. The lack of strong regulatory oversight resulting from frag- mented supervision aggravated the problem."5 While the first recapitalizations were limited in their cost, the final con- solidation plan was costly in its timing, coinciding with a period when the government was running up large fiscal deficits in 1993-94 to cover high social welfare costs. Perhaps more damaging was the notion that govern- ment efforts at least indirectly created a moral hazard by precipitating expec- tations of future assistance among large enterprise loss makers, particularly 15. The Mlinistry of Finance established the Bank Control Unit in Januarv 1994 to monitor compliance with the consolidation agreements, although its authority was unclear. This lack of clarity was rooted in the fragmentation created by the 1991 Bank- ing Act, which split responsibility for regulation, supervision, and enforcement be- tween the Bank Supervisory Committee and the state banking supervision agency Michael Borish and Fernando Montes-Negret 99 in strategic sectors with high employment levels. Such an approach under- mined bank management, as government intervention undercut attempts to enforce financial discipline. Even though the 1993 bank consolidation prograrn increased capital ad- equacy ratios to acceptable levels, true capital adequacy ratios were still uncer- tain because of the incomplete scope of audits. The SOCBs were still struggling to maintain adequate capital levels in 1995, which ultimately contributed to the decision to privatize the banks under the May 1995 Law on Privatization. The state banks accounted for the major share of losses and nonperforming loans. By September 1994 the stock of bad loans had reached Ft 597 billion (US$5.5 billion), or 26 percent of total loans (69 percent of the total stock of enterprise loans). The stock of bad loans not serviced for more than a year was Ft 268 billion (US$2.5 billion), or 12 percent of outstanding loans. 16 The proportion of troubled loans in the portfolios of the four large SOCBs increased from less than one-third at the end of 1993 to nearly 40 percent by September 1994. The progressive deterioration of their portfolios meant that the 1993 consolidation program may not have been sufficient in achieving an 8 percent capital ad- equacy ratio based on end 1994 balance sheets, as it was based on end 1993 data (see IMF 1995). Large state banks generated pretax profits in 1994, but these were thin--return on average assets was only 0.85 percent-and well below levels the SOCBs needed to achieve 8 percent capital adequacy ratios. In light of these weaknesses, the Ministry of Finance initiated policies to improve the financial positions of the large state owned banks by focusing on hard budget constraints and privatization. It initiated the process in 1995 with the partial privatization of several SOCBs, including the acquisition of Budapest Bank by General Electric Capital and EBRD for US$87 million, and the partial privatization of OTP, which raised more than US$140 mil- lion in new capital. MHB, which is now majority private, made strong ef- forts to reduce problem loans in advance of full privatization. 17 It succeeded 16. Annual operating losses approximated Ft 20 billion at 16 banks, most of which were SOCBs. Nonperforming loans were estimated to be (a) Ft 186.4 billion (US$2.2 bil- lion), or 11.5 percent of outstanding loans at the end of 1992; (b) Ft 229 billion (US$2.5 billion), or 13 percent of outstanding loans in mid-1993; and (c) Ft 352 billion (US$3.5 bil- lion) at the end of 1993, before recapitalization; (d) Ft 143.8 billion (US$1.4 billion), or 11.1 percent of outstanding loans at the end 1993, after recapitalization. Classified loans were mostly to large SOEs. 17. Between mid-1995 and March 1996, K&H resolved more than half of its troubled credit portfolio (a reduction in nominal value from Ft 45 billion to Ft 22 billion), while MHB launched a subsidiary ("Risk Kft") to restructure Ft 82 billion of its troubled assets (Ft 16.8 billion of book value, net of provisions). More than half of the assets transferred to Risk Kft were resolved of approximately two-thirds of the net value of the assets. 100 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine in attracting investment from ABN-Amro in 1996, partly because of these efforts. In the case of ABN-Amro's investment, another US$200 million to US$250 million or more are expected to be invested to strengthen the bank's position in the retail banking market. The strengths of Hungary's new approach include the restored commit- ment to macroeconomic stabilization, the sustained political commitment to an enabling macroeconomic enviromnent, the improved governance and management at banks and firms, the recognition of the need to be competi- tive as markets become increasingly linked, and the establishment of a time- table for full privatization. Notwithstanding some of the weaknesses of the recapitalization approach, Hungary is well on its way to being the most fully privatized and competitive banking sector in Central Europe. Poland: From Shock Therapy to Gradualism In Poland, the banking system's portfolio deteriorated substantially in 1991, at which time the government was confronted with the fiscal implications of the loss of some 40 percent of bank assets. The government wanted to find a solution that minimized the loss of jobs and productive capacity. After an initial period when licensing requirements encouraged the entry of private banks to stimulate competition (but for which Poland's legal and regulatory framework and supervisory capacity were unprepared), Poland's bank reform program stressed the gradual restructuring of nine regional SOCBs and several specialized banks prior to privatization. The government originally decided to delegate to the nine SOCBs the task of restructuring the enterprises that had failed to adapt to new market condi- tions, and to recapitalize the commercial banks for that purpose. After two banks (Bank Slaski and Bank Wielkopolski) were partly privatized with foreign investment, the number was reduced to seven commercial banks. The Law on Financial Restructuring of Enterprises and Banks became ef- fective in March 1993, and established the basis for a program not only to recapitalize the banks and restructure their balance sheets, but also to deal with state enterprises with bad debts. Key features of the plan were as follows: * Classifying loans. The authorities instructed the state banks to sepa- rate out loans classified by auditors as loss and doubtful, and to create provisions amounting to 100 percent for the loss category and 50 percent for the doubtful category on the basis of a December 1991 portfolio analysis conducted by international accounting firms. * Setting up workout u7itsfor loan restructuring. The SOCBs set up internal workout units to manage their bad loan portfolios and to restructure Michael Borish and Fernando Montes-Negret 101 these loans within about a year. Banks refrained from lending to bor- rowers with doubtful or unrecoverable loans unless they presented a clear and acceptable plan to the bank (and to the Ministry of Finance, the banks' owner) to enhance their creditworthiness and justify new lending. Each workout unit had 15 bank staff, although not all bad debts were assigned to these units. * Issuing bonds (recapitalization). The government recapitalized seven SOCBs with about US$600 million in 1993, approximating 12 per- cent capital adequacy after provisioning for loan losses and accrued interest. Poland also recapitalized three specialized banks, although the restructuring of these banks differed considerably from that of the SOCBs, in that it follows a more centralized approach. The cost of these recapitalizations thus reached about US$1.1 billion. * Establishing the Bank Privatization Fund. Poland established the Ba:nk Privatization Fund at the end of 1992 to service the bonds issued to the SOCBs in 1993. It established this fund to reduce prospective investors' fears of debt service complications that would interfere with bank privatization. Resources from the fund were to be transferred to the government to service the Treasury bords any SOCB held as a result of its recapitalization. As an in- centive to privatize as rapidly as feasible, funds are made avail- able only after a bank has been privatized. By the March 1994 deadline spelled out in the law, the commercial banks had financially restructured the 800 or so enterprises that accounted for most of their bad loan portfolios. About 200 of these enterprises, which accounted for more than half of the bad loans, entered into conciliatory agreements with their creditors under an out-of-court reorganization pro- cedure modeled on the U.S. bankruptcy code (Chapter 11). The agree- ments typically entailed (a) transforming state enterprises into joint stock companies; (b) rationalizing and reorienting troubled enterprises' activi- ties by reshaping product lines, closing unproductive units, and trim- ming the work force; (c) rescheduling financial obligations to make re- payment financially feasible; and (d) diluting state ownership through debt-equity swaps in about 150 cases. Other enterprises underwent liq- uidation, had their collateral executed, regained their creditworthiness, or had their debts auctioned off by banks. The recapitalization of the seven SOCBs enabled them to recognize their losses and work out their bad loans. The results are still hard to gauge, as many enterprises have shown improved performance that might not have occurred if the economy had not grown at 6 to 7 percent real 102 Restrictuiring Distressed Banks in Transition Economies: Central Europe and Llkraine rates. However, the process helped to "commercialize" the banks prior to privatization based on the adoption of strengthened corporate governance and management, a program to restructure loan portfolios, and incen- tives for employees through the issuance of shares. The enterprise and bank restructuring program has strengthened financial discipline and forced commercial banks to develop risk-assessment capacity. By most accounts, the commercial banks have succeeded in restructuring their loans and strengthening their financial condition. As of March 1994, the SOCBs had effectively dealt with 98 percent of their larger bad loans, and by some estimates, their average capital adequacy ratios stood at 27 percent, well above international standards. They signifi- cantly reduced their stocks of bad loans, from 30 percent of portfolios in 1993 to about 8 percent in 1995. About 83 percent of classified SOCB loans have been or are being restructured, while 17 percent have been declared unre- coverable. Regional SOCBs are currently extremely profitable, largely be- cause of the large net interest spreads they have enjoyed between govern- ment securities and deposits-these rates were in the 8 to 9 percent range until 1996-as well as from efficiency improvements. Since late 1995 lending has increased, while spreads have narrowed as competition has intensified. The strengths of the Polish approach include the sustained political com- mitment to improved bank governance and management; the introduc- tion of suitable incentives to stakeholders, that is, shares for bank employ- ees and retained proceeds by banks from recoveries after recapitalization; the beneficial institutional effects of internal workout units, which have contributed to better credit management and operations after the restruc- turing period (see Montes-Negret and Papi 1997); the extensive use of train- ing and technical assistance from internationally recognized Western banks; and the recognition that recapitalization should be up-front, one-time, fis- cally feasible, and focused on privatization. All this has been reinforced by a steadily improving macroeconomic context characterized by 6 to 7 per- cent real growth rates, declining inflation rates, and reduced fiscal deficits. However, questions remain about the extent of SOEs' operational and physical restructuring and their ability to compete as the market opens up. Some of the financial restructuring of bad loans to long-term status may have delayed loss recognition, because some enterprises have yet to undergo sufficient operational restructuring to demonstrate real competitiveness and positive cash flow in open market conditions. This contributes to a level of nonperforming loans in the range of 15 to 20 percent of loan portfolios. With regard to the bank privatization that has occurred, it has been par- tial, largely because of fears of the domestic banking sector being overrun by Mic1hael Borish and Fernando Montes-Negret 103 more competitive foreign institutions without ample time being provided to allow Polish banks to become competitive. Signs that this attitude is chang- ing are encouraging, as shown in the increasing stakes held by ING of the Netherlands and Allied Irish Bank in the two state banks that did not require recapitalization under the program, and the initial privatization of Poland's mostprofitable bank, Bank Handlowy, in mid-1997. However, the privatization of five SOCBs has been put on a consolidation track, which may be compli- cated by remaining questions about the (a) degree of expertise SOCBs have attained regarding internal and managerial controls, loan evaluation, risk man- agement, and overall service delivery, although here technical assistance in the form of twinning arrangements with West European banks has led to vast improvements; (b) basic bank governance, particularly for those banks that still hold troubled SOE loans in important industrial sectors; (c) general levels of SOCB competitiveness without protection from greater foreign en- try; and (d) impact on changing portfolios as net spreads on government securities decline, and as interest income from lending increases as a share of the income stream. Polish banks were partly shielded from foreign competi- tion during 1992-95 to give Polish banks time to recover, reduce bad debts, and increase capital in advance of more open competition. Such protection is not likely to recur. Another major challenge is the reform of the specialized banks-BGZ and PKO BP-which still accounted for more than a third of banking system assets in 1996. Thus weaknesses to the Polish approach in- clude continued weaknesses in the enabling environment (although these are steadily improving), slow privatization after several years of institutional strengthening, and still low levels of financial intermediation. Ukraine: The Perils of Delayed Economic Reform Since 1992 the number of "private" commercial banks in the Ukraine has proliferated; however, most of these banks have long been in serious finan- cial trouble, many are unable to meet minimum capital requirements, and some are sti]l managed according to noncommercial criteria. In general, banks' profits are weakened by the persistent accumulation of nonperform- ing loans to enterprises. Thus Ukraine did not pursue structural reforms as intently as its counterparts in CEE countries. The first commercial banks with private ownership were registered in 1989. In 1990 three of the five sectoral banks in Ukraine-Bank Ukraina (agriculture), Prominvest (heavy industry), and Ukrsotsbank (social invest- mentbank)-transformed themselves into joint stock companies. These three banks accounted for more than 80 percent of Ukraine's banking activity 104 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine from 1991-93. By the end of 1994, the banking sector consisted of the Na- tional Bank of Ukraine, five specialized banks, (two state owned and three majority owned by state enterprises), and about 350 locally owned com- mercial banks. The number of commercial banks had shrunk to around 220 by early 1995 because of widespread insolvencies, including some of the larger "private" banks, which failed because of foreign exchange losses and fraud. Among the remaining banks, nearly half (about 100) were reported to be in serious financial trouble in 1996 because of the large share of nonperforming loans in their portfolios. Only 14 banks were sufficiently strong to meet Ukraine's minimum capital requirements. Meanwhile, as of the end of 1994, two large banks-Oschadny Bank (the savings bank) and Exim Bank (foreign trade)-remained state owned, although under more commercialized incentive structures. Fundamental problems facing the Ukrainian banking sector have been both structural and macroeconomic in nature, although the country's macroeconomic circumstances have improved in the last few years and it has made some progress at the structural level in terms of building institutional capacity and skills. The current situation can be summarized as follows: • Ownership, management, and governance. Many of the commercial banks were formed to raise funds for specific enterprises. Thus their criteria for resource management were usually captive to the prerogatives of enterprise owners, and were often noncommercial. For these banks the introduction of hard budget constraints altered incentives and decisiomnaking, thereby helping to contain some of the problem af- ter significant damage had already been done to the economy. * Funding sources. Banks' liabilities consisted mainly of enterprise de- posits that have contractual maturities of three months or longer, with an option to withdraw funds and forfeit interest. In 1991-92 newly formed commercial banks had to respond to enterprises' demands for cash. This shortage of cash forced enterprises and state banks to write IOUs to their employees rather than pay wages in cash. Enterprise and household deposits have diminished since as local currency values depreciated, and as foreign currency hold- ings were kept outside the banking system to meet liquidity needs. Banks have tried to restore deposits since that time, but with little success. Meanwhile, Ukraine has a limited capital market to tap as a source of funding, with most of this market dominated by trad- ing in short-term government securities. At great risk, only the Na- tional Bank of Ukraine and the Savings Bank are meaningful in the Michael Borish and Fernando Montes-Negret 105 interbank market. Here too, resources are scarce. Through the in- terbank market, the Savings Bank has suffered significant losses and has been forced to reschedule what were originally short-term loans to other banks, some of which have failed. • Lending. Liabilities are matched by 3- to 12-month loans to state en- terprises or new, privately owned firms. These short-term loans have been priced at variable interest rates that can be reset every three months. Long-term lending is virtually nonexistent. The introduc- tion of the karbovanetz in January 1992 enabled the National Bank of Ukraine to create currency that was channeled into the economy through commercial banks. Commercial banks lent aggressively through 1993 with little or poor security, and paid no attention to cash flow in a market sense. This was subsequently followed by a period when banks recognized bad debts, triggering a decline in new lending flows. Since then, enterprises have had fewer incen- tives to place deposits with banks, which has added to the cycle of resource scarcity in the banking system. * Pricing. Commercial bank spreads on local currency (loans against deposits) were as high as 45 to 85 percent in 1995, although they have fallen since. High interest rate margins on loans have partially com- pensated for large classified debt levels since 1994, although they may not actually cover the full range of costs associated with portfolio prob- lems, taxes, reserve requirements, and other features that undermine banks' profitability (see Montes-Negret and Papi 1996). However, these high nominal rates were unsustainable for many enterprises, and weakened their cash flow and their ability to service interest fully and ultimately repay principal. (Effective interest rates on U.S. dollar loans ranged from 30 to 60 percent, and interest rates on Ukrainian cur- rency loans were also high.) These high nominal margins were pri- marily a function of a lack of deposits to fund loans, the poor quality of banks' loan portfolios, and the undeveloped level of competition within the banking system. Today, Ukraine's commercial banks are not a rnajor source of investment or working capital financing. Until 1994, foreign exchange transactions were the major source of in- come for barks. Spreads on these transactions ranged up to 10 percent, although margins had dropped to 2.4 percent by late 1994. Banks that had a greater capacity to obtain funds or approval for transactions from the National Banlc of Ukraine prospered. Contacts in the government played a significant role in this process. 106 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine Today, the Savings Bank with its extensive national network of about 15,000 branches is possibly the most important bank functioning in the market. It has served as a conduit for public payments such as households' accounts and pensions, and as a keeper and distributor of privatization certificates. Since May 1991, the Savings Bank has actively tried to develop the asset side of its operations by making housing loans and by extending loans to small enter- prises owned by people who hold deposits with it. It has also solicited com- mercial banks for loans and negotiated more favorable rates with them than what it could earn on credit to the National Bank of Ukraine. This has helped to develop the interbank market, whose size was constrained by regulatory lending rate ceilings. The Savings Bank remained the only bank that could offer its clients government guaranteed deposit insurance. This prompted an aggressive pricing strategy-lower interest rates on deposits than other com- mercial banks-the outcome of which was a decline in its share of household deposits to less than 60 percent by the end of 1994 despite the government insurance guarantee. The majority of savings deposits accumulated during 1991-94 were forwarded to the National Bank of Ukraine for its own internal use or for on-lending to comnercial banks. The Savings Bank incurred large losses as a result of its aggressive pricing and diversification strategy. In general, Ukraine's banking sector has yet to experience meaningful restructuring. This was both a cause and a consequence of Ukraine's weak economy. Since 1995-96 the authorities have made some progress in strength- ening the banking sector, largely through efforts to build supervisory capac- ity reform accounting standards, and enforce regulations when capital is severely impaired or when gross violations occur. Along with the moderate improvement in macroeconomic conditions, this represents a step forward. However, as seen elsewhere, macroeconomic stabilization is necessary but not sufficient. Ukraine will ultimately have to make faster and greater progress at the structural level to achieve competitiveness in the banking sector. This Table 7.6. The Distribution of Responisibilityfor Successftl Bank-ing Sector Refor-m Ingredient Governmnents Banks Macroeconomic Provide sound monetary, fiscal, Respond with efficient, well- environment and exchange rate framework managed, diversified operations to contain inflation rates, limit as net interest margins shrink fiscal deficits, and maintain sta- ble current and capital accounts Links to markets Encourage linkage across mar- Should be focused on feasible kets, products "universal" options Michael Borish and Fernando Montes-Negret 107 Table 7.6. (Continued) Ingredient Governments Banks Recapitalization Should have little or no adverse Should be one-time, up-front, macroeconomic impact linked to privatization time- table, and focused on strong governance and management to attract strategic domestic and/or foreign investment Enabling Build on stable macroeconomic Take advantage of opportunities, environment framework with open markets, including joint ventures with well-functioning institutions for experienced specialists; be inno- oversight, and supportive infra- vative, stress open information structure and recognize that markets are subject to constant change Political Sustained commitment to reform Sustained commitment to mar- commitment with long-term vision; rejection ketplace competitiveness based of short-term political pressures on global standards, not interim distortions Stakeholder Better banking performance Rewards (financial and other- incentives strengthens macroeconomic wise) for improved performance fundamentals Technical Assistance and training essential Training to professionalize assistance anid for policymakers and regulators standards and personnel; corres- training for safe and sound banking in pondent links with prime-rated support of a stable macroeco- banks formalize competitive nomic framework systems Sequencing of Institutional capacity and finan- Restructuring should be as reforms cial sector infrastructure needed; rapid as is feasible within the links between macroeconomic context of prudential standards framework and structural issues for safety and soundness should be recognized from the start; should include liquidation Governance and Incentives need to emphasize Should be focused on share ap- management global standards for sustaina- preciation and long-term com- bility, accountability, safety, and petitiveness based on global soundness; should reinforce en- standards abling environment efforts Privatization Institutional capacity and finan- Privatization should be as rapid cial sector infrastructure needed; as is feasible within the context links between macroeconomic of prudential standards for framework and structural issues safety and soundness and the should be recognized from the ability to attract strategic invest- start ment Table 7.7. Evaliiation of Banking Sector Restructuring Resiults Item Czech Republic Hungary Poland Ukraine Approach Centralized; up-front Decentralized; multiple Partly decentralized, up- No recapitalization; rapid carve-out and recapitali- recapitalizations; rapid front recapitalization and ownership transformation, zation; "good bank-bad privatization since 1995; slow, partial privatization but lacking in transparency bank"; partial SOCB most competitive bank- of SOCBs; partly central- and unaccompanied by privatization ing sector in region be- ized for large enterprises needed changes in incen- cause of enabling envi- in sensitive sectors tive structures ronment, foreign invest- ment Technical Limited assistance; more Some; high levels of Significant via twinning Some; more needed in ab- assistance and needed at state banks, foreign investment accel- and other forms sence of significant foreign co foreign regulatory agencies erated competitiveness investment participation Financial impact Bank solvency restored; Bank solvency restored; Bank solvency restored; Weak capital and liquidity on banks balance sheets restruc- balance sheets restruc- balance sheets restruc- because of high level of tured; more capital tured; recognition of tured; transparency nonperforming loans; in- needed because of in- nonperforming loans strengthened by loan sufficient provisioning, creased recognition of with introduction of loss provisioning, but transparency; limited role nonperforming loans stricter standards in undermined by doubts and level of financial inter- with introduction of 1993-94; SOCB earnings about enterprise restruc- mediation in economic stricter standards in 1994, less than those of private turing and ongoing credit development findings of extemal audi- banks despite holding 72 quality; earnings based tors with more severe percent of banking sys- on securities rather than assessment than banks' tem assets (1994); fiscal on lending internal views pressures lead to faster reforms Table 7.7. (Continued) Item Czech Reputblic Hungary Po1and Ukraine Bar'- Ongoing problems with SOCBs could not com- Slow privatization; Ongoing loan portfolio operations, SOCB loan portfolios; im- pete with private banks SOCBs partly protected problems; supervision management, and proved supervision re- after recapitalizations; from new foreign compe- improving, but continued governance quired; weak governance rapid privatization since tition from 1992-95; risk of weak governance because of ownership 1995-96; highly competi- SOCBs now have strong because of ownership structures; additional tive standards now in capital; SOCBs benefited structures; increased for- technical assistance place; weak supervision from twinning, strong eign investment, systems, needed governance since 1993 know-how needed Financial cost of US$10 billion (1991, US$3.5-5.0 billion Less than US$1 billion for No recapitalization; costs o recapitalizations 1993) = 15-20% of GDP (1991-94) = 2-3% of GDP SOCBs (1993) = less than from losses, high net in those years during the period 1% of 1993 GDP; other spreads, fiscal subsidies, costs from higher spreads, foregone growth disintermediation; more and efficiency for specialized banks Financial Serious questions about Improving performance Improving performance Weak performance; need perfornance of degree of restructuring, because of competition, because of real GDP for major overhaul at the enterprises levels of competitiveness high levels of foreign growth; concerns about structural level; informal because of governance; investment, and hardened delays in privatizing sector accounts for 60% of serious current account budget constraints since "strategic" firms, poten- economy, mostly micro and deficit in 1996 prompted 1995-96 tial costs in mining, steel, small-scale; foreign invest- emergency proposals for shipyards, agriculture ment needed reform in April 1997 Source: World Bank. 110 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine could be facilitated by a firm commitment to reorganize the banking sector, to impose stricter capital and governance requirements, to change the Bank- ing Law and the Central Bank Act, to provide more thorough oversight of bank management, and to attract the capital and expertise needed from for- eign banks to invigorate the banking market. Summary of Structural Reforms CEE reform programs have adopted a series of measures to transform ex- isting financial systems to serve a market-oriented economy, restructure and privatize enterprises, and to create stable macroeconomic conditions for growth. Their approaches to reform have met with varying degrees of success. Table 7.6 highlights some of the critical ingredients needed for successful reform. Table 7.7 evaluates results by country and approach. The Enabling Environnientfor Successful Banking All four countries have improved the enabling environment, particularly by restoring macroeconomic stability and confidence in national curren- cies. With the exception of Ukraine, they have introduced meaningful laws, regulations, and accounting frameworks consistent with market principles, and Ukraine has shown some movement in this direction. The CEE coun- tries have made progress with institution building, for example, banking supervision, credit rating systems, bankers' associations, and training pro- grams (Poland); with overall levels of competitiveness resulting from in- vestment by prime-rated banks and accelerated privatization since 1995 (Hungary); and with capital markets showing increasing investment inter- est in domestic banks and insurance companies (Hungary, Poland). Un- derlying all these developments is maintenance of a stable macroeconomic framework, as was the case in the Czech Republic until recently when its current account deteriorated, and is increasingly the case in Hungary and Poland. Accelerating the pace of industrial and financial sector privatization in Poland, sustaining a commitment to stable macroeconomic fundamen- tals in Hungary, and following through with ambitious structural reforms in the financial banking and enterprise sectors in the Czech Republic should enhance competitiveness and increase the countries' links to regional and global markets. This bodes well for banking, on the condition that they are prepared for global competition, can manage risk properly, and are willing to adopt the kinds of governance and accountability standards investors and regulators increasingly require. Michael Borish and Fernando Montes-Negret 111 In contrast, Ukraine (and most non-Baltic FSU cotntries) has failed to provide domestic and foreign investors with sufficient stability for long- term confidence. Notwithstanding modest improvements in the legal frame- work, recent regulatory interventions, and a generally open commitment to external investment, countries like Ukraine have far to go in establish- ing an enabling environment. Institutional capacity is weak, financial sec- tor infrastructure is limited, and capital markets activity is limited. Delays in developing basic financial infrastructure have been a serious drawback, particularly in accounting and legal matters. Achieving macroeconomic stability naas been a major challenge, and revamping the economy into one that is competitive will require a major overhaul. None of this bodes well for the banking sector in the near term. Adding to these problems is the general weakness of most of Ukraine's traditional trading partners, although its links with Poland and other countries of Central Europe are increasing as are its good relations with countries of the Organisation for Economic Co-operation and Development. Parameters of Competitionfor Successful Banking Transition countries show varying levels of competition in their financial mar- kets. Hungary has long been the most open to foreign investment, and has been the most profitable (along with Poland more recently) among transition countries' banking sectors (see Borish, Ding, and Noel 1996). Privatization of large SOCBs has accelerated quickly since 1995, and Hungary is likely to have virtually nc state ownership in the banking sector by late 1998 or early 1999. The Czech Republic has also had an open environment for investment in the banking system, much of it associated with mass privatization and derived from neighboring countries to provide financing and services for joint ven- tures and direct investment. However, the Czech Republic's interlocking direc- torates have reduced the importance of the stock exchange and information disclosure as a catalyst for heightened competitiveness (see OECD 1996). The emergency measures proposed in April 1997-full privatization of banks and enterprises, enhanced transparency and disclosure, strengthened regulation- reflect increased recognition of many of the structural weaknesses that need to be reformed. Poland's licensing policy was less open from 1992-95, although foreign banks remained the most profitable during that period. Poland is now increasingly focusing on the consolidation of remaining SOCBs, andbanks rep- resented a disproportionate 40 percent or more share of the Warsaw Stock Exchange's market capitalization in late 1996. How long Poland will take to privatize, liquidate, and/or merge its remaining SOCBs and specialized banks, 112 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine and the potentially distortionary effect these delays might have on further fi- nancial sector development, particularly given the depth of problems at the largest specialized banks, remains to be seen. However, Poland has long had a vision of privatization by the end of the century, and understands the need for satisfactory institutions to regulate banks and enterprises properly in an in- creasingly liberalized environment. The recentprivatization of Bank Handlowy and plans for additional banking privatization in 1998-99 reflect continued commitment to this direction. Noteworthy among CEE countries is the trend toward convergence in real interest rate trends. As they sustain stable frameworks and competi- tion, average real interest margins have increasingly converged. Ukraine has had an open environment for investment in the banking sec- tor from domestic and foreign sources, but its weak economy and lack of institutional capacity have undermined serious banking development. Along with most of the rest of the FSU, Ukraine is beginning to recognize that safe and sound banking practices are preconditions for stable and orderly growth. Proper oversight by investors and regulators (and later by depositors) is needed to achieve this, but to date Ukraine has lacked such oversight. Most countries initially had low capital requirements and flexible li- censing standards to promote entry and competition. This was particu- larly evident in Poland and Ukraine, where the number of banks increased quickly, but was also true in the Czech Republic and Hungary. However, this exemplifies an error in the sequencing of reforms, that is, liberalizing markets before establishing adequate legal, regulatory, supervisory, and enforcement capacity for stable market development. Recognizing this er- ror, CEE countries tightened minimum capital requirements. The Czech Republic and Hungary have capital standards higher than EU minimums, while Poland's criteria are currently consistent with the EU. However, a second key point is the development of capacity for regulatory enforce- ment in a forward-looking manner to contain systemic risks that could have a contagion effect and prompt lender of last resort or other emergency sce- narios with significant macroeconomic implications. Such risk management on the part of banks, investors, and regulators involves a range of vari- ables that, if exposed to excess volatility, could have a significant adverse impact on individual banks' capital and earnings and spread to the system at large. Regulatory capacity to prevent such problems is currently rela- tively weak in the Czech Republic. The view in Hungary is that foreign investment into the sector partly externalizes these risks. Poland has fo- cused on developing domestic capacity in these areas. Ukraine, which is not likely to become a member of the EU in the foreseeable future, has Michael Borish and Fernando Montes-Negret 113 greater capacity needs. As CEE countries make progress toward accession to the EUJ (having already joined the Organisation for Economic Co-opera- tion and Development), they have begun to implement recommendations from the Basle Committee. This will need ongoing strengthening as the Basle Committee itself begins to incorporate new risk-based techniques that go beyond traditional rules-based approaches. Proper application of prudential regulations has been a key institutional priority for orderly financial sector development. As systems become more open to competition, more rather than less regulation is needed. Poland dis- covered this in the early 1990s after following extremely flexible licensing prac- tices. Ukraine faced a similar set of circumstances and is now beginning to deal with these problems. However, in all countries, effective supervision and oversight is a dynamic process that must take the changing, interrelated, and increasingly complex features of financial sector development into account. For transition countries this requires time to develop, particularly as these countries have only recently introduced new laws, regulations, and account- ing framewvorks and moved away from passive, compliance-based oversight to more active forms of oversight. 18 Poland has made great strides toward effective supervision of banking. However, supervision requires strengthen- ing in Hungary. In the Czech Republic banking supervision has limited pre- ventive capacity given that structural weaknesses may be deeper than previ- ously thought, fraud may be more prevalent than can be tolerated, and bank debt represents a central part of economic financing. Ukraine has slowly moved to improve supervision, although enforcement may be more difficult than in CEE countr:ies. Ongoing development in this area will be needed to protect depositors and to manage risk for those issuing debt and equity. This will also require coordination with supervisory authorities across national borders. Methods of Recapitalization The performance of CEE countries and the methods chosen to recapitalize banks has varied. In some cases, such as the Czech Republic, the macroeco- nomic envire-nment appeared to be stable until late 1996-early 1997, when structural weaknesses were disclosed more openly. Despite high levels of eco- nomic growth, high and growing levels of nonperforming loans continued to 18. Most transition countries have only recently begun to introduce rules-based supervision corsistent with international standards. Some countries, such as Poland, are also introducing more recently pioneered risk-based supervision techniques in vary- ing degrees, pa:rticularly as they move closer to integration with more affluent EU countries through international agreements. 114 Restructuring Distressed Banks in Transition Econzomies: Central Europe and Ukraine weaken the financial condition of the large SOCBs in the Czech Republic. Recognition of a massive 8.6 percent current account deficit exposed the mag- nitude of structural weaknesses and demonstrated that macroeconomic sta- bility is not sufficient for successful reforn. Thus the up-front carve-out may have helped with the initial partial privatization of banks, but it does not appear to have been sufficient to create the conditions and incentives required for necessary structural reforms. In other cases, Hungary, for instance, has benefited from a strong en- abling environment and high levels of investment, which have raised the level of banking sector competition. However, macroeconomic fundamen- tals deteriorated in 1993-94, demonstrating that macroeconomic stability enhances structural reforms and competitiveness, while macroeconomic deterioration weakens or undermines them. The recapitalizations Hungary pursued largely reflected the recognition of losses at state banks and enter- prises and the government's desire to restore bank solvency prior to privatization and limit some of the more drastic restructuring requirements, for instance, layoffs, of loss-making enterprises. In the end, the govern- ment recognized that accelerated privatization and strategic investment were more efficient than recapitalization. Fortunately for Hungary, ongo- ing reforms dating back to the 1980s and a favorable enabling environ- ment made this approach feasible. Meanwhile, Poland has pursued a gradual approach to privatization, but a steady approach to institutional strengthening, development of regulatory capacity, and improvement of macroeconomic fundamentals. Reform of state commercial banks was fairly swift and time-bound, lead- ing to internal changes that made these banks more privatizable than in 1993. However, privatization has taken place more slowly. Poland's ef- forts to reform its SOCBs were relatively low-cost in terms of GDP, al- though intermediation rates remain low and questions persist about the ability of these banks to manage credit in an increasingly competitive environment. Poland's approach to the SOCBs points to much of what is necessary in terms of increasing macroeconomic stability and improv- ing structural capacity. However, how these banks will fare in more open markets remains to be seen. Most recapitalization has involved the issuance of bonds (and guarantees inthe case of Hungarv) to replacenonperformingloans onbankbalance sheets, thereby strengthening banks' solvency positions. Where these were interest- bearing bonds, carve-outs eased some of the liquidity problems banks experi- enced as a result of nonperforming loans. In the Czech Republic, the carve-out was conducted prudently so as to not create macroeconomic imbalances. In Michael Borish and Fernando Montes-Negret 115 Hungary, the highest financial costs coincided with worsening fiscal deficits, and theri2by may have contributed to macroeconomic imbalances. However, even more worrisome at the time was the risk of moral hazard and the institu- tional weaknesses of Hungary's supervisory system. The government of Hun- gary reacted successfully in 1995 with a vibrant commitment to privatization in the banlking sector. In Poland the recapitalization had less of an adverse effect on rnacroeconomic imbalances, although it did not address the two larg- est and mDst troubled specialized banks. Ukraine did not pursue recapitaliza- tion. Table 7.8 presents a review of CEE country performance. Restructuring, Privatization, and the Importance of Governance and Management In all three CEE countries, recapitalization occurred prior to privatization. In the Czech Republic, privatization was rapid as part of the 1992 mass privatization program, but only partial. In Poland, privatization was more gradual. Several SOCBs remain in state hands, although this is likely to change in the next two years. In Hungary, no timetable for privatization existed until 1995, although sustained commitment to reform and firm com- mitmnent to privatization from 1995 on has been successful. All the examples point to the need for improved bank governance, management, skills, sys- tems, and competitiveness, and show the limitations of recapitalization as a panacea for banking sector problems. Some of the more persistent prob- lems reflect the need for further and significantly greater investment in bank automation and skill development. In the Czech Republic, improved management and governance relate to the ongoing role of the state and the cross-ownership structure of banks, insurance companies, investment funds, and enterprises. While restrictions on the concentration of ownership and on lending to shareholders do ex- ist, these restrictions are probably insufficient to maintain disciplined fi- nancial management, to avert arm's length deals, and to prevent wider systemic risk in the banking and financial sector. Much of the current weak- ness in the economy relates to these structural weaknesses, which are now more openly reflected in the macroeconomic data that had previously pre- sented a picture of stability and competitiveness. In Hungary, the authorities made significant efforts to impose financial discipline on loss-making enterprises through strict bankruptcy laws and pro- cedures. Likewise, their recapitalization efforts included revised business plans, performance indicators, and operational reforms that were meant to make companies more competitive and creditworthy. However, even if bank man- agement was satisfactory in its credit decisionmaking, government ministries Table 7.8. Review of Bank Restruicturing Resuilts in CEE Courntries Country, costs, benefits, and results Positive Negative Czech Republic Considered only partly successful, with serious structural weaknesses remaining to be resolved. Costs Fiscally responsible. To pay for the recapitaliza- tion, offsetting cuts in public expenditure were made to achieve fiscal balance. C Benefits Five major banks were privatized (up to 63 per- Continued large share of assets held by SOCBs, cent private share ownership) along with enter- partial (not full) privatization of SOCBs, and prises in the 1992 mass privatization program. limited growth of private Czech banks reflect the limitations of the Czech approach to bank privatization. Governance still a major concern. Results Some improvement in the financial sector ena- Large banks still only partly privatized. Konsoli- bling environment combined with initial macro- dacni still needs high capital because of portfolio economic stabilization, fiscal balance, and bene- weaknesses. Major concerns about cross- fits of investment in the enterprise sector. ownership risks (National Property Fund, banks, insurance, funds). Weak supervision. Limited use of bankruptcy and liquidation to force serious restructuring. Table 7.8. (Continued) Country, costs, benefits, and results Positive Negative Hungary Recapitalization considered less suc- cessful than general commitment to enabling environment and strong commitment to privatization after recapitalizations occurred. Costs US$3.5 to US$5.0 billion in recapitalization aver- aged more than 2 to 3 percent of GDP per year from 1991-94. Main risk was a moral hazard, Coincided with period of costly fiscal deficits. May have crowded out some private investment. Benefits Policy of ongoing recapitalization from public resources reversed in 1995 in favor of a policy based on accelerated privatization. Results Hungary has created a favorable enabling envi- Failed to make SOCBs profitable or competitive, ronment for banks. The rising number of private although they were privatizable. Uncertain and foreign banks has added enhanced banking whether recapitalization (restored solvency) sector capacity and competition. directly assisted privatization, as acquisition prices would have been discounted relative to individual bank insolvency. (Table continues on thefollowing page.) Table 7.8. (Continued) Country, costs, benefits, and results Positive Negative Poland Up-front recapitalization for SOCBs considered successful. Recapitalization of specialized banks not successful. Costs Comparatively low cost in terms of GDP. Hard US$2 billion, most of which was for specialized budget constraints helped clean up state com- banks that should probably be liquidated or mercial bank portfolios. downsized. Weak or inadequate legal framework for specialized banks in housing and agriculture. , Benefits Restored solvency, strengthened governance and Some of the cost associated with bank restruc- Q0 management, tightened lending criteria, intro- turing was imposed on the private sector, be- duced incentives for employees and other cause of foregone investment resulting from stakeholders. difficulties in accessing credit from banks. Banks recapitalized from the purchase of government securities, a safe route to profitability that has helped finance fiscal deficits at the expense of private sector investment. Results SOCBs sufficiently recapitalized, and credit man- Poland's approach to banking reform has been agement strengthened. Supervision improved. gradual. Intermediation rates remain low. The Fast transfer of know-how through twinning portfolios of the troubled specialized banks re- arrangements with prominent West European main problematic. Management skills are some- commercial banks. times thin; seniority is often more important than performance. Restrictions on foreign bank entry from 1992-95 may have slowed progress. Weak automation. Michael Borish and Fernando Montes-Negret 119 responsible for the banks and loss-making enterprises sometimes overruled such decisions. Today, such risks appear to have vanished due to the strong commitment to privatization that has left all but one bank in the hands of the state. Privatization has been accompanied by strategic investment, which has shored up governance and management standards, thereby increasing confi- dence in tihe competitiveness of Hungary's financial sector in the long term. In Poland, governance in SOCBs has been fairly strong under the direc- tion of the Ministry of Finance. The Enterprise and Bank Restructuring Law made provisions for strict governance and management, set timetables for portfolio restructuring, and included incentives for banks in their recovery and collect ion efforts. This included shares to employees so that they could benefit from capital appreciation once privatization occurred, as well as so- phisticated training from Western institutions, which gave these individuals greater value in the marketplace as the banking sector continued to open up to competition. Hard budget constraints were employed to clean up state commercial bank portfolios. However, the portfolios of the troubled special- ized banks remain problematic, and intermediation rates remain low. Lessons Learned The following represent key lessons learned from banking reform in tran- sition countries and their approaches to distressed banking sectors. - Macroeconomic and structural issues are linked. Macroeconomic sta- bility is essential for savings mobilization, investment, and growth. Likewise, governments' willingness and ability to impose hard budget constraints on loss-making banks and enterprises is criti- cal to promote the larger objectives of macroeconomic stabiliza- tion, fiscal and external balance, and economic growth. Hungary is now benefiting from improvements in its macroeconomic fun- damentals, which are reinforced by more prudently managed banks and more creditworthy enterprises. Poland has shown progress at the structural level largely because of the imposition of hard budget constraints and strong governance in the banking sector. All this has occurred while macroeconomic fundamentals have steadily improved. Enhanced competitiveness in the Czech banking sector will depend on improvements at the structural level, and this should be reinforced by the Czech Republic's long-stand- ing emphasis on macroeconomic stability. Ukraine shows that achieving meaningful structural reform in the absence of macro- economic stability is difficult or impossible. 120 Restrncturing Distressed Banks in Transition Economies: Central Europe and Ukraine An enabling environment is essentialfor successfiul banking reform. Legal and regulatory reform are necessary, but not sufficient for banking reform. Also needed are court and out-of-court processes that are ef- fective in resolving debt disputes; property registries that provide needed information related to prospective collateral properties; effec- tive supervisory enforcement of prudential regulations; credit rating agencies that provide needed information and risk assessment; ana- lysts who review bank operations for investment purposes; banking associations that help to professionalize and maintain standards; ac- counting firms that audit banks; and financial media that actively cover financial sector issues for the public. Having an enabling environment for competitive, market-based banking includes equal treatment on issues of entry (licensing) and exit (liquidation, consolidation). Trans- parent criteria need to be established and implemented for bank reso- lution to maintain standards of competitiveness and service delivery. The enabling environment is positive in Hungary and is getting bet- ter in Poland. The Czech Republic will likely need to implement much of the package proposed in April 1997 to provide the needed trans- parency, accountability, and institutional framework for long-term sta- bility and growth. Ukraine lags behind these countries in virtually all categories, although it is making progress in terms of institutional ca- pacity building, for example, banking supervision. * Sustained political commitment for banking reform is indispensable. Governments, budgets, and the economy benefit from a safe, sound banking system. Conversely, an unstable banking system is detri- mental to governments, budgets, and the economy. Nevertheless, some of the draconian measures required during transition un- dercut public sector determination to accelerate reform and privatization in the banking sector. Countries that have pursued incomplete approaches to banking reform have incurred high costs in the form of unrestructured banks and enterprises, deferred prob- lems, and foregone competitiveness, all of which can result in con- tinuously poorly performing portfolios, the run-up of arrears, and efforts to tap scarce monetary and fiscal resources that could be better allocated elsewhere. Ukraine represented many of these characteristics until recently. Hungary and Poland represent good examples of ongoing commitment to reforms. Hungary since 1995 in particular represents a good example of commitment to rapid reform. Governments must sustain the political commitment needed for banking reforms. This will be important in the Czech Michael Borish and Fernando Montes-Negret 121 Republic as it closes the circle on a number of ownership, gover- nance, management, and regulatory issues. * The sequencing of reforms is important. Most transition countries have raised their capital requirements and tightened their licensing stan- dards in the banking sector after an early burst of liberalization. Po- land represents the best example of this among the three CEE coun- tries, with a proliferation of private banks before the authorities put in place the necessary laws, regulations, and institutions to monitor for risk. Since then, Poland has tightened requirements, and is now poised to privatize, liquidate, and/or merge most of its large state commercial banks by the end of the century. Profits are up, the stock exchange is heavily invested in banks (compared to other sectors), market participants and regulators are more confident that renewed commitment to open competition can be launched. Hungary already has sound institutions, which is one of the reasons why private in- vestmnent has steadily flowed into the financial sector. However, bank- ing supervision still needs some strengthening to assist with overall financial sector modernization. The Czech Republic will need to strengthen standards and establish enforcement capacity that is more preventive and forward looking. To date, standards in the Czech Re- public have been somewhat lax. All these challenges and develop- ments point to the need for sufficient institutional capacity for safe and sound banking and orderly markets. Ukraine is beginning to com- mit to institutional strengthening and an enabling environment for financial sector development, having recognized that ownership trans- formation does not work without new capital, market-based gover- nance and management, adequate legislation and regulations, and instituiJons that support and reinforce market development. * In a competitive environment governance and management must be market- basedfor banks to befinancially viable on a sustainable basis. The applica- tion of noncommercial criteria in an environment of scarce resources (hard budget constraints, run-up of arrears) usually leads to a misal- location of resources that ultimately reduces liquidity and solvency. Even if resources are not scarce, the emergence of market-based in- vestors requires that banks manage resources efficiently to meet re- turn targets. If not, their funding will decline and their position in the market will erode. Accountability and transparency are essential for strengthened governance and management, and for continued expan- sion of banks' funding bases. Hungary and Poland have faced up to this reality now that hard budget constraints are generally in place. 122 Restructuring Distressed Banks in Transition Economies: Central Europe and Ukraine The Czech Republic will likewise need to focus on this as an exten- sion of reforms already initiated to add liquidity to its capital markets and to reduce the centrality of bank debt as a means of financing in the economy. Ukraine is also facing up to this because of the weak- ness of its banks and the recognized need for a major transformation that has yet to occur. * Market-based privatization, not share-based ownership transformation or privatization using vouchers, is the best way to ensure market-based governance. Private investors have more incentives to manage their real cash capital prudently (or to have it managed) than do public officials or coupon holders. Ukraine's "privatization" of banks was conducted without serious governance and management changes, and cash capital put at risk was limited. This approach failed, as it has in many other parts of the FSU. While Poland has demon- strated that private ownership is not necessary for adequate gov- ernance, the temptation public sector officials face to soften bud- get constraints to increase the franchise value of their banks or to ensure lending to preferred firms, friends, and political allies on noncommercial bases jeopardizes macroeconomic fundamentals and safe banking practices. This is in evidence in Poland in some strategic sectors, and was true in Hungary on the justification of protecting jobs in an economy with official double-digit unem- ployment rates. There is concern that this is also the case with some of the Czech Republic's largest state banks and enterprises, par- ticularly since 1996, when the current account deficit started spi- raling to worrisome levels. e Suitable incentives n eed to be in place for a wide range of stakeholders for banking reform to succeed in the long term. Investors, lenders to banks, and depositors all represent the funding base of banks. The extent to which banks can generate confidence relative to the competition in eamings, debt service capacity, and overall safekeeping of deposits is the degree to which banks will provide needed incentives for long- term success. To make that happen, management and employees at all levels need to have incentives for competitive performance. In Poland, many bank employees (and employees of SOEs) were awarded shares at discounted prices to make privatization an ac- ceptable alternative and to promote improved performance for share appreciation purposes. Likewise in Hungary, significant training has been provided to improve professional skills. In addition, both coun- tries have improved their standards for accountability in recent years, Michael Borish and Fernando Montes-Negret 123 which was made possible with increasing disclosure as investment and stock market activity intensified. * Technical assistance and training from experienced market-based practi- tioners are essential building blocksfor institutional development. The fastest approach to developing the financial sector is to accelerate institutional development predicated on a viable legal and regula- tory framework and to promote competition among world-class institutions. However, this takes time. Poland's impressive enhance- ment of SOCB competitiveness (in contrast to that of its large, spe- cialized banks) and achievements in developing supervisory and enforcement capacity reflect the importance and value of technical assistance and training. Similarly Hungary has benefited signifi- canltly from foreign investment in the banking sector, and this has had a beneficial effect on overall competitiveness. By contrast, while the Czech Republic has attracted foreign investment, it has de-em- phasized the role of technical assistance. Consequently, the Czech Republic may have missed out on some essential building blocks for structural competitiveness along the way, particularly in the area of banking supervision and regulation. In Ukraine, foreign investment has been sparse, including in the banking sector. For Ukraine to become competitive, several changes will have to be made. One of the most important will be developing institutional capacity for safe and sound banking, which will require continued techrical assistance in banking as well as in supervision. * Links to regional and global markets are becoming increasingly crucial to financial sector competitiveness and economic growth. While banks in transition countries are sometimes just emerging from decades of centratl planning, the world market is becoming increasingly inter- twined. CEE countries will have the benefit of joining the EU in the near fature, and this is serving as an anchor for reform. FSU coun- tries generally lack this kind of anchor. In all cases, foreign invest- ment would greatly assist with new capital, management, systems, technologies, and market linkages. Hungary is the most notewor- thy beneficiary to date, and benefits have also accrued in the Czech Republic and Poland. Attracting foreign investment into the bank- ing sector is even more crucial for Ukraine (and other FSU coun- tries) to make up for greater weaknesses, and because accession to the EU is not likely to occur any time soon. * Banking, reform is closely linked to capital markets development. Banking and other financial sector activities are increasingly converging. Most 124 Restructuring Distressed Ban in Transition Economies: Central Europe and Ukraine of the world has followed universal banking practices, and interna- tional norms to address risk management across financial subsectors are gradually evolving. In terms of financing, the establishment of viable bond markets would ease demand on banks for riskier, long- term investment (project) financing. More active stock exchanges would improve bank funding, enhance overall risk management ca- pacity and oversight, and improve the capital structure of companies to which banks might lend and/or provide services. Poland and Hun- gary are moving forward in this respect, with prospects improving as their macroeconomic fundamentals improve and their governance re- quirements match international standards. However, stock market capitalization still remains relatively small as a percentage of GDP. In the Czech Republic, stock market turnover is limited and high levels of capitalization were based on a number of false premises. Ultimately, greater transparency and accountability will be required to increase liquidity and to provide more accurate measures of capitalization. Recapitalization of banks should be up-front, one-time,fiscally balanced, in- stitutionallyfocused, and linked to timetablesforfirll privatization. The Czech recapitalization, while the most expensive, was partly successful in that it was generally up-front and one-time (the 1993 recapitalization related to ruble-based debt), and fiscally balanced. However, its weak- ness is that it failed to achieve full privatization, and institutional strengthening was insufficient to impose market-based governance and management structures. In Poland, recapitalization of SOCBs was successful in most cases, but privatization has still not occurred in several of the banks. With regard to the specialized banks, Poland should consider a liquidation or downsizing scenario for nonsalvageable assets as part of its privatization strategy. The case of Hungary, the most successful in the region, shows the importance of timetables for full privatization supported by an enabling environ- ment and strong institutions. While Ukraine did not recapitalize its banks, its efforts should focus on the above principles. References Barbone, Luca, and Dominico Marchetti, Jr. 1994. "Economic Transformation and the Fiscal Crisis: A Critical Look at the Central European Experience of the 1990s." Policy Working Paper no. 1286. Europe and Central Asia Country Department II, World Bank, Washington, D. C. Borish, M., and M. Noel. 1996. Private Sector Development during Transition: The Visegrad Countries. Discussion Paper no. 318., Washington, D. C.: World Bank. Michael Borish and Fernando Montes-Negret 125 Borish, Michael S., Wei Ding, and Michel Noel. 1996. On the Road to EU Accession: Financial Sector Development in Central Europe. Discussion Paper no. 345., Wash- ington, D. C.: World Bank. Borish, M'ichael S., Millard F. Long, and Michel Noel. 1995. Restructuring Banks and Enterprises: Recent Lessons from Transition Countries. Discussion Paper no. 279., Washington, D. C.: World Bank. Dittus, Peter. 1994. "Bank Reform and Behavior in Central Europe." Journal of Com- parative Economics 19 (December): 335-61. IMF (International Monetary Fund). 1995. "Hungary. Recent Economic Develop- ments and Background Issues." Washington, D. C. Montes-Negret, F., and L. Papi. 1996. "Are Bank Interest Rates too High?" Policy Note. World Bank, Financial Sector Development Department, Washington, D. C. .1997. "The Polish Experience with Bank and Enterprise Restructuring." Policy Research Working Paper no. 1705. World Bank, Washington, D. C. OECD (Organisation for Economic Co-operation and Development). 1993. "OECD Economic Survey, Hungary." Paris. - 1996. "OECD Economic Survey, the Czech Republic." Paris. Saunders, A., and A. Sommariva. 1993. "Banking Sector and Restructuring in Eastern Europe." Journal of Banking and Finance 17 (September): 931-57. World Bank. 1997. World Development Indicators. Washington, D. C. 8 Preventing Banking Crises by Supporting the Truth Ivan Rernsik When analyzing processes that are part of the economic transforma- tions in Central European countries, many believe that problem loans, which are generally accepted as the major contributor to banking cri- ses, are always connected to previous regimes rather than to loans granted in the new market economy era. Therefore they believe that carving out old, bad loans from commercial banks and putting them into loan "hospitals" solves the problem. In the real world, new loan exposures and new securities exposures are just as threatening to the commercial banking sector's stability as inherited loans. In the early days of economic transformation any process used to iden- tify problem assets is often subjective and exposed to various political and other pressu;res. Promising assets can become nonperforming later because of changed circumstances, and commercial bankers are often tempted to evergreen inherited loans for certain clients in the hope of cashing in fat interest margins and/or lucrative, fee-based export-import transactions. The new market economy era brings with it new pressures. In the ab- sence of private capital, commercial banks cannot resist political pressures and have to substitute for missing capital by extending privatization loans to a new generation of "entrepreneurs" who have no money and no idea how to run a business. It is thus understandable that large parts of privatization loan portfolios are today classified as lost. Commercial bank- ers also have to extend new loans to existing enterprises to pay workers so as to prevent social disturbances in fragile democracies. 127 128 Preventing Baiking Crises by Supporting the Truth Liberalizing banking and creating new, private banks put more pressure on the management of already established banks. They are expected to provide new entrants with money market lines, and some managers agree to do so as not to be labeled as being opposed to com- petition or reform. A lot of these new banks do not exist any longer, but their unpaid debt does. Even decisions on new "greenfield" projects using the sophisticated tools of Western project finance analysts is still subjective and open to failure. In the absence of financial data, decisionmaking is often more intuitive than scientific, and monitoring the performance of such projects is difficult. What has happened in some countries and what will happen in others obviously cannot be changed. However, what they must all do is use fi- nancial reporting as a basis for making decisions and measuring progress, and thus assessing the success of the economic transformation process. As concerns loan loss provisioning, transition economies need to set up well-defined loan classification systems and enforceable loan loss provi- sioning rules at the beginning of an economic transition. The Czech Re- public introduced its first such system and rules in 1994, six years after it had introduced a two-tier banking system. A sound loan classification and loan loss provisioning system should have been in place much earlier to enable the disclosure of old loan exposures or their residuals, as well as of new exposures, at their fair value. This would have helped prevent the over-rating of the success of the Czech economic transformation. Let us consider a simplified example of the loan loss provisioning rules currently in effect in the Czech Republic to demonstrate some of the prob- lems of the country's banking sector and economy. Banks in the Czech Re- public are allowed to set aside a maximum of 1 percent provision per year if a loan is 30 days overdue, a maximum of 5 percent if it is 90 days over- due, a maximum of 10 percent if it is 180 days overdue, and a maximum of 20 percent if it is 360 days overdue. This illustrates sufficiently that the whole concept of loan loss provisioning is not quite correct. Central banks around the world require minimum levels of loan loss provisions to main- tain the health of commercial banking systems, not maximum levels. How- ever, in the Czech Republic more provision allowances are permissible if bankruptcy has been initiated, but this is still rare. Some central banks in transition economies take the tax approach to loan loss provisioning rules as they try to solve two problems at the same time: what is the required level of loan loss provisions and what should their tax treatment be in profit and loss accounts. This creates more prob- lems than it solves: some Czech banks have not built adequate levels of Ivan Remsik 129 loan loss provisions simply because they have not been required to do so. However their interpretation is different. They say that they have not been allowed to do so. Another point in connection with the tax treatment of loan loss provisions is that the maximum amoumt of tax allowable loan loss provisions is limited to 3 percent of a loan portfolio per year. Such a low limit is more than conservative for any transition economy. On the other side there is no limit to loan loss provisions without tax relief. When I was responsible for restructuring the balance sheet of the Zivnostenska Bank when preparing the bank for privatization, the only possibility was to build adequate loan loss provisions out of the bank's equity reserves. This was perfectly legal. However, commercial bankers are reluctant to use their equity reserves as loan loss provisions in the be- lief that they will lose their negotiating power when trying to convince governments to increase the level of tax allowable loan loss provisions if their balance sheets look fine. They are also afraid that if they build loan loss provisions from their equity reserves, and thus state their assets at fair value, they will not be able to claim tax relief on provisions in subsequent years because additional provisions would no longer be needed. No de- ferred tax credit system exists. Banks are not allowed to recognize deferred tax assets a-nd claim deferred tax credits in following tax years as they can in the United Kingdom. Another notable regulation in the Czech Republic is that unpaid inter- est must be recognized as income and tax must be paid on unpaid interest. Some years ago I refused to sign the Zivnostenska Bank's accounts show- ing unpaid interest as income, because as far as I know it is considered a crime in sorne countries. However, I instructed the Finance Department to calculate and pay tax on it so as not to break Czech tax laws. The Czech banking industry has a track record of attempting to deter- mine realistic asset values. The result is not yet perfect, but the industry is moving in the right direction. The most recent step was in 1997, when the Central Bank introduced a system for classifying securities holdings in in- vestment and trading books with relevant reporting, valuation, and provi- sioning policies that were in line with international practices. This hap- pened some eight or nine years after the creation of a two-tier banking system and after grandiose privatization that had involved a massive issu- ance of securities. Corporate financial reporting in Central Europe is generally far from any accepted standard. Assets are stated at cost, including the assets of newly privatized entities; mark-to-market valuation for any class of assets, including financial assets such as shares or fixed income securities, is not 130 Preventing Banking Crises by Supporting the Truth required; and requirements for the provisioning of a debtor's book do not exist. Income statements and audits of such statements are tax driven. Some of the more than 1,000 licensed auditors in the Czech Republic can audit a large company in a single day as they just look at income statements and check whether the company has calculated its taxes correctly. In such an environment, it is difficult for bankers to carry out sensible corporate credit analyses and to make prudent credit decisions. What is even more danger- ous is that industrial holdings and privatization and mutual funds, which emerged as part of the transition economies' transformation, value their assets at cost and are not required to do mark-to-market valuation or to create provisions for undervalued assets. Thus we need to look at the entire economy: enterprises and compa- nies do not tell the truth; their owners, such as privatization and mutual funds, do not tell the truth; the banks do not tell the truth. So we wonder, can official macroeconomic data be correct? The recent crisis in the Czech banking sector and the less than optimistic economic reality indicate that having had true information available during the various stages of the country's economic transition would have been better. To prevent simi- lar scenarios from occurring again we should do everything to support the truth. Central banks have done something. It was not on time, it is not perfect, but they are on the right track. I am not so sure about international monetary authorities. When I was in banking most International Monetary Fund and World Bank missions came to see me to find out what was really going on in the country. They would receive a lot of promising information in the morning while visiting government officials and realistic information in the afternoon when visit- ing commercial bankers. The commercial bankers asked for guidance from these missions on how to set up prudential regulations and policies. Un- fortunately, some reluctance to provide such help has been apparent. However, the most critical period is still ahead of us. For the Czech government, proud of its privatization results, introducing sensible account- ing and valuation policies for the nonbanking sector will be difficult. They are unlikely to tell public shareholders that under their new guidelines the shareholders' securities in company ABC or mutual fund XYZ have no value. The reality is that they are unlikely to do so voluntarily. International monetary institutions should take a greater role in the process of discovering and telling the truth to policymakers, bankers, and the general public. Progress will not occur without the truth, and support- ing dissemination of the truth is the best way to prevent banking crises. Part IV. Bank Failures in Industrial Countries 9 Lessons from Bank Failures in the United States James R. Barth and Robert E. Litan In mid-1997 the U.S. Congress and the Clinton administration were actively considering proposals to modernize the financial services industry. At the time, support for easing the restrictions separating banks, securities firms, and insurance companies was fairly broad. There was far less support for mixing banking and commerce. Indeed, House Banking and Financial Ser- vices Committee Chairman James Leach testified that "Mixing commerce and banking simply doesn't fit our kind of democracy" (Leach 1997, p. 5).1 The debate in the United States about the appropriate role of banks in the overall financial system was not occurring in isolation. Many other indus- trial and developing countries were also assessing various banking laws and regulations to determine which were most appropriate in an increasingly tech- nological and global banking world. The intense and widespread focus on bank regulation at this particular time was largely due to the occurrence of significantbanking problems in more than half the world's countries between 1. Despite this view, in late June most members of the House Banking and Finan- cial Services Committee voted to permit a limited degree of bank ownership of nonfi- nancial firms and vice versa. However, many considered the prospect for legislation being enacted in 1997 that contained even a limited amount of such mixing to be doubt- ful. A majority of members also voted to eliminate many of the barriers separating banks, securities firms, and insurance companies. 133 134 Lessons from Bank Failires in the United States 1980 and 1997 (see, for example, Caprio and Klingebiel 1996; Lindgren, Garcia, and Saal 1996). These problems were still ongoing in many countries and had developed in both industrial countries and emerging market economies. To prevent future problems, countries were working individually to design regulatory structures that better ensured more stable, efficient, and competi- tive banking industries. Countries were also working jointly, frequently with the assistance and advice of such international agencies as the International Monetary Fund and the World Bank, to ensure that such structures less- ened the likelihood that financial disruptions in one country would spread to other countries (see Goldstein 1997). This chapter attempts to contribute to our understanding of the relation- ship between banking problems and bank regulation by examining the bank- ing problems that occurred in the United States during 1980 through 1996. As has been well documented in the literature, many factors were respon- sible for the problems (see, for example, Barth 1991; Barth and Bartholomew 1992; Barth and Brumbaugh 1994; Barth, Brumbaugh, and Litan 1992; Bartholomew and Whalen 1995; Brewer 1995; Brumbaugh 1988; Kane 1985, 1989b; Kaufman 1995; Kroszner and Strahan 1996; Litan 1987, 1991, 1994; Romer and Weingast 1992; White 1991). Most agree, however, that bank regulation bears an important part of the responsibility for what happened. In particular, overly restrictive laws and regulations contributed to thou- sands of depository institutions being exposed to substantial interest rate risk in the late 1970s and early 1980s. Subsequently, especially in the early to mid-1980s, lax regulation and supervision enabled many inadequately capi- talized institutions to grow rapidly by engaging in high-risk activities. Bank Failures, Resolution Costs, and Related Developments Tables 9.1-9.4 and figure 9.1 present selected information about U.S. feder- ally insured banking institutions from 1980 through 1996. In particular, the tables indicate the extent of the problems of commercial and savings banks, savings and loan institutions (S&Ls), and credit unions. Note the following important points about the information contained in tables 9.1-9.4 and figure 9.1. First, between 1980 and 1996, 5,207 federally insured institutions with US$920 billion in assets failed and cost an estimated US$192 billion in nominal dollars to resolve. (These totals are obtained by simply summing the appropriate figures under the heading "Resolution" at the bottom of tables 9.1-9.3.) When expressed in inflation-adjusted dollars, that is, 1992 dollars, the costs are US$216 billion, or about 3.5 percent of 1992 GDP. Not all the different types of depositories experienced the same degree James R. Barth and Robert E. Litan 135 Table 9.1. LI.S. Federally Insured Commercial Bank Industry, 1980-96 Category 1980 1981 1982 1983 1984 1985 Number of -nstitutions 14,435 14,408 14,446 14,460 14,483 14,407 Number of branches 38,738 40,786 39,783 40,853 41,799 43,293 Number of full-time equivalent employees (thousands) 1,442 1,489 1,499 1,509 1,527 1,562 Total assets (US$ billions) 1,856 2,029 2,194 2,342 2,509 2,731 Equity capital (US$ millions) 107,599 118,241 128,698 140,459 154,103 169,118 Net after tax income (US$ millions) 14,010 14,722 14,844 14,931 15,502 17,977 Taxes (US$ rrillions) 4,658 3,904 3,037 4,017 4,721 5,629 Percentage of real estate loans to total assets 14.5 14.4 14.0 14.4 15.4 16.1 Percentage of commercial & industrial loans to total assets 21.1 22.4 23.0 22.4 22.5 21.2 Percentage of agriculture production loans to totzd assets 1.7 1.7 1.7 1.7 1.6 1.3 Percentage of loans to individuals to total asse:s 10.1 9.5 9.1 9.6 10.6 11.3 National commercial banks Percentage of number of institutions 30.7 30.9 31.7 32.8 33.8 34.4 Percentage of total assets 56.9 57.2 57.4 58.0 59.7 59.8 State, federal memnber commercial banks Percentage of number of institutions 6.9 7.1 7.2 7.3 7.3 7.4 Percentage of total assets 17.4 17.0 17.6 16.6 18.2 18.1 Equity capital to asset ratio categories More than 8% Number 7,981 7,941 7,976 7,674 7,423 7,324 Total assets (US$billions) 278 300 320 344 353 377 6% to 8% Number 5,401 5,411 5,292 5,280 5,441 5,339 Total assets (US$ billions) 434 466 494 560 648 710 3% to 6 % Number 1,038 1,041 1,141 1,426 1,507 1,571 Total assets (US$ billions) 1,141 1,260 1,375 1,432 1,498 1,633 1.5% to 3% Number 12 16 27 53 61 84 Total assets (US$ billions) 2 4 3 4 7 6 0% to 1.5% Number 2 5 12 21 23 36 Total assets (US$ bilhons) 1 2 1 2 Less than 0% Number 1 2 5 14 11 29 Total assets (US$ billions) 1 I Number of problern commercial banks - 196 326 603 800 1,098 Assets of problem commercial banks (US$ billions) - - - - - - Resolutions of commnrcial and savings banks Number 10 10 42 48 80 120 Total assets (US$: millions) 236 4,859 11,632 7,027 3,276 8,735 Estimated present value cost (U3S$ millions) 31 782 1,169 1,425 1,635 1,044 Number of months rated 4 or 5 before closure 15 19 16 19 15 15 (Table continues on thefollowing page.) 136 Lessonsfrom Bank Failiures in the Tnited States Table 9.1. (Continued) Category 1986 1987 1988 1989 1990 1991 Number of institutions 14,199 13,703 13,123 12,709 12,343 11,921 Nurnber of branches 44,392 45,357 46,381 48,005 50,406 51,969 Number of full-time equivalent employees (thousands) 1,563 1,545 1,527 1,532 1,518 1,487 Total assets (US$ billions) 2,941 3,000 3,131 3,299 3,389 3,431 Equity capital (US$ millions) 182,144 180,651 196,545 204,823 218,616 231,699 Net after tax income (US$ millions) 17,418 2,803 24.812 15,575 15,991 17,935 Taxes (US$ millions) 5,266 5,404 9,988 9,540 7,704 8,265 Percentage of real estate loans to total assets 17.5 20.0 21.6 23.1 24.5 24.8 Percentage of comnmercial & industrial loans to total assets 20.4 19.7 19.2 18.8 18.2 16.3 Percentage of agriculture production loans to total assets 1.1 1.0 1.0 0.9 1.0 1.0 Percentage of loans to individuals to total assets 11.4 11.7 12.3 12.1 11.9 11.4 National cornmercial banks Percentage of number of institutions 34.3 33.7 35.3 32.9 32.2 31.8 Percentage of total assets 59.3 59.1 58.9 59.7 58.6 56.5 State, federal member commercial banks Percentage of number of institutions 7.7 7.9 8.1 8.1 8.2 8.2 Percentage of total assets 18.2 17.6 17.1 16.4 16.5 16.9 Equity capital to asset ratio categories More than 8% Number 6,699 6,790 6,646 6,741 6,377 6,422 Total assets (US$ billions) 384 438 437 488 554 615 6% to 8% Number 5,169 4,944 4,703 4,407 ,340 4,309 Total assets (US$ billions) 755 846 944 1,113 1,088 1,377 3% to 6 % Number 2,028 1,604 1,401 1,295 1,234 1,048 Total assets (US$ billions) 1,776 1,588 1,680 1,572 1,684 1,421 1.5%, to 3% Number 124 165 158 109 92 66 Total assets (US$ billions) 14 111 39 76 41 13 0% to 2.5% Number 88 108 96 73 66 43 Total assets (US$ billions) 6 5 10 26 6 3 Less than 0% Number 71 76 104 82 35 32 Total assets (US$ bilhons) 4 10 21 23 15 2 Number of problem commercial banks 1,457 1,559 1,394 1,092 1,012 1,016 Assets of problem commercial banks (US$ billions) 286 329 305 188 342 528 Resolutions of commercial and savings banks Number 145 203 221 207 169 127 Total assets (US$ millions) 7,638 9,231 52,683 29,402 15,729 62,524 Estimated present value cost (US$ millions) 1,728 2,028 6,866 6,215 2,889 6,037 Number of months rated 4 or 5 before closure 20 21 24 28 34 29 James R. Barth and Robert E. Litan 137 Table 9.1. (Continued) Category 1992 1993 1994 1995 2996 Number of institutions 11,462 10,958 10,451 9,940 9,528 Number of branches 51,935 52,868 55,145 56,513 57,215 Number of full-time equivalent employees (thousands) 1,478 1,494 1,488 1,484 1,489 Total assets 'US$ billions) 3,506 3,706 4,011 4,313 4,578 Equity capital (US$ millions) 263,403 296,491 312,088 349,578 375,295 Net after tax income (US$ millions) 31,987 43,036 44,624 48,749 52,390 Taxes (US$ millions) 14,481 19,838 22,426 26,176 28,227 Percentage of real estate loans to total assets 24.8 24.9 24.9 25.0 25.9 Percentage of commercial & industrial loans to total assets 15.3 14.5 14.7 15.3 15.5 Percentage of agriculture production loans to total assets 1.0 1.0 1.0 0.9 0.9 Percentage of loans to individuals to total assets 11.0 11.3 12.2 12.4 12.4 National commercial banks Percentage of number of institutions 31.5 30.3 29.4 28.8 28.6 Percentage of total assets 57.2 56.7 56.3 55.7 55.3 State, federal member commercial banks Percentage of number of institutions 8.4 8.8 9.3 10.5 10.7 Percentage ot total assets 18.2 19.6 21.1 22.8 24.6 Equity capital to asset ratio categories More than 8% Number 6,857 7,542 6,969 7,497 7,104 Total assets (US$ billions) 994 1,365 1,160 1,719 1,757 6% to 8% Number 3,924 3,151 3,074 2,281 2,261 Total assets (LtS$ billions) 1,894 2,09D 2,185 1,998 2,116 3% to 6 % Number 624 263 398 159 158 Total assets (US$ billions) 626 250 667 59 709 1.5% to 3% Number 24 7 8 3 4 Total assets (US$ billions) 2 1 0.5 0% to 1.5% Number 14 6 4 2 1 Total assets (US$ billions) 3 1 Less than 0% Number 23 2 0 0 0 Total assets (USS billions) 8 0 0 0 Number of problem commercial banks 787 426 247 144 82 Assets of problem cornmercial banks (US$ billions) 408 242 33 17 5 Resolutions of comm, rcial and savings banks Number 122 41 13 6 5 Total assets (US$ umillions) 45,485 3,527 1,402 753 190 Estimated present value cost (US$ millions) 3,707 655 208 104 - Number of mon&,s rated 4 or 5 before closure 32 - - - - - Not available. * Less than US$500 million. Source; Barth and Brumbaugh (1994); Federal Deposit Insurance Corporation, Office of the Comptroller of the Currency, Bank Research Division data. The formatis adapted from Barth, Brumbaugh, and Litan (1990, p. 25). 138 Lessonsfrom Bank Failutres in the United States Table 9.2. U.S. Federally Insuired Savings and Loan Industry, 1980-96 Category 1980 1981 1982 1983 1984 1985 Number of institutions 3,993 3,751 3,287 3,146 3,136 3,246 Total RAP assets (US$ billions) 604 640 686 814 978 1,070 GAAP capital (US$ billions) 32 27 20 25 27 34 Tangible capital (US$ billions) 32 25 4 4 3 9 Net after tax income (US$ millions) 781 (4,631) (4,142) 1,945 1,022 3,728 Net operating income (US$ millions) 790 (7,114) (8,761) (46) 990 3,601 Net non-operating income (USS millions) 398 964 3,041 2,567 796 2,215 Taxes (US$ milions) 407 (1,519) (1,578) 576 764 2,087 Percentage of home mortgages to total assets 66.5 65.0 56.3 49.8 44.9 42.4 Percentage of mortgage-backed securities to total assets 4.4 5.0 8.6 10.9 11.1 10.4 Percentage of mortgage assets to total assets 70.9 70.0 64.9 60.7 56.0 52.8 Stock institutions Percentage of number of institutions 20 21 23 23 30 33 Percentage of total assets 27 29 30 42 52 56 Federally-chartered Percentage of number of institutions 50 51 51 53 54 53 Percentage of total assets 56 63 70 69 66 65 Tangible capital to asset ratio categories More than 6% Number 1,701 1,171 787 661 643 806 Total assets (US$ billions) 181 101 59 84 62 95 3%0 to 6% Numnber 1,956 1,766 1,202 1,091 945 1,009 Total assets (US$ billions) 379 348 190 222 227 259 2.5% to3% Number 230 524 592 569 526 460 Total assets (US$ billions) 39 113 136 185 168 212 0% to 1.5% Number 63 178 291 310 327 266 Total assets (US$ billions) 4 50 81 88 153 135 Less than 0% Number 43 112 415 515 695 705 Total assets (USS billions) 0.4 29 220 234 336 335 Conservatorships at Resolution Trust Corporation Number _ _ _ _ - _ Total assets of institutions taken over (US$ billions) - - - - - - Number of problem savings and loans 330 499 744 689 748 679 Assets of problem savings and loans (US$ billions) 43 104 182 189 265 270 Resolutions Number 11 28 76 54 27 36 Total assets (US$ millions) 1,458 13,908 27,748 19,655 5,783 7,066 Estimated present value cost (US$ millions) 167 1,018 1,213 1,024 833 1,025 Number of months reporting tangible insolvent before closure 5.4 5.2 12.9 16.4 23.4 25.9 James R. Barth and Robert E. Litan 139 Table 9.2. (Continued) Category 1986 1987 1988 1989 1990 1991 Number of institutions 3,220 3,147 2,949 2,616 2,359 2,110 Total RAP assets (US$ billions) 1,164 1,251 1,352 1,187 1,029 895 GAAP capital (US$ billions) 39 34 46 52 52 53 Tangible capital (US$ billions) 15 9 23 - - 42 Net after tax income (US$ miUlions) 131 (7,779) (12,057) (6,783) (3,817) 1,195 Net operating income (US$ millions) 4,562 2,850 907 (8,308) (4,022) 2,265 Net non-operating income (US$ millions) (1,290) (7,930) (11,012) 2,198 1,347 1,356 Taxes (US$ millions) 3,141 2,699 1,952 673 1,142 2,426 Percentage of home mortgages to total assets 38.9 37.8 38.6 41.2 43.0 45.6 Percentage of mortgage-backed securities to total assets 13.1 15.6 15.4 14.2 14.5 14.2 Percentage of mortgage assets to total assets 52.0 53.4 54.0 55.4 59.5 59.8 Stock institutions Percentage of number of institutions 37 40 43 43 44 45 Percentage of total assets 62 65 68 69 74 76 Federally-chartered Percentage of number of institutions 53 56 59 61 64 65 Percentage of total assets 66 66 72 76 83 84 Tangible capital to asset ratio categories More than 6% Number 972 1,113 1,136 1,180 1,132 1,148 Total assets (US$ billions) 156 188 196 206 195 227 3% to 6% Number 995 891 864 813 837 763 Total assets (l]S$ billions) 316 356 418 480 484 468 1.5% to 3% Number 354 277 281 245 163 105 Total assets (US$ bilions) 191 196 244 206 154 104 0% to 2.5% Number 227 194 160 120 101 47 Total assets (US$ billions) 144 143 182 59 83 36 Less than 0% Number 672 672 508 239 109 33 Total assets (US3$ billions) 324 336 283 192 89 41 Conseroatorships at Resolution Trust Corporation Number - - - 318 207 123 Total assets of in-stitutions taken over (US$ billions) - - - 142 127 70 Number of problem savings and loans 637 574 - 404 450 340 Assets of problem savings and loans (US$ billions) 249 217 - 202 235 224 Resolutions Number 51 47 222 327 213 144 Total assets (US$ millions) 24,182 10,921 113,965 146,811 134,766 82,626 Estimated present value cost (US$ millions) 3,605 4,509 52,203 51,140 21,473 10,823 Number of months reporting tangibleinsolventbeforeclosure 30.6 35.7 42.0 36.0 43.0 41.0 (Table continues on thefollowing page.) 140 Lessonsfrom Bank Failures in the United States Table 9.2. (Continued) Category 1992 1993 1994 1995 1996 Number of institutions 1,871 1,669 1,543 1,437 1,334 Total RAP assets (US$ billions) 807 775 774 771 769 GAAP capital (US$ billions) 56 58 58 62 61 Tangible capital (US$ billions) 52 54 55 57 56 Net after tax income (US$ millions) 5,103 4,917 4,275 5,360 4,750 Net operating income (US$ millions) 6,855 7,141 6,597 7,460 - Net non-operating income (US$ millions) 1,047 595 422 835 - Taxes (US$ millions) 2,779 2,819 2,744 2,935 1,748 Percentage of home mortgages to total assets 45.7 45.8 47.0 47.4 49.9 Percentage of mortgage-backed securities to total assets 14.5 15.4 16.5 16.3 14.4 Percentage of mortgage assets to total assets 60.2 61.2 63.5 63.7 64.3 Stock institutions Percentage of number of institutions 49 54 56 59 60 Percentage of total assets 80 82 86 86 90 Federally-chartered Percentage of number of institutions 70 75 78 82 82 Percentage of total assets 87 90 93 95 96 Tangible capital to asset ratio categories More than 6% Number 1,246 1,342 - - - Total assets (US$ billions) 310 397 3% to 6% Number 559 323 - - - Total assets (US$ billions) 435 372 1.5% to 3% Number 39 2 - - - Total assets (US$ billions) 33 4 - - - 0% to 1.5% Number 7 2 - - - Total assets (USS billions) 13 3 - - Less than 0% Number 3 0 - - - Total assets (US$ billions) 4 0 - - - Consenatorships at Resolution Trist Corporation Number 50 8 0 0 - Total assets of institutions taken over (US$ billions) 35 6 0 0 - Number of problem savings and loans 203 101 53 41 29 Assets of problem savings and loans (US$ billions) 134 77 30 11 5 Resolutions Number 59 9 2 2 1 Total assets (US$ millions) 45,980 6,339 142 456 - Estimated present value cost (US$ millions) 4,741 532 14 66 Number of months reporting tangible insolvent before closure 38.0 - - - - - Not available. GAAP Generally accepted accounting principles. RAP Regulatory accounting practices. Source: Barth (1991); Barth and Brumnbaugh (1994); Resolution Trust Corporaton (various years); datafrom the Federal Home Loan Board, the Office of Thrift Supervision, and the Office of the Comptroller of the CurTency. James R. Barth and Robert E. Litan 141 Table 9.3. U.S. Federally Insured Credit Union Industry, 1980-96 Category 1980 1981 1982 1983 1984 1985 Number of institutions 17,350 16,960 16,424 15,804 15,180 15,033 Total assets (US$ billions) 61 65 70 82 93 120 Capital (USS billions) 3.7 4.3 4.7 5.3 6.2 7.8 Net income (US$ millions) 314 677 714 747 1,131 1,303 Net operatirg income (US$ millions) 502 882 922 930 1,316 1,538 Net non-operating income (US$ millions) 0.8 7 1 12 11 66 Provision foi loan losses (US$ millions) 190 212 209 194 195 301 Federal incorne taxes (US$ millions) 0 0 0 0 0 0 Percentage of 1st mortgages to total assets 4.7 4.4 3.3 3.7 3.9 4.8 Percentage of mortgage-backed securities to total assets - - - - - - Percentage of total real estate assets to total assets - - - - - - Federally chartered Percentage of number of institutions 71.7 70.8 69.4 69.1 69.4 67.3 Percentage of total assets 65.8 65.9 65.3 66.5 68.6 65.3 Federally insur ed state-chartered Percentage of number of institutions 28.3 29.2 30.6 30.9 30.6 32.7 Percentage of total assets 34.2 34.1 34.7 33.5 31.4 34.7 Capital to asset ratio categories Mere than 6% Number 10,286 11,282 11,134 10,460 10,763 10,552 Total assets (US$ billions) 26.2 30.7 34.2 38.1 46.2 58.3 3% to 6% Number 4,563 3,811 3,725 3,894 3,490 3,676 Total assets (US$ billions) 24.3 22.8 27.4 34.4 39.2 52.6 1.5% to 3% Number 1,357 1,083 944 960 612 560 Total assets (US$ billions) 4.8 5.4 4.9 6.3 5.2 6.7 0% fo 1.5% Number 872 582 481 347 208 172 Total assets (U",$ billions) 5.1 4.5 2.7 1.7 1.1 1.0 Less fhan 0% Number 272 202 140 143 107 73 Total assets (US$ billions) 0.6 1.1 0.4 1.4 1.3 1.2 Number of problem credit unions 1,180 1,174 1,192 1,124 872 742 Assets of problem credit unions (US$ billions) 2.4 3.0 4.6 4.7 4.1 4.1 Resolutions Number 239 349 327 253 130 94 Total shares (USS millions) - 136 156 102 208 47 Estimated presentt value cost (US$ millions) 33 44 79 55 20 12 Number of months rated 4 or 5 before closure - - - 69.6 80.8 64.9 (Table continues on thefollowing page.) 142 Lessonsfrom Bank Failures in the United States Table 9.3. (Conitiniued) Category 1986 1987 1988 1989 1990 1991 Number of institutions 14,687 14,335 13,878 13,371 12,860 12,960 Total assets (US$ billions) 148 162 175 184 198 227 Capital (US$ billions) 9.2 10.6 12.0 13.5 15.0 17.4 Net income (US$ millions) 1,366 1,464 1,659 1,653 1,691 2,066 Net operating income (US$ millions) 1,746 2,074 2,310 2,445 2,609 3,026 Net non-operating income (US$ millions) 113 (36) 38 21 28 36 Provision for loan losses (US$ millions) 494 574 688 813 946 996 Federal income taxes (US$ millions) 0 0 0 0 0 0 Percentage of 1st mortgages to total assets 7.4 10.1 11.9 12.6 12.3 11.5 Percentage of mortgage-backed securities to total assets - - - - - _ Percentage of total real estate assets to total assets 12.4 16.5 19.6 21.7 21.9 20.6 Federally chartered Percentage of number of institutions 66.4 65.6 65.7 66.0 66.2 63.5 Percentage of total assets 64.6 64.9 65.4 65.7 65.6 63.4 Federally insured state-chartered Percentage of number of institutions 33.6 34.4 34.3 34.0 33.8 36.5 Percentage of total assets 35.4 35.1 34.6 34.3 34.4 36.6 Capital to asset ratio categories More than 6% Number 9,719 9,673 9,984 10,496 10,367 10,399 Total assets (US$ billions) 66.9 81.3 101.7 121.4 139.7 168.0 3% to 6% Number 4,156 3,875 3,257 2,370 2,032 2,141 Total assets (US5 billions) 69.9 69.6 64.5 54.2 50.3 50.7 1.5% to 3% Number 575 549 413 294 255 262 Total assets (US5 billions) 8.0 8.1 6.0 4.1 3.7 5.6 0% fo 1.5% Number 175 188 155 130 133 102 Total assets (US$ billions) 1.4 2.6 1.9 2.5 2.5 1.0 Less than 0% Number 62 50 69 81 74 56 Total assets (US5billions) 1.5 0.6 1.2 1.5 2.1 1.8 Number of problem credit unions 794 929 1,022 794 678 685 Assets of problem credit unions (US$ billions) 6.6 8.1 10.6 8.4 9.4 10.4 Resolutions Number 94 88 85 114 164 130 Total shares (USS millions) 116 327 297 285 339 267 Estimated present value cost (US$ millions) 29 52 33 74 49 77 Number of months rated 4 or 5 before dosure 55.4 44.1 30.1 24.0 17.5 - James R. Barth and Robert E. Litan 143 Table 9.3. (Continued) Category 1992 1993 1994 1995 1996 Number of institutions 12,594 12,317 11,991 11,687 11,392 Total assets (US$ billions) 258 277 290 307 327 Capital (US$ billions) 20.9 24.9 27.7 31.7 35.2 Net income (US$ millions) 3,364 3,743 3,438 3,377 3,530 Net operating income (US$ millions) 4,148 4,419 4,181 4,172 4,620 Net non-operating income (US$ millions) 95 76 (59) (20) 15 Provision for loan losses (US$ millions) 879 752 684 775 1,105 Federal inconme taxes (US$ millions) 0 0 0 0 0 Percentage of 1st mortgages to total assets 11.3 11.9 12.9 12.8 14.0 Percentage of mortgage-backed securities to total assets - 3.5 3.6 3.1 2.8 Percentage of total real estate assets to total assets 19.0 18.7 20.0 20.1 21.6 Federally chartered Percentage of number of institutions 62.8 62.5 62.5 62.7 62.8 Percentage of total assets 62.7 62.4 63.1 63.2 63.2 Federally insured state-chartered Percentage of number of institutions 37.2 37.5 37.5 37.3 37.2 Percentage of total assets 37.3 37.6 36.9 36.8 36.8 Capital to asset ratio categories More than 6% Number 10,356 10,901 11,074 11,146 10,990 Total assets (US$ billions) 208.5 250.9 268.1 296.6 321.0 3%to6% Number 1,971 1,279 824 465 321 Totalassets((US$billions) 46.1 24.3 20.1 9.6 5.6 2.5% to 3% Number 178 80 54 37 35 Totalassets(US$billions) 1.5 1.4 0.4 0.4 0.1 0% to 1.5% Number 59 39 24 16 30 Total assets (US$ billions) 1.7 0.5 0.2 0.1 0.1 Less than 0% Number 29 17 14 23 15 Totalassets (US$billions) 0.6 0.2 0.1 0.1 0.03 Number of problem credit unions 608 474 319 267 - Assets of problem credit unions (US$ billions) 7.4 4.3 2.4 2.0 - Resolutions Number 114 71 33 26 19 Total shares (US'S millions) 223 265 255 545 19 Estimated presenit value cost (US$ millions) 107 20 36 13 2 Number of months rated 4 or 5 before dosure - - - - - -Not availab e. Note: Capital includes undivided earnings, regular reserves, and other reserves, but excludes allowances for loan and investn-ent losses, As of 1996 the allowance for investment losses no longer exists. Effective January 1, 1995, the National Credit Union Share Insurance Fund insurance year changed from October 1 through September 30 to January 1 through December 31. Source: Barth and Brumbaugh (1991, 1994); National Credit Union Administration (1995); correspondence with Dr. Tun A. Wai, National Association of Federal Credit Unions. 144 Lessonsfrom Bank Failures in the United States Table 9.4. Selected Performance Measuresfor Commercial Banks and Savings and Loan Institutions, 1950-96 Commercial banks Savings and loan institutions Equity Equity Returni on7 Return on capital to Return oz Return on capital to Year assets (%) equity (%) assets (%) assets (%) eqzuity (%) assets (%) 1950 0.67 9.80 6.75 1.19 16.60 7.18 1951 0.61 9.04 6.71 1.22 16.71 7.30 1952 0.59 8.73 6.73 1.01 13.89 7.25 1953 0.57 8.73 6.93 1.02 14.51 7.05 1954 0.75 8.30 7.11 1.06 15.40 6.87 1955 0.64 10.72 7.16 1.10 16.41 6.73 1956 0.69 9.03 7.04 1.02 15.20 6.74 1957 0.72 9.55 7.70 0.95 13.84 6.85 1958 0.91 11.82 7.65 0.97 14.06 6.92 1959 0.65 8.31 7.89 0.97 14.08 6.92 1960 0.90 11.33 8.05 0.86 12.49 6.92 1961 0.89 11.11 7.97 0.98 14.16 6.93 1962 0.82 10.24 8.02 0.98 14.07 6.95 1963 0.79 9.78 8.08 0.70 10.26 6.81 1964 0.79 10.04 7.72 0.72 10.83 6.69 1965 0.79 10.43 7.53 0.67 9.82 6.82 1966 0.78 10.45 7.42 0.50 7.24 6.88 1967 0.82 11.34 7.09 0.38 5.52 6.88 1968 0.80 11.40 6.89 0.60 8.80 6.85 1969 0.85 12.02 7.18 0.68 9.73 6.99 1970 0.88 12.36 7.12 0.57 8.17 6.98 1971 0.87 12.37 6.95 0.71 10.86 6.53 1972 0.83 12.24 6.62 0.77 12.16 6.36 1973 0.85 12.75 6.68 0.76 12.18 6.22 1974 0.76 12.43 5.70 0.54 8.65 6.21 1975 0.68 11.78 5.90 0.47 8.09 5.81 1976 0.69 11.50 6.11 0.63 11.14 5.68 1977 0.70 11.72 5.92 0.77 13.99 5.70 1978 0.76 12.94 5.77 0.82 14.21 5.77 1979 0.80 13.96 5.73 0.67 11.55 5.80 1980 0.79 13.77 5.75 0.11 2.61 5.47 1981 0.76 13.18 5.82 -0.75 -17.28 4.34 1982 0.71 12.18 5.84 -0.65 -20.44 3.18 1983 0.64 11.11 6.00 0.27 8.11 3.33 1984 0.65 10.52 6.14 0.11 3.97 2.77 1985 0.69 11.12 6.19 0.39 12.30 3.17 1986 0.61 9.92 6.19 0.08 2.40 3.34 1987 0.09 1.55 6.02 -0.59 -21.69 2.72 1988 0.81 13.16 6.28 -0.96 -44.24 2.17 1989 0.48 7.76 6.21 -0.54 -13.17 4.10 1990 0.48 7.55 6.45 -0.35 -6.97 5.02 James R Barth and Robert E. Litan 245 Table 9.4. (Continued) Commercial banks Savings and loan institutions Equity Equity Ret; frn on Return on capital to Return on Return on capital to Year assets (%) equity (%) assets (%) assets (%) equity (%) assets (%) 1991 0.53 7.97 6.75 0.13 2.25 5.94 1992 0,92 12.93 7.51 0.61 9.33 6.93 1993 1.20 15.50 8.01 0.63 8.66 7.50 1994 1.17 14.61 7.78 0.56 7.36 7.48 1995 1.17 14.61 8.11 0.70 9.00 8.01 1996 1.19 14.46 8.20 0.62 7.78 7.92 Source: Federal Deposit Insurance Corporation and Office of Thrift Supervision data. of difficulties, however. Tables 9.1-9.3 show that the major contributors to the total failure resolution costs were failed S&Ls. These institutions alone accounted for 80 percent of the total costs, whether the costs are expressed in nominal dollars or in 1992 dollars. Indeed, when all figures are expressed in 1992 dollars, S&L failure resolution costs from 1980 through 1996 actually exceed the losses borne by all uninsured depositors in both commercial and savings banks and in S&Ls from 1921 through 1933, although they are smaller as a percentage of GDP, and they are only slightly less than these losses plus the total losses borne by both uninsured depositors and the insurance funds for these institutions from 1934 through 1940. (This particular comparison is made because some might consider the later losses to be carried over from the period before insurance was in effect.) The losses associated with banking failures, in other words, were less with- out federal deposit insurance, but with market discipline, than those with fed- eral deposit insurance and regulatory discipline. Even including the losses for the first several years after the implementation of federal deposit insurance to the calculations does not alter the conclusion one draws. The evidence clearly indicates that the regulatory structure for S&Ls that existed during most of the 1980s and early 1990s was inadequate in assuring that the industry would operate safely and soundly. Worse yet, the federal deposit insurance fund for S&Ls itself reported insolvency in 1985 for the first time since its establishment in 1934 and was replaced with a new insurance fund in 1989. As a result of this insolvency, taxpayers were required to help pay for resolving failed S&Ls. 2 2. The enormity of the failure resolution problem led to the establishment of the Resolution Trust Corporation in 1989 to dispose of failed S&Ls and/or their assets. It ceased operations in 1995. Table 9.2 presents information on the conservatorships the Resolution Trust Corporation handled during its existence. 146 Lessonsfrom Bank Failures in the United States Figure 9.1. Return on Assets and Equity and Equity Capital to Assets, 1950-96 0 -0.5 - Commercial banks - Savings and loan institutions 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 20 - I-1 ~--20- - - Commercial banks -40 - Savings and loan institutions 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 9 6- 3 - ~ ~ Commercial banks 2 - Savings and loan 1 institutions 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 Source: Table 9.4. James R. Barth and Robert E. Litan 147 By contrast, the regulatory structure for credit unions worked much better during the same period. By recapitalizing their own federal deposit insurance fund, credit unions themselves provided more than enough resources to cover all failure resolution costs. Note that unlike in the case of S&Ls and commer- cial and savings banks, there was no widespread concern being expressed in the 1980s that taxpayer funds would be needed to resolve failed credit unions. 3 Second, despite the serious problems federally insured depositories ex- perienced during the 1980s and the early 1990s, they were nonetheless tem- porary arid, most important, did not significantly disrupt aggregate eco- nomic activity. I The temporary nature of the problems is evident in figure 9.1, whichi shows the rate of return on assets, rate of return on equity, and the equity capital-to-asset ratios for commercial banks and S&Ls from 1950 through 1996. Figure 9.1 and table 9.4 also show that S&Ls experienced far worse problems than commercial banks based upon these three widely used measures of performance. Furthermore, figure 9.1 clearly shows two distinct timnes when S&Ls suffered severely. The first period was when short- term interest rates rose relative to long-term interest rates in the late 1970s and the early 1980s. The second period was when asset quality deterio- rated significantly in the latter half of the 1980s. A substantial literature documents that the problems experienced by S&Ls were due to a combination of factors, including both sharp declines in energy prices and adverse tax rate changes in 1986.5 Most observers agree that one of the most important contributors to the S&Ls, problems were laws and regulations that exposed them to substantial interest rate risk in the late 1970s and early 1980s. Subsequently, lax regulation and su- pervision, especially in the early to mid-1980s, enabled inadequately capi- talized instil:utions to grow rapidly by engaging in high-risk activities. As 3. The insurance fund for commercial and savings banks reported insolvency in 1990 and 1991, but then returned to solvency as the financial condition of these institu- tions significarttly, albeit unexpectedly, improved. 4. In this sense, federal deposit insurance accomplished its goal of preventing wide- spread runs and the associated disruption in the credit system and payments mecha- nism. Disruptions in the latter, of course, also disrupt aggregate economic activity. The failures of federal deposit insurance during this time were that taxpayers were not adequately protected because of excessively costly failure resolutions and that eco- nomic resources were misallocated. 5. With respect to tax rate changes, the Economic Recovery and Tax Act of 1981 short- ened depreciable life from 40 to 15 years for real property and increased the depreciation rate from 125 to 175 percent declining balance for real property. The Tax Reform Act of 1986, by contrast, lengthened depreciable life from 19 to 27.5 years for residential property and to 31.5 years for commercial property. Also, the depreciation rate was decreased from a 175 percent declining balance to straight-line for real property, and a passive loss rule was added that prevented the direct offset of most real estate losses against ordinary income. 148 Lessonsfrom Bank Failures in the United States table 9.2 shows, even though several hundred institutions holding about US$200 billion or more in assets were reporting insolvency or were known by the regulatory authorities to be problem institutions during 1982 through 1986, total industry assets grew by 70 percent, or US$478 billion. Although the banking problems were severe, tables 9.1-9.3 and figure 9.1 show that they were clearly over by the mid-1990s. Indeed, at that time the regulatory authorities of all three types of federally insured deposi- tory institutions were issuing reports on the overall healthy condition of institutions. This situation reflected a combination of largely unanticipated factors, including a favorable interest rate environment, widespread and noninflationary economic growth, and a significant shift away from tradi- tional banking activities as institutions essentially reinvented themselves through their own efforts and the efforts of both the regulatory authorities and Congress. Gilbert (1997, p. 1), an economist with the Federal Reserve Bank of St. Louis, reports that: "Almost all of the rise in profitability of the banking industry from 1992 to 1996 can be attributed to the relatively low interest rates paid on savings and small time deposits." Third, several important changes have occurred in the structure of the federally insured depository industry. Since 1980, the number of all types of depository institutions has fallen substantially. Indeed, the total num- ber of institutions has declined by 36 percent. This reduction has been due not only to failures, but also to consolidation through mergers and acqui- sitions brought about by increased domestic and global competition, im- proved information technology, and eased restrictions on both geographi- cal expansion and intradepository industry entry. At the same time, as table 9.5 and figure 9.2 show, the depository in- stitutions' share of total financial assets held by all U.S. financial service firms declined from 65 percent in 1950 to 30 percent in 1996. During the same period the share accounted for by only U.S.-chartered commercial banks declined from 50 to 17 percent. The U.S.-chartered commercial banks' share today is nearly the same as the share accounted for by money market and other mutual funds. This latter development emphasizes the declining role of banks in the traditional intermediation function as the U.S. financial system has evolved over time. Similarly, the role of S&Ls has also diminished in terms of its traditional intermediation function, but in a more fundamental way. As table 9.6 shows, as recently as 1980 S&Ls were the major holder of home mortgages. The securitization of such mortgage loans, however, has contributed to the decline in the traditional role of S&Ls as major holders of home mortgages. Indeed, the share of home mortgages ac- counted for by all savings institutions declined from 50 percent in 1980 to James R. Barth and Robert E. Litan 149 Table 9.5. Percentage Distribution of Total Financial Assets Held by All U.S. Financial Service Firms, Selected Years, 1950-96 Institutioll 1950 1960 1970 1980 1990 1995 1996 Depository Institutions I Commercial banks 50.9 37.8 38.2 34.1 27.5 24.4 23.0 U.S.-chartered 50.3 37.1 36.1 29.1 21.8 18.0 16.8 Foreign offices in the U.S. 0.4 0.6 0.7 2.3 3.0 3.6 3.5 Bank holding companies - - 1.1 2.4 2.5 2.5 2.5 Banks in the U.S. possessions 0.3 0.1 0.3 0.3 0.2 0.2 0.2 Savings institutions 13.7 19.6 19.7 18.2 11.2 5.5 5.1 S&Ls 5.9 11.8 12.8 14.4 9.1 _ e _ Savings banks 7.7 6.8 5.9 3.8 2.1 _e Credit unions 0.3 1.0 1.3 1.6 1.8 1.7 1.6 Contractual intermediaries Life insurance companies 21.3 19.2 14.8 10.7 11.3 11.4 10.9 Other insurance companies 4.0 4.3 3.7 4.2 4.4 4.0 3.9 Private pension funds b 2.4 6.3 8.3 11.6 13.4 14.3 14.8 State and local government retirement funds 1.7 3.3 4.4 4.5 6.8 8.3 8.5 Others Finance companies 3.2 4.6 4.7 4.7 5.0 4.5 4.4 Mortgage companies - - - 0.4 0.4 0.2 0.2 Mutual funds c 1.1 2.8 3.5 1.4 5.0 10.1 11.5 Money market mutual funds 0.0 0.0 0.0 1.8 4.1 4.0 4.4 Closed-end funds - - - 0.2 0.4 0.7 0.7 Security brokers and dealers 1.4 1.1 1.2 1.0 2.2 3.1 3.1 Real estate investment trusts - - 0.3 0.1 0.1 0.1 0.2 Issuers of asset-backed securities - - - 0.0 2.3 3.7 4.0 Bank personal irusts d -. 5.6 4.3 4.0 3.7 Total assets (lJS$ billions) 294 605 1,356 4,349 12,152 18,414 20,482 - Not available. a. Commercial banks consist of U.S.-chartered commercial banks, domestic affiliates, Edge Act corporations, and agencies and offices in U.S. possessions. Foreign banking offices in the United States include Edge Act corporations and offices of foreign banks. International banking facilities are excluded from domestic banking and treated like branches in foreign countries. Savings and loan associations include all savings and loan associations and federal savings banks insured by the Savings Association Insurance Fund. Savings banks include all federal and rmutual savings banks insured by the Bank Insurance Fund. b. Private pension funds include the Federal Employees' Retirement Thrift Savings Fund. c. Mutual funds are open-end investment companies (including unit investment trusts) that report to the Investment Company Institute. d. Bank persanal trusts are assets of individuals managed by bank trust departments and nondeposit, noninsured trust companies. e. The flow of funds accounts were restructured in the second quarter of 1993, thereby omitting this breakdown. Source: Board of Governors of the Federal Reserve System, flow of funds accounts. Figure 9.2. Percentage Distribution of Total Financial Assets Held by All U.S. Financial Service Firms, 1950-96 100 - 90 - 80 - All other 70- 60 - Private pension and state and local >~~~~~~~~~~~~~~~~~~~~~~~gvrmn reti =rement funds 60- 4 50 -\ Other depository institutions 40 -Mta ud 30- 20- 10 - U.S.-chartered commercial banks OI I I I .- _ I.--l - 1950 1960 1970 1980 1990 1995 1996 Source: Table 9.5. Table 9.6. Percentage Distribution of U.S. Real Estate Mortgage Loans by Lender. Selected 'ears, 1950-96 1950 1960 1970 Lender HM MM CM HM MM CM HM MM CM Commercial banks 21.0 10.2 18.1 13.6 5.4 20.4 14.4 5.5 27.2 Savings institutions 38.5 32.0 11.8 53.5 28.5 15.2 56.2 36.9 22.6 S&Ls 29.0 2.5 2.5 39.0 10.5 7.5 41.9 22.9 13.4 Savings banks 9.5 29.5 9.3 14.5 18.0 7.7 14.3 13.0 9.2 Credit unions 1.0 0.0 0.0 0.3 0.0 0.0 0.3 0.0 0.0 Life insurance companies 18.8 28.1 29.4 17.5 18.6 30.1 9.1 26.6 30.4 Private pension funds 0.1 0.3 0.2 0.4 1.7 1.0 6.0 2.0 1.4 State and local government retirements ftnds 0.1 0.3 0.1 0.6 2.2 0.5 1.0 3.3 1.1 Finance companies 0.8 0.4 0.1 1.0 0.8 0.2 2.0 2.0 0.4 Real estate investment trusts 0.0 0.0 0.0 0.0 0.0 0.0 0.2 2.1 2.4 Mortgage pools 0.0 0.0 0.0 0.0 0.0 0.0 1.0 0.1 0.0 Government-sponsored enterprises 0.0 0.0 0.0 2.0 0.0 0.0 5.3 0.5 0.0 Issuers of asset-backed securities 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 U.S. government 3.3 0.4 0.1 3.0 4.9 0.1 2.1 5.1 0.5 State and local governments 0.5 0.0 0.0 1.0 0.1 0.0 0.6 3.3 0.2 lHouseholds 16.8 28.2 39.2 7.2 37.9 32.1 7.6 13.2 13.7 Total assets (US$ billions) 45.0 9.0 13.0 142.0 21.0 33.0 294.0 60.0 86.0 (Table continues on thefollowing page.) Table 9.6. (Continued) 1980 1990 1996 Lender HM MM CM HM MM CM HM MM CM Commercial banks 16.8 9.1 31.6 17.5 12.0 44.5 18.4 15.8 52.7 Savings institutions 50.1 38.1 24.1 24.0 29.7 14.4 13.5 20.8 7.5 S &Ls 43.0 26.8 17.7 - - - - - - Savings banks 7.1 11.3 6.4 - - - - - - Credit unions 0.5 0.0 0.0 1.9 - - 1.9 - - Life insurance companies 1.9 10.0 31.6 0.5 9.4 28.4 0.2 8.1 24.3 Private pension funds 0.1 0.3 0.7 0.3 1.3 1.5 0.1 0.3 2.0 State and local governmnent retirements funds 0.4 2.7 1.4 0.1 1.6 1.0 0.1 1.6 1.0 Finance companies 2.6 1.5 1.2 2.5 - - 2.6 - Real estate investment trusts 0.0 1.1 0.9 0.0 0.7 0.7 0.1 3.8 0.6 Mortgage pools 11.2 4.2 0.0 37.9 9.5 - 44.0 11.0 - Government-sponsored enterprises 6.1 4.7 0.0 4.4 4.3 - 5.6 6.5 - Issuers of asset-backed securities 0.0 0.0 0.0 2.3 1.6 0.3 6.5 11.4 9.0 U.S. government l.9 7.3 2.2 1.4 7.4 1.8 0.4 4.9 1.1 State and local governments 2.1 7.5 0.7 2.3 13.2 0.9 1.6 15.3 1.1 H ouseholds 6.3 10.1 2.9 4.6 6.9 3.2 5.1 0.6 0.9 Total assets (US$ billions) 955.0 142.0 256.0 2,616.0 309.0 758.0 3,818.0 297.0 706.0 - Not available. HM Hlome mortgages. MM Multifamnily mortgages. CM Commercial mortgagees. Source: Board of Governors of the Federal Reserve System, flow of funds accounts. James R. Barth and Robert E. Litan 153 14 percent in 1996. At the same time, as tables 9.1 and 9.3 show, real estate assets have increased in importance for commercial banks and credit unions. These and other recent developments are continuing to blur many of the distinctions among the different types of financial service firms as they all attempt to adapt to the changing, competitive, domestic, and in- creasing.ly global, financial marketplace. 6 Lessons from U.S. Banking Problems As the experiences of the 1980s and early 1990s demonstrate, despite the existence of an elaborate regulatory structure, significantbankingproblems can develop. In trying to understand how to design bank regulation to pre- vent future problems, deriving some lessons from actual problems of the past is helpful. The lessons discussed here are based on U.S. banking prob- lems, and therefore may not be applicable to other countries at different stages of development and with different cultures and financial systems. Banks' Participation in a Wider Range of Activities The ongoing debate over whether to allow banks to engage in a wider range of activities highlights the fact that regulation has significantly limited fed- erally insured depositories in what they can do and when they can do it. This necessarily means that depository institutions have been unable to adapt freely to changing, competitive financial markets. The situation of the S&Ls in the late 1970s and early 1980s represents the most extreme case in which institutions were unable to adapt to a changing environment in a timely manner. As table 9.7 shows, despite repeated attempts to broaden their range of permissible activities, prior to the early 1980s S&Ls could essentially offer only long-term, fixed rate home mortgages. Only the threat of their very survival prompted Congress to grant S&Ls, albeit too late for many of them, greater freedom to reduce their interest rate risk exposure. This was largely done by permitting institutions to diversify away from home mortgages, authorizing the use of various derivative instruments, and allowing variable rate home mortgage loans. 6. A commercial bank is legally defined to be an institution that accepts demand deposits, makes commercial loans, and whose deposits are insured by the Bank Insur- ance Fund of the Federal Deposit Insurance Corporation. Yet today only about 15 per- cent of the assets of commercial banks are commercial loans and only about 15 percent of their assets are funded with demand deposits. In other words, by 1997 legal rather than market factors were needed to distinguish a commercial bank from manv other types of firms in the financial marketplace. 154 Lessonsfrom Bank Failures in the United States Table 9.7. Evolution of Federal S&L Powers, 1933-82 Enacted Enacted Enacted Enacted Enacted Proposed Admini- stration's Home Federal Oumer's Housing Houising Public Law Public Law Savings Loan Act, Act, Act, 87-779, 88-560, Bank Bill, Power 1933 1959 1961 1962 1964 1965' Home mortgage Yes Yes ($20,000 ($35,000 or loan limit) 2% of assets)' Nonresidential Yes Yes Yes mortgages (15% of (20% of assets) assets) Land development Yes Yes (Loans up (Loans) to 5% of assets) Direct investment Yes Yes Yes (5% of assets (2% of assets (Up to in bus. dev. in urban 50%. of Corp.) renew. area) capital) Service corporation Yes Yes (1% of (No limit) assets) Construction loans Corporate bonds Yes Education loans Yes (Up to 5% of assets) Consumer loans Yes (Up to US$5000) Commercial loans Leasing of personal property Investment in commercial paper Credit cards James R. Barth and Robert E. Litan 255 Table 9.7. (Continued) Proposed Proposed Proposed Proposed Enacted Admini- stration's Hunt proposed 'Fine study" House Commission "Financial legislation Banking Public Law Study, Institution passed Senate Committee 95-630, Power 1972 Act," 1973' 19754 print, 1976' 1978 Home mortgage Yes Yes Yes Yes (Statutory (Statutory (Statutory ($60,000 loan limit loan limit loan limit loan dele.) dele.) dele.) limit) Non-residential Yes Yes Yes Yes Yes mortgages (30% of (20% of (20% of assets) assets) assets) Land development Yes Yes Yes (Loans) (5% of assets) Direct investment Yes Yes Yes Yes Yes (Up to 3% (For (For (For (2% of of assets) community community community assets for development) development) development) community development) Service corporation Yes Yes Yes (1% of (1% of (1% of assets) assets) assets) Construction loans Yes Yes Yes Yes Yes (20% of (5% of assets) assets) Corporate bonds Yes Yes Yes Yes (Up to 10% of assets) Education loans Yes Yes Yes Yes (Up to 3% (30% of (20% of (5% of of assets) assets) assets) assets) Consumer loans Yes Yes Yes Yes (Up to 10% (Up to 10% (30% of (20% of of assets) of assets) assets) assets) Commercial loans Yes (Up to 3% of assets) Leasing of personzl property Investment in Yes Yes Yes Yes commercial paper (Up to 3% (Up to 10% (30% of of assets) of assets) assets) Credit cards Yes Yes Yes (Table continues on thefollowing page.) 156 Lessonsfrom Bank Failures in the United States Table 9.7. (Continued) Proposed Enacted Proposed Enacted Admini- Depository Admini- stration Institutions stration Garn- proposal, Deregulation proposal, St. Germain, Year first Power 1979f Act, 1980 1981' 1982 authorized Home mortgage Yes Yes Yes 1933 (90% loan-to- (Statutorv (Statutory value limit) loan-to- loan-to- value limnit value limit dele.) dele.) Non-residential Yes Yes Yes 1933 mortgages (20% of (No limit) (40% of assets) assets) Land development 1959 Direct investment Yes Yes Yes 1961 (2% of (In small (1% assets in assets for business small business community investment investment development.; corporation) corporation) 5% business development) Service corporation Yes Yes 1964 (3% of (5% of assets) assets) Construction loans Yes 1978 (5% of assets) Corporate bonds Yes Yes Yes Yes k 1980 (10% of (20% of (No limit) assets) assetsi) Education loans Yes Yes 1978 (5% of (No limit) assets) Consumer loans Yes Yes Yes Yes 1980 (10% of (20% of (No limit) (30%. of assets) assets) assets) Commercial loans Yes Yes' 1982 (No limit) (10% of assets) Leasing of personal Yes Yes 1982 property (10% of (10% of assets) assets) Investmnent in Yes Yes 1980 commercial paper (109% of (20% of assets) assets) Credit cards Yes Yes 1980 James R. Barth and Robert E. Litan 157 Table 9.7. (Continued) Note: Enacted legislation in bold, everything else is proposed legislation. a. H.:R. 14 and H.R. 11508, 89th Congress b. Report, Presidential Conimittee on Financial Structure and Regulation (1972) c. S. 2591, 93rd Congress (1973) d. S. ].267,94th Congress (1975) e. Financial Reform Act of 1976, 94th Congress (1976) f. Depository Institutions Deregulation Act, 96th Congress (1979) g. S. 1703, 97th Conigress (1981) h. The bill also provided an 80% loan-to-value limit that could be waived by regulation. i. The limit was raised to $75,000 in 1979, Public Law 96-161 j. By regulation, up to 1% unrated bonds k. While the Gam-St. Germain Act continues the placement of corporate bond investments in the 20% of assets basket, the Bank Board interpreted the Act to permit up to 100% of assets in corporate bonds, with up to 1% in unrated bonds. 1. By regulation, the BankBoard interpreted this authority to include investmentin unrated corporate bonds up to 10% of assets. Source: Barth (1991). Figure 9.3 presents a broader perspective on bank regulation by pro- viding information on some of the more important developments that affected all depository institutions from 1781 through 1996. Based on stud- ies of various aspects of this historical record, one learns that most bank regulation has not been proactive, but rather a reaction to actual or per- ceived banking problems. Yet the attempts to resolve the identified prob- lems all too often created new and potentially even more serious prob- lems. The reason for this is that changes in the regulatory structure not only change the opportunities for banking institutions to engage in what they consider to be the most profitable activities, but also change their incentives with respect to risk-taking behavior. Their proponents view changes in the range of opportunities available to institutions to pursue various activities as necessary to achieve a safe and sound banking in- dustry. Viewed in a static context, such changes may appear to be able to achieve their goal, but financial markets must be viewed in a dynamic context. Financial markets are subject to changes that the regulatory au- thorities cannot control, or even anticipate. In such a situation changes in the regulatory structure may breed new and unanticipated problems. This was the case when S&Ls were first required to specialize in fixed rate home mortgages and then encouraged to diversify into new activities, many of which they were allowed to do without either sufficient exper- tise or adequate equity capital contributed by owners. In view of this situation, there is considerable merit to Federal Reserve Bank of Cleveland President Jerry Jordan's (1996) view that: "Banking com- panies should not be required to get permission from regulators before doing 158 Lessonsfrom Bank Failures in the United States Figure 9.3. Timeline of Banking Developmentts in the United States U.S. Constitution - Congress has the power "to coin money and regulate the value thereof." (1787) First mutual savings bank First commercial (1816) bank (1781) First S&L (1831) 1780 1800 1820 1840 1860 1880 Wildcat banking Natonal (1836-1863) Bank Art (1864) First Bank Second Bank of of the U.S. the U.S. (1791-1811) (1816-1836) National Currency Act (1863) - Fiscal agent for U.S. Treasury - Loans to state banks with temporary liquidity problems Limit state bank note issuance - Part private bank and part central bank Office of the Comptroler of the Currency - Federally chartered banks - Uniform currency - Tax on state bank notes James R. Barth and Robert E, Litan 159 Figure 9.3. (Continued) Federal Credit Union Act (1970) Banking provides Intemational Stock market federal deposit Act (1978) crash insurance for puts foreign banks on (1929) credit unions equal footing with McFadden Act (1927), Savigs and Loan Holding U.S. banks whereby national banks Company Act (1968) Depository Institutions could branch to same permits unitary S&L Deregulation and extent permitted state holding companies Monetary Control Act banks to engage in any activity, (1980) phases out even those unrelated deposit rate ceilings by to S&L business April 1, 1986; allows negotiable order of withdrawl Federal Resei/v Baaccounts at all depository Federal Reserve Bank Merger acts / institutions; allows S&Ls to Act (1913) (1960,1966) establish make consumer loans and furnishes "elastic merger guidelines and issue credit cards currency" denote competition as a criterion Competitive Equality lst credit unicn Banking Act (1987) (1909) \ limits growth of \\ l nonbank banks 1900 1920 1940 1960 1980 2000 Riegle-Neil | Interstate Banking .It and Branching Bank Holding Efficiency Act Company acts (1994) whereby (1956, 1970) bank holding restricted interstate companies can ownership of banks; bank acnuire banks Great Depression holding companies could nationwide after = Securities & Exchange engage in activities Sept. 29,1995, Commission deemed by the Federal and branch - Federal Deposit Insurance Reserve to be "closely nationwide after for commercial banks and related to banking"; June 1, 1997, S&Ls defined a bank unless state opts - Federally chartered out S&Ls and credit unions - Banking Act of 1933 (Glass-Steagall) separates Federal Deposit commercial banking from insurance investment banking Garn-St. Germain Depository Corporation - Federal Home Loan Bank Institutions Act (1982) Improvement Act System - Allows possibility of interstate (1991) mandates and interinstitutional mergers prompt corrective - Gives S&Ls authority to action" make some commercial loans Financial Institutions Reform, Recovery, and Enforcement Act (1989) changes structure of S&L institution regulation; replaces Federal Home Loan Bank Board with Office of Thrift Supervision; replaces Federal Savings and Loan Insurance Corporation with Savings Association Insurance Fund Note: Before the Civil War the federal government only issued coins, and only these were legal or lawful mnoney. During the Civil War "greenbacks" were added, and finally in 1933 federal reserve notes were added. 160 Lessons from Bank Failures in the United States something new. Rather, they should notify authorities of their intentions. If regulators want to prevent the action, the burden should be on them to inter- vene in a timely way to demonstrate that the costs exceed the benefits." There is also considerable merit to Federal Reserve Bank of Kansas City President Thomas Hoenig's (1996, p. 11) view that: "In light of the costs and difficul- ties of implementing prudential supervision for larger institutions who are increasingly involved in new activities and industries, the time may have come to sever the link between these institutions and the safety nets, mak- ing it feasible to significantly scale back regulatory oversight of their op- erations." Of course, if banks were permitted unrestricted access to new activities, the authorities would want to be sure that they conducted such activities under appropriate conditions. In this regard, the authorities could require that new activities, whether conducted in a subsidiary of a bank or in a holding company, be capitalized by funds other than those used to meet the bank's required capital standards. The authorities could also im- pose interaffiliate lending restrictions on banks and any new nonbank af- filiates. Moreover, they could prohibit nonbank affiliates that are creditors from reaching banks' assets by "piercing the corporate veil." In short, any expansion by banks into new activities should be accompanied by prudent limitations on the overall way in which such activities are conducted. This qualification also applies to the design of an appropriate bank regulatory structure for the emerging market economies. Federal Safety Net Many individuals believe that banking institutions should be restricted to a fairly narrow range of activities because they have access to the fed- eral safety net, that is, deposit insurance, discount window borrowing, and the Fedwire payments system. Indeed, much of the debate about whether or not to ease, if not eliminate, the restrictions separating banks and nonbank firms relates to the safety net. A specific concern is that any subsidy associated with the safety net could flow from banks to any af- filiated nonbank firms. Yet disagreement exists about whether any such subsidy even exists. Federal Reserve Chairman Alan Greenspan has tes- tified that a subsidy does exist (Greenspan 1997); Comptroller of the Currency Eugene Ludwig has testified that "no net [taking into account the cost of the regulatory burden imposed on banks] subsidy exists" (1997, p. 2); and Federal Deposit Insurance Corporation Chairman Ricki Helfer has testified that "if a net subsidy exists, it is very small" (1997, p. 2). Outside the bank regulatory agencies, the Shadow Financial Regulatory James R. Barth and Robert E. Litan 161 Committee has concluded that the net subsidy is "probably not particu- larly large" (1997, p. 2). Obviously some effort should be made to measure the net subsidy, and if a net subsidy is found to exist, it should be eliminated in an effi- cient and timely manner. Once eliminated, the danger of the subsidy spreading to other affiliates is also eliminated. The more important point, however, is that the mere existence of a subsidy should not be used to deny banks the opportunity to engage prudently in a wider range of ac- tivities, and correspondingly, for other firms to own banks. As table 9.8 shows, the United States is clearly out of step with almost all the other 19 nonoverlapping Group of Ten and European Union countries with re- spect to the extent to which banks are permitted to engage in securities, insurance, and real estate activities. The United States is also out of step with respect to permitting banks to own nonfinancial firms and vice versa (see Barth, Nolle, and Rice 1997 for further international comparisons regarding these and related banking issues). Adverse Selection and Moral Hazard Various types of adverse selection, principal-agent, and moral hazard prob- lems arise in banking. It is therefore incumbent upon the regulatory au- thorities to examine, supervise, and regulate federally insured depositories to promote a stable, efficient, and competitive banking industry. The au- thorities must also, of course, resolve troubled institutions in a timely and cost-effective manner so as to limit losses to the insurance funds and thereby better protect taxpayers and minimize the misallocation of resources. As the S&L situation so vividly demonstrates, regulatory forbearance can ex- acerbate an existing problem. In the early 1980s, when virtually the entire S&L industi y was deeply troubled, the authorities lowered regulatory capital requirements and broadened the measurement of capital for meeting the requirements to include items unacceptable under generally accepted ac- counting principles. As figure 9.4 shows, these actions were taken by the regulatory authorities (the Federal Home Loan Bank Board at that time) after the U.S. Congress first lowered the minimum statutory capital require- ment and then eliminated it entirely. As a result of these and other regula- tory actions, many institutions engaged in high-risk activities with little, if any, equity contributed by owners at stake. As table 9.2 shows, S&Ls were publicly reporting tangible insolvency for several years before being re- solved. Indeed, the institutions resolved in 1988 had been reporting tan- gible insolvency on average for three-and-a-half years prior to resolution. Table 9.8. Permissible Banking Activities and Bank Ownership in the European Union and Group of Ten Countries, 1995 Commercial bank Nonfinancialfirm Power and investmenit in investmiient in country Securities Insurance Real estate nonfinancialfirm commercial banks Very wide powers Austria Unrestricted Permitted Unrestricted Unrestricted Unrestricted France Unrestricted Permitted Permitted Unrestricted Unrestricted Netherlands Unrestricted Permitted Permitted Unrestricted Unrestricted Switzerland Unrestricted Permitted Unrestricted Unrestricted Unrestricted United Kingdom Unrestricted Permitted Unrestricted Unrestricted Unrestricted Wide powers Denmark Unrestricted Permitted Permitted Permitted Unrestricted Finland Unrestricted Restricted Permitted Unrestricted Unrestricted Germany Unrestricted Restricted Permitted Unrestricted Unrestricted Ireland Unrestricted Prohibited Unrestricted Unrestricted Unrestricted Luxembourg Unrestricted Permitted Unrestricted Unrestricted Restricted Portugal Unrestricted Permitted Restricted Permitted Unrestricted Spain Unrestricted Permitted Restricted Unrestricted Permitted Somewhat restricted powers Belgium Permitted Permitted Restricted Restricted Unrestricted Canada Permitted Permitted Permitted Restricted Restricted Greece Permitted Restricted Restricted Unrestricted Unrestricted Italy Unrestricted Permitted Restricted Restricted Restricted Sweden Unrestricted Permitted Restricted Restricted Restricted Table 9.8. (Continued) Commercial bank Nonfinancial firm Power and Investment in investment in country Securities Insurance Real estate nonfinancialfirm commercial banks Restricted powers Japan Restricted Prohibited Restricted Restricted Restricted > United States Restricted Restricted Restricted Restricted Restricted Unrestricted A full range of activities in the given category can be conducted directly in the bank. Permitted A full range of activities can be conducted, but all or some must be conducted in subsidiaries. Restricted Less than a full range of activities can be conducted in the bank or subsidiaries. Prohibited The activity cannot be conducted in either the bank or subsidiaries. Note: Securities activities include underwriting, dealing, and brokering all kinds of securities and all aspects of the mutual fund business. Insurance activities include underwriting and selling insurance products and services as principal and as agent. Real estate activities include investment, development, and management. Source: Barth, Nolle, and Rice (1997). 264 Lessonsfrom Bank Failhres in the United States Figure 9.4. Federal Savings and Loan Insurance Corporation Insured Institu- tions, Capital to Asset Ratios, 1940-88 Regulatory accounting practices 10l ................ Generally accepted accounting practices - - - - Tangible capital 9 - 1956: In addition to the statutory capital requirement, the FHLBB (Federal Home Loan Bank Board) introduced a regulatory 1982: Ga-S. Germain Act 8 - capital requirement based on the financial eliminated 3-6% statutory attributes of an institution. r within which the 7 / FHLss set requirements. 7- ,_\ hrequirements to 3% from 4%. 6-\ 1972: The FHLBB 6 replaced previous \ ~~~~~~capital requirements 5 From time of of with a capital establishmentthe equireent that was FHLBB in 1934, the te greaer f 5% of 1980 Monetary 4- tory20% of scheduledl replaced 5% reuiremenit wasI 5% ofisue tems) or adjusted statutory I 3 deposits. capital net worth index. =equirementrwith 3- ~~~~~~~~~~~~~~3-6 sautr range. The F-LBB set a 4°/% 2- requirement. 1985. The FHiLBB replaced the separate statutory reserve and regulatorv capital requirements with I ~~~~~~~a single revised capital requirement.s 0 I 1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990 Source: Barth (1991). The enormous failure resolution costs discussed earlier reflect the conse- quences of allowing insolvent institutions to remain open for lengthy peri- ods. Figure 9.5 shows the relationship between regulatory delay and fail- ure costs for S&Ls during the 1980s. Many believe that the enactment of the Federal Deposit Insurance Cor- poration Improvement Act in 1991 eliminates the possibility of any simi- lar regulatory forbearance in the future. While the act certainly has its desirable features, one should not be overly optimistic that it will al- ways work as intended. The reasons for some healthy skepticism are twofold. First, when adverse movements in interest rates in the late 1970s and early 1980s devastated the S&Ls, existing statutory and regulatory capital standards were deemed to be too stringent, and were therefore simply eased, with the effect of papering over the problem. Second, as table 9.2 shows, sufficient information was publicly available that docu- mented the severity of the problems in the S&L industry throughout the 1980s, and yet decisive action to resolve the situation once and for all was not taken until the end of the decade. This suggests that even James R. Barth and Robert E. Litan 165 Figure 9.5. Failed S&Ls: Regulatory Delay and Failure Costs, 1980-92 60 45 50 Failure costs as a percentage 3 of failure assets 40 3 // ~~~~~~~20. 20 -/ 15 Average number of months tangible _. ,insolvent before closure 0 . . . , . . . - . . - 0 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 statutorily mandated regulatory discipline may be less than a perfect substitute for market discipline. Attempts should therefore be made to rely as much as possible on stable market discipline and less on regula- tory discipline, not vice versa, to prevent future banking problems (see Calomiris 1989, 1992, 1997; Kane 1989a, 1992).7 The Perspective of Policy Decisions All too often policy decisions about bank regulation seem to be made from a relatively narrow perspective. To demonstrate this point, figure 9.6 shows that funds from savers do not flow to investors only through banks. In- stead, funds may flow from savers to investors through money and capital markets and through a variety of financial intermediaries. Given the im- portance of investment for long-term economic growth, and hence for im- proved living standards, the flow of funds from savers to investors must not be disrupted. Yet, as indicated in the center of the figure, problems arise that may indeed impede the flow of funds from savers to investors. As also indicated, however, private and public methods exist that are used to try to resolve problems that impede the flow of funds. The important point that nonetheless remains is that disruptions in the credit system and payment mechanism, or more generally, in the entire financial system, can 7. Note tha-t Argentina, for instance, has recently implemented a subordinated debt requirement for its banking system. Figure 9.6. Designing a Financial System Returns and risk Retums and risk Money and Capital Markets Funds (formnal and informal credit systems) Ftinds / Transaction costs Privat\ 8 r / * ~~~~~~~~Property rights - Contas\ Information/accounting system - Technoloe / * ~~~~~~~~~~~~~~~~Inappropriate govermmenitpolicies * Info r mto disclosure\ > Savers / Lenders Customs * Takeovers Borrowers / Investors (o sehodsefrms, aLndgoersnment)m (households, firzm, andl government) Marketfailure6 Plwblic (households, firms, and govemment) Informational asyrnmetries Legal system * Moral hazard *Polit(cal system RetuX s ad rk pPpal ag ynt conflicts m a Enforcentpoici R *Conflicts of interest *Subsidies adtaxe / \ *Barriers to entry and exit/ \ *Public goods/ Funds F, inancial Firrns Funds . ~~~~~~~~~~~(credit system and Returns and risk payme.nt mechanism) Returns and risk James R. Barth and Robert E. Litan 167 interfere with the smooth and efficient flow of funds from savers to inves- tors. This in turn can adversely affect economic growth. Based on a broad perspective, as illustrated by figure 9.6, designing appropriate bank regulation should be viewed as part of the process of designing an appropriate overall financial system (see, for example, Barth and Brumbaugh 1997; Herring and Litan 1995; Kaufman and Kroszner 1996). As the different entities identified in the figure are interrelated, the authorities should not focus exclusively on any one entity or subset of entities, such as depository institutions, when designing regulatory struc- tures. Furthermore, this figure can be viewed from either a domestic or a global perspective. In any event, by focusing too narrowly on just banks, for example, the authorities might consider certain regulations as appro- priate that from a broader perspective would be inappropriate. A few examples might be useful to help make this point clearer. First, as figure 9.3 shows, prior to 1956, the mixing of commercial bank- ing and commerce was permitted through holding companies. In that year and subsequently in 1970, however, legislation was enacted that permittecd commercial banks only to be affiliated with nonbanking firms that were "closely related to banking." Thus the mixing of commercial banking and commerce was effectively terminated. Yet in 1968 Congress enacted legislation that permits a holding company that owns a single S&L to engage in any activity, even activities unrelated to the S&L busi- ness. As a result of this freedom and other important differences (such as unrestricted nationwide branching and an eased qualified thrift lender test in 1996), the value of an S&L charter has been enhanced relative to a commercia l bank charter; however, the relative enhancement has resulted from legislative and regulatory actions, not market forces. By focusing too narrowly on one particular type of depository institution, in other words, policymakers can enact legislation that unintentionally alters the competitive landscape in significant ways. Second, when the authorities recognize that many different types of financial firms exist, they naturally must ask the question: what is a bank? Legally, a U.S. bank is defined to be a firm that makes commercial loans, accepts demiand deposits, and whose deposits are federally insured by the Bank Insurance Fund of the Federal Deposit Insurance Corporation. Yet, today, as noted earlier, commercial loans and demand deposits each only amount to about 15 percent of the total assets of commercial banks. What was once a traditional bank no longer exists. Banks have been rein- venting themselves in recent years to remain viable in a changing finan- cial marketplace. They now compete with a variety of other less-regu- lated financial and nonfinancial firms, as well as the money and capital 168 Lessons from Bank Failures in the United States markets, by increasingly offering more services that generate fee income and by relying less on net interest income. Who would have thought only a few years ago that an automobile firm and an electric company would be direct competitors of banks. Today, how- ever, more than half of Ford Motor Company's net income comes from its financial services operations, and General Electric Company earns about 40 percent of its net income from its capital services operations. In view of this situation, the authorities must broaden their focus beyond the legal definition of a bank to encompass the functions banks actually perform when designing appropriate bank regulation. As the saying goes, "we need banking services, not banks," and laws and regulations can prevent legally chartered banks from offering banking services. Third, various restrictions on banks undoubtedly contributed to the development and growth of competing nonbank firms and capital mar- kets. For example, the branching restrictions on banks in the 1800s lim- ited their size (see Roe 1997). This in turn limited banks' ability to pro- vide bigger loans to the large-scale commercial firms that were coming into existence. Together with the prohibitions on nonbank ownership by banks, this situation undoubtedly contributed to the growth of U.S. capi- tal markets as commercial firms increasingly sought outside funds to sup- port their growth. Of course, this development has produced desirable, albeit unanticipated, benefits; however, when the regulatory barriers to nationwide branching were finally reduced in 1995, banks were of di- minished importance in the overall financial system. Conclusions The United States has recently experienced its worst bank problems since the Great Depression. The problems occurred despite an elaborate bank regulatory structure. The obvious conclusion is that the existing structure was not appropriate for fulfilling its assigned responsibilities. Although banking institutions are now in overall good financial condition and the authorities have significantly improved bank regulation, debate about the exact way to modernize the legal definition of a bank persists. Perhaps the most important lesson from the recent past in the United States is that the most appropriate way for all countries to proceed is by viewing banks not in isolation, but instead as an integral part of a much larger financial sys- tem: a financial system that is increasingly global in nature and constantly evolving in response to new developments. Such a broader perspective suggests that relying less on extensive bank regulation and more on mar- ket discipline is the best way to proceed. James R. Barth and Robert E. Litan 169 References Barth, James R. 1991. The Great Savings and Loan Debacle. Washington, D. C.: The American Enterprise Institute Press. Barth, James R., and Philip F. Bartholomew. 1992. "The Thrift Industry Crisis: Re- vealed Weaknesses in the Federal Deposit Insurance System." In James R. Barth and R. Dan Brumbaugh, Jr., eds., The Reform of Federal Deposit Insurance: Disci- plining the Government and Protecting Taxpayers. New York: Harper Business. Barth, Jarnes R., and R. Dan Brumbaugh, Jr. 1991. The Credit Union Industry: Finan- cial Condition and Policy Issues. California Credit Union League. . 1994. "Moral Hazard and Agency Problems: Understanding Depository In- stitution Failure Costs." In George G. Kaufman, ed., Research in Financial Ser- vices, vol. 6. Greenwich, Connecticut: JAI Press. - 1997. "Development and Evolution of National Financial Systems: An Inter- national Perspective." Latin American Studies Association meeting, April 17- 19, Guadalajara, Mexico. Barth, James R., R. Dan Brumbaugh, Jr., and Robert E. Litan. 1990. The Banking In- dustry in Turmoil: A Report on the Condition of the U.S. Banking Industry and the Bank Insurance Fund. Report of the Subcommittee on Financial Institutions Su- pervision, Regulation, and Insurance of the Committee on Banking, Finance, and Urban Affairs. Washington, D. C.: Government Printing Office. _ . 1992. The Future of American Banking. Armonk, New York: M. E. Sharpe. Barth, James R., Daniel E. Nolle, and Tara N. Rice. 1997. "Commercial Banking Structure, Regulation, and Performance: An International Comparison." Eco- nomics WVorking Paper no. 97-6. Office of the Comptroller of the Currency, Wash- ington, E). C. Bartholomewv, Philip F., and Gary W. Whalen. 1995. "Analysis of Bank Failure Data: Commercial Bank Resolutions: 1980-1994." Office of the Comptroller of the Currency. Washington, D. C. Brewer, Elijah. 1995. "The Impact of the Deposit Insurance System on S&L Share- holders' Risk/Retum Tradeoffs." Journal of Financial Services Research 9 (1): 65-89. Brumbaugh, R. Dan, Jr. 1988. Thrifts under Siege. Cambridge, Massachusetts: Ballinger. Calomiris, Charles W. 1989. "Deposit Insurance: Lessons from the Record," Eco- nomic Perspectives (Federal Reserve Bank of Chicago) (May/June). _ . 1992. "Getting the Incentives Right in the Current Deposit-Insurance Sys- tem: Successes from the Pre-FDIC Era." In James R. Barth and R. Dan Brumbaugh, Jr., eds., The Reform of Federal Deposit Insurance: Disciplining the Gov- ernment ancd Protecting Taxpayers. New York: Harper Business. - 1997. "Designing the Post-Modern Bank Safety Net: Lessons from Devel- oped and Developing Economies." Conference paper from the Bankers' Roundtable Program for Reforming Federal Deposit Insurance, May 23, Ameri- can Enterprise Institute, Washington, D. C. 170 Lessons from Bank Fa2lures in the United States Caprio, Gerard,Jr., and Daniela Klingebiel.1996. "Bank Insolvency: Bad Luck, Bad Policy, or Bad Banking?" World Bank, Development Research Group, Washington, D. C. Gilbert, R. Alton. 1997. "Banks Profit from Low Rates on Time and Savings Depos- its." Monetary Trend (Federal Reserve Bank of St. Louis) June). Goldstein, Morris. 1997. The Casefor an International Banking Standard. Washington, D. C.: Institute for International Economics. Greenspan, Alan. 1997. Statement before the Subcommittee on Financial Institutions and Consumer Credit of the Committee on Banking and Financial Services, United States House of Representatives, February 13. Helfer, Ricki. 1997. Oral Statement before the Subconrmittee on Capital Markets, Securi- ties, and Government Sponsored Enterprises, Committee on Banking and Financial Services, United States House of Representatives, March 5. Herring, Richard J., and Robert E. Litan. 1995. Financial Regulation in the Global Economy. Washington, D. C.: The Brookings Institution. Hoenig, Thomas M. 1996. "Rethinking Financial Regulation." Economic Review (Fed- eral Reserve Bank of Kansas City) (Second Quarter). Jordan, Jerry L. 1996. "The Future of Banking Supervision." Economic Commentary (Federal Reserve Bank of Cleveland) (April 1). Kane, Edward J. 1985. The Gathlering Crisis in Federal Deposit Insurance, Cambridge, Massachusetts: MIT Press. - 1989a. "Changing Incentives Facing Financial-Services Regulators," Journal of Financial Services Researcih 2, September, 1989. . 1989b. The S&L Insurance Mess: How Did It Happen? Washington, D. C.: The Urban Institute Press. . 1992. "The Incentive Incompatibility of Government-Sponsored Deposit In- surance Funds." In James R. Barth and R. Dan Brumbaugh, Jr., eds., The Reform of Federal Deposit Insuirance: Disciplining the Government and Protecting Taxpayers. New York: Harper Business. Kaufman, George G. 1995. "The U.S. Banking Debacle of the 1980s: An Overview and Lessons." Financier 2 (2): 9-26. Kaufman, George G., and Randall S. Kroszner. 1996. "How Should Financial Insti- tutions and Markets Be Structured? Analysis and Options for Financial System Design." Working Paper no. WP-96-20, Federal Reserve Bank of Chicago. Kroszner, Randall S., and Philip E. Strahan. 1996. "Regulatory Incentives and the Thrift Crisis: Dividends, Mutual-to-Stock Conversions, and Financial Distress." Jouirnal of Finance 51 (4): 1285-319. Leach, James A. 1977. Statement before the Subcommittee on Capital Markets, Se- curities, and Governmnent-Sponsored Enterprises, United States House of Rep- resentatives, March 22. Lindgren, Carl-Johan, Gillian Garcia, and Matthew 1. Saal. 1996. Bank Soundness and Macroeconomic Policy. Washington, D. C.: International Monetary Fund. James R. Barth and Robert E. Litan 171 Litan, Robert E. 1987. What Should Banks Do? Washington, D. C.: The Brookings Institution. - 1991. The Revolution in U.S. Finance. Washington, D. C.: The Brookings Insti- tution. . 1994. "U.S. Financial Markets and Institutions in the 1980s: A Decade of Tur- bulerLce." In Martin Feldstein, ed., American Economic Policy in the 1980s. Chi- cago: The University of Chicago Press. Ludwig, Eugene A.. 1997. Oral Statement before the Subcommittee on Capital Mar- kets, Securities, and Govemment-Sponsored Enterprises, Committee on Bank- ing and Financial Services, United States House of Representatives, March 5. National Credit Union Administration. 1995. Annual Report. Alexandria, Virginia. Resolution Trust Corporation. Various years. Statistical Abstract. Washington, D. C. Roe, Mark J. 1997 "The Political Roots of American Corporate Finance." Journal of Applied Corporate Finance 9 (4): 8-22. Romer, Thomas, and Barry R. Weingast. 1992. "Political Foundations of the Thrift Debacle." In James R. Barth and R. Dan Brumbaugh, Jr., eds., The Reform of Fed- eral Deposit Insurance: Disciplining the Government and Protecting Taxpayers. New York: Harper Collins. Shadow Financial Regulatory Committee. 1997. Statement No. 137. May 5. White, Lawrence J. 1991. The S&L Debacle. New York: Oxford University Press. 10 The Banking Crisis in Japan Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito In the 1970s many industrial and developing economies embarked on a tran- sition of financial institutions and markets from administratively controlled to more open and competitive structures. The transition was driven by con- flicts between existing financial structures, inflationary macroeconomic poli- cies, and the emergence of a new economic and technological environment. Despite considerable progress toward liberalized financial structures, the transition has not progressed smoothly, marked in rnany cases by se- vere financial disruptions and banking problems (see Borio, Kennedy, and Prowse 1994; Economist 1997; Lindgren, Garcia, and Saal 1996). Japan initiated financial liberalization in the mid-1970s, and until the mid-1980s it proceeded at a slow, steady pace with few disruptions to the real or financial sectors, especially when compared to the United States (Cargill and Royama 1988). What started as a smooth transition changed dramati- cally in the second half of the 1980s, however, with the sharp run-up in asset prices and the booming economic and monetary growth that char- acterized the "bubble economy" (Cargill, Hutchison, and Ito 1997). The bubble economy followed the pattern of a classic speculative bubble, and the subsequent fall in asset prices in the early 1990s and the associated recession had an adverse impact on banks' balance sheets. Financial in- stitutions were saddled with a massive nonperforming loan problem es- timated at some US$500 billion (as of late 1995), with some estimates putting the figure at more than twice that amount. Regulatory inertia, forbearance, and forgiveness despite awareness of the same policy failures earlier in the United States have exacerbated Japan's 173 174 The Banking Crisis in Japan banking crisis and nonperforming loan problem. The piecemeal and tenta- tive approach the Japanese Ministry of Finance initially followed in dealing with the banking crisis exhausted the deposit insurance funds'-the De- posit Insurance Corporation (DIC) and the insurance agency for small insti- tutions-limited resources without confronting the underlying problems.' Against the background of a nonperforming loan problem, depressed asset markets, and the weakened condition of financial institutions, Japan has been confronted with its most serious financial crisis since the early 1950s. The turning point was reached in 1995-96, when the authorities took more decisive action. They departed from the official no failure of finan- cial institutions policy, committed public funds explicitly to problem in- stitutions, restructured the two deposit insurance agencies, and passed legislation to deal with the special problems of the jusen (largely insol- vent financial institutions that had mainly provided credit for real estate) and the credit cooperatives. Moreover, in November 1996 Prime Minister Ryutaro Hashimoto-at the head of a newly reorganized Liberal Democratic Party elected follow- ing a campaign heralding broadly based reform-announced extensive deregulation of Japan's financial system by the year 2001. Senior officials have likened this proposal to the "Big Bang" financial deregulation in the United Kingdom more than a decade ago. The far-reaching objective of the initiative is to make Tokyo a global financial center that rivals New York and London, based on three basic principles of reform: market mechanisms, global in nature, and transparency. Acting on this initiative, the Diet passed legislation in mid-1997 that, among other things, created the new Financial Supervisory Agency (FSA), thereby greatly reducing the role of the Minis- try of Finance in monitoring and supervising the financial system and fur- ther deregulating the foreign exchange market (see Choy 1997 for a sum- mary of these legislative events). The changes enacted and those proposed for future legislation, even if they resulted in major benefits to the economy as a whole, necessarily en- tail losers as well as winners. The potential losers, long protected by regu- lation from market competition in Japan's compartmentalized financial services industry, are likely to oppose further change vigorously. Financial deregulation has been on the agenda for many years, has proceeded only gradually, and some skeptics argue that a sense of deja vu surrounds the present push for deregulation. 1. The DIC is the major deposit insurance agency in Japan. Organized in 1971, the DIC insures the deposits of banks and nonbank depositories. Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 175 Japan's financial system, however, is at a juncture today that is not com- parable to any other episode during the past 45 years. Stress in the finan- cial system, especially failure to resolve the nonperforming loan problem quickly, continues to hold back the economy, and a large part of the real estate market has stagnated as a result. The shortcomings of the existing regulatory and supervisory structure, especially in the context of the major changes in financial institutions and markets during the past 15 years, are readily apparent. Market forces and competition among financial institu- tions make the existing financial structure incompatible with Japan's regu- latory and supervisory structure. The recent legislation creating the FSA recognizes this and attempts to address the problem. Moreover, economic and political pressure for fundamental reform will likely continue, even if the present wave of popular opinion against Japan's financial institutions and regulators, especially the Ministry of Finance, wanes. This chapter discusses the conflict between the existing financial sys- tem and regulatory structure, and focuses on whether recent changes are likely to resolve this problem. Our research suggests that a central ele- ment of any successful reform is the introduction of a new policy ap- proach that avoids past mistakes, especially regulatory delays, moral hazard, and inadequate funding, when dealing with insolvent financial institutions burdened with nonperforming loans. A more explicit and transparent accounting, supervisory, and regulatory framework would help to avoid future banking problems, and the creation of the FSA ap- pears to be an important step in this direction. Financial iDeregulation and the Bubble Economy In the second half of the 1980s asset inflation was evident in many countries, though not to the same extent as in Japan. Indexes of real asset prices for 13 industrial countries consisting of equity, residential real estate, and commercial real estate components illustrate a broadly based process of asset inflation during this period (Borio, Kennedy, and Prowse, 1994). The international character of asset inflation suggests common ex- planatory factors. The coincidence of financial liberalization and asset inflation and deflation has led a number of observers to argue that liber- alization played a major role in the financial disruptions of the 1980s and in the problems that the 1990s inherited from the boom and bust period. In the contexc- of liberalization in the 1980s, the removal of binding port- folio constraints permitted banks and other depositories to take on riskier 176 The Banking Crisis in Japan investment and loan portfolios, including high loan to value ratios.2 Banks, directly or indirectly, provided imprudent levels of credit to real estate and equity markets in an effort to offset declining profit margins and declining market shares and to maintain the franchise value of commer- cial bank charters supported in the past by a regulated and administra- tively controlled financial environment. The following features of Japan's financial system in the mid-1980s made asset inflation more probable in the context of a newly liberated financial structure and accommodating monetary policy: * Market participants had more portfolio flexibility than they had ever had in the past. This was especially true for small depositories like savings and loan institutions. In the 1980s small depositories of the credit union type aggressively pursued lending in speculative real estate ventures during the bubble phase, and like their U.S. counter- parts, did so without oversight. * The shift to a slower growth path after the first oil price shock in 1973 reduced the corporate sector's reliance on bank credit and ser- vices. As a result, banks sought out new markets outside traditional corporate finance and were willing to assume new, and often higher, risks for which they had little previous experience. * The main bank system began to unravel in response to financial liber- alization. In the past, this system had served as an effective tool for evaluating and monitoring risk. No widely available financial disclo- sure framework was available to replace the main bank system. a Despite a common perception that Japanese banks were subject to a Glass-Steagall type of rule, they actually have considerable author- ity to purchase and hold equities directly, thereby allowing bank credit to flow easily into equity markets. In addition, banks' hidden reserves were tied directly to the fortunes of the stock market. I The regulatory monitoring system lagged behind market develop- ments and administrative guidance could not keep pace with the rapidly changing financial environment. 2. "In those countries where the asset price boom was most marked in the 1980s (Finland, Sweden, Norway, Japan and the United Kingdom) or where the disruption caused by the downward correction in valuations has caused great concem (Australia and the United States) there is a relatively close correlation between the ratio of private credit to GDP and asset price movements.. .To a large extent, the major expansion of credit during the past decade reflected a relaxation of credit constraints in the financial industry in wake of both market-driven and policy-determined structural develop- ments" (Borio, Kennedy, and Prowse 1994, pp. 27-28). Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 177 C omplete deposit guarantees encouraged risk taking at the very time the Bank of Japan provided the liquidity and financial liberalization p;rovided the asset diversification powers. The risk incentive nature of government deposit guarantees (the moral hazard problem) plays an important role in accounting for the coinci- dence of asset inflation and financial liberalization in Japan and elsewhere (see Iwainura 1993 for a simulation study of the effects of moral hazard on portfolio choice). In most cases liberalization failed to change the sys- tem of government deposit guarantees, which had been designed for a more reg,alated and administratively controlled financial environment. As a result, government deposit guarantees provided incentives to as- sume risk:, while at the same time, regulatory and market innovations permitted depositories to manage and assume more risk. In addition, regu- latory authorities responsible for administering the government deposit guarantee system were subject to perverse incentives in how they dealt with troubled institutions, which in turn encouraged greater risk taking on the part of depository institutions. Financial Liberalization, Other Fundamentals, and Bubbles Table 10.1 summarizes the factors that generated the asset price bubble and the consequent nonperforming loan problem. The initial jump in stock and land prices during 1985-86 was probably related to changes in funda- mentals: aggregate productivity gains in Japan were extremely strong at the time; demand for real estate, particularly in the Tokyo area as a direct result of libe:ralization, increased significantly; and expansionary monetary policy pushed interest rates to low levels. Financial deregulation, changes in the flow of funds, and banks' increased risk taking activity also played a role at the beginning of the run-up in asset prices, but whether these factors were fundamental determinants or speculative activities is unclear. The increased risk taking was rational in the sense that banks aggressively expanded into the real estate business given the newly expanded powers over asset allocation, the changes in traditional business relationships, and so on. In addition, financial deregulation combined with lax supervision and extensive deposit guarantees allowed expanded lending into real estate on a speculative basis. At some point, probably in late 1986 or 1987, the asset inflation process became a speculative bubble with little restraint shown by the financial institutions, the regulatory authorities, or the Bank of Japan. Expectations 178 The Banking Crisis in Japan Table 10.1. Factors that Created the Asset Price Buibble and Economy Contributing to the Nonperforming Loan Problem Factor Outcome Financial liberalization starting in the Increased asset diversification powers mid-1970s and continuing to the present. for bank and nonbank depositories; increased ability to manage and assume risk unequaled in the postwar period. Downward shift in potential GDP growth Banks sought out new markets to re- path after 1973 oil price shock reduced establish and enhance their market corporate dependence on the banking share. Specifically, banks moved system as the rate of investment fell from aggressively into real estate lending rates prevalent during the high-growth and competed with jusen in the 1980s. period. The decline of the main bank system and Reduced banks' ability to monitor risk. the overall reduced role of banks in enterprise groups. Banks were permitted to hold equities in Banks' capital position became depen- nonfinancial businesses as part of the dent on the equity market, so that enterprise group framework. Banks during equity price increases banks could apply 45 percent of the appreciation would expand loans and investments, of equity holdings to capital referred to and during equity price declines as hidden or latent capital. banks would contract loans and investments. Reliance by the regulatory authorities on Regulatory authorities became less administrative guidance to monitor bank able to monitor risk as bank and non- and nonbank depositories. bank depositories engaged in greater asset diversification than in the past. Bank and nonbank depositories had few incentives to limit risk. Pervasive system of government deposit The reduced oversight and enhanced guarantees. asset diversification powers became more sensitive to the risk incentives embedded in the deposit guarantee system. Monetary easing in 1987 and 1988 as the The Bank of Japan provided liquidity Bank of Japan placed greater emphasis that supported the asset inflation. on exchange rate objectives than on domestic considerations. Incentives for regulatory authorities to Incentives to assume risk were adopt forbearance and forgiveness in enhanced. dealing with troubled institutions. Source: Cargitl, Hutchison, and Ito (1997). Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 179 of asset price increases fed upon themselves, and price to dividend and price to rent ratios increasingly deviated from fundamental values until the crash in the 1990s. Speculators typically believed that even though the levels oi stock and land prices were abnormally high and would eventu- ally fall, they believed that further investment was warranted as long as other investors thought that prices would continue to rise. Many specula- tors assumed that they would be among the first to sell their asset hold- ings, the:reby realizing large capital gains, when the market started to fall. This kincl of behavior has been termed as stochastic bubbles, herd instinct, momentLm trading, or bandwagon behavior. Eventually asset prices reversed course, and declined rapidly in the early 1990s. The Nikkei 225 stock price index reached its peak of 38,915 on the last business day of 1989, and then tumbled. By October 1, 1990, the Nikkei stood barely above 20,000, having lost almost 50 percent of its value in nine months. The Nikkei fell below 15,000 in mid-1992, and only broke the 20,000 point again in 1996. Land prices began to fall in late 1991, and by 1996 prices were frequently only half of their peak values. At that time, the land prices ave.raged those prevalent 10 years earlier. A combination of policy actions and the self-correcting mechanism of the speculative process (deflating the bubble) were responsible for the asset price decline. By mid-1989 the monetary authorities became fully aware of, and concerned about, asset price inflation and started to raise interest rates. The Ministry of Finance also introduced several measures to slow land price rises. The Iraqi invasion of Kuwait on August 2, 1990, further weakened the world economic outlook, and the prospects for oil- dependent Japan in particular. Furthermore, once the decline in asset prices began, banks had an incentive to reduce lending for real estate and other purposes. When the Basel risk-based capital ratio was negotiated in 1988 as an international minimum standard for banks with interna- tional busin(ess, Japanese banks were allowed to count 45 percent of their equity holdings as part of their Tier Two capital. As the value of these equities was devoted to meeting capital asset requirements, the cost of capital increased, reducing banks' incentive to make loans and contribut- ing to the drop in credit expansion. Despite these factors, the most important reason for the collapse in asset prices was the self-correcting mechanism inherent in stochastic speculative processes. Expectations of further price declines generated selling, which in turn led to price declines. In early 1990, for example, Nikkei futures tended to lead the decline in cash markets. This was re- sponsible for the view that futures transactions were making the stock market too volatile, and led to a tightening of margin requirements in the Nikkei futures market in Osaka in 1991. 180 The Banking Crisis in Japan Nonperforming Loans, the Jusen Problem, and Regulatory Inertia Asset deflation dramatically affected the profitability of Japanese financial institutions and led to serious concerns about the stability of the entire financial system. Deterioration in the quality of loans to the real estate sec- tor was the primary problem, but was compounded by the drop in the value of banks' large equity holdings and growing loan problems associ- ated with a prolonged recession.3 The Ministry of Finance, as the primary regulatory agency, was slow in reacting to the nonperforming loan problem confronting financial institu- tions. After the initial sharp decline in stock and land prices in 1991-92, the ministry initially adopted a forbearance policy, allowing banks to hold nonperforming loans without special write-offs in the hope of a quick re- covery of the economy and the real estate market (Folkerts-Landau, Ito, and others 1995, p. 2). However, smaller financial institutions with large real estate exposure began showing signs of distress in 1993, soon followed by even larger problems for jusen or housing loan companies (nonbank subsidiaries of financial institutions specializing in housing loans). At that time the Ministry of Finance arranged a 10-year rehabilitation plan forjusen. Rather than recognizing loan losses, however, the plan was predicated upon a land price recovery. When land prices failed to recover, and with nonperforming loans growing in size and number, the ministry quickly abandoned the rehabilitation plan. As the financial crisis unfolded, in 1994 and 1995 the regulatory au- thorities launched a new approach for resolving these problems. The Ministry of Finance closed some of the lowest quality institutions, cre- ated the Tokyo Kyodo Bank as a bridge bank that would receive the remaining assets of failed smaller institutions, allowed (or encouraged) massive write-offs by some banks, and decided to close down the jusen. An infusion of public funds, proposed for the first time in the 1996-97 budget, was earmarked to cover the costs of closing seven jusen compa- nies, and a variety of other measures were enacted in the wake of the nonperforming loan and jusen problem. The jusen companies had been created in the mid-1970s as subsidiar- ies of banks, securities firms, and life insurance companies. At that time, 3. Japan's most severe postwar recession itself can largely be traced to the decline in asset prices. Consumption fell as households saw the real value of their equity and real estate holdings decline precipitously. The fall in consumption induced a fall in fixed investment, which was further weakened by the overhang of excess capacity accumulated during the asset inflation phase and the credit crunch induced by the decline in banks' hidden capital. 7homas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 181 banks were concentrating on corporate lending, and were not generally interested in expanding their operations to household lending, either for mortgage or consumer credit. The jusen companies provided consumer credit, and much like consumer finance companies in the United States, borrowed from other institutions, because they were not permitted to accept deposits. As the corporate sector reduced its dependence on bank credit after 1975, however, banks began to turn to consumer finance as a new line of business, and in the 1980s became aggressive lenders to indi- viduals. In response to the competition for personal loans, jusen compa- nies turned to real estate lending in the second half of the 1980s to substi- tute for thie lost consumer lending business. In April 1990 the Ministry of Finance introduced regulations to limit total bank lending to the real estate sector; however, jusen lending was exempted, During 1990-91 jusen lending increased rapidly as a result of funds provided by agricultural cooperatives and their prefectural fed- erations. Concerns about jusen asset quality were raised as early as 1992, and the authorities arranged a 10-year rehabilitation plan for seven of the eight jusen companies in early 1993. (The eighth jusen, established by ag- ricultural cooperatives, is not considered to be a problem institution and was exempt from the plan.) Lending to jusen was restructured so that parent banks (jusens' major shareholders) were required to reduce the interest rate on outstanding loans to zero, other banks (nonparent banks that lent to the jusen) were required to reduce the interest rate on out- standing loans to 1.5 percent, while agricultural cooperatives and their prefectural federations were to receive interest income of 4.5 percent from jusen. These arrangements were designed to provide liquidity to jusen until the expected future recovery in land prices permitted the outstand- ing loans to be paid off in 10 years. Land prices, however, continued to fall, and the nonperforming loans held by jusen rose dramatically. For all practical purposes, the rehabilitation plan itself became bankrupt. The jusen problem became the focus of intense policy debate in 1995, and even overshadowed the banks' nonperforming loan problem. In Au- gust 1995 the Ministry of Finance conducted a special examination of the jusen problem. Of the total Y 13 trillion of jusen assets, it estimated nonperforming loans to be worth Y 9.6 trillion, of which Y 6.4 trillion was unrecoverable and Y 1.2 trillion was a possible loss. This amounts to more than a quarter of all losses financial institutions had incurred to that date. The Ministry of Finance and the suppliers of funds to jusen companies agreed to dissolve the seven housing loan companies, and in July 1996 the Housing Loan Administration was established to assume Y 6.4 trillion of unrecoverable loans extended by failed jusen companies. 182 The Banking Crisis in Japan Financial markets became increasingly concerned about the burden sharing of the jusen losses among banks and other financial institutions that further reduced the creditworthiness of Japanese banks because poli- ticians and ministry officials suggested that banks should shoulder more than a pro rata share of jusen losses. The precedent for this approach had already been set when banks were required to provide a more than pro rata share in schemes for dealing with failing institutions in July and August 1995 (Cosmo credit cooperatives, Kizu credit cooperatives, and the Hyogo Bank). Banks unrelated to these institutions were asked to con- tribute to the loss-sharing scheme by contributing capital or making be- low market rate loans to banks that assumed the assets of the insolvent institutions. This request was rationalized on the public good character of the financial system and the need to maintain stability. The policy of requiring unrelated banks to contribute directly to bail out schemes was referred to as the all-Japan rescue scheme. Uncertainty about future bank losses because of doubts about the specific jusen resolution scheme and the extent of the all-Japan compo- nent, however, led to a downgrading of Japanese banks' creditworthi- ness. The perception of increased risk in Japanese banks at large resulted in a Japan premium in the Euro-dollar market over the London inter- bank offered rate. The Japan premium appeared in late July to mid- August 1995, and persisted for several months, even though the Minis- try of Finance attempted to reassure the market by announcing a complete deposit guarantee to 2001. Regulatory Response to the Nonperforming Loan Problem The first official response to the nonperforming loan problem was a series of uncoordinated actions predicated on a short recession and a rapid re- covery of asset prices. On August 18, 1992, the government announced a temporary rule change that allowed corporations to defer reporting their stock portfolio losses until the end of the fiscal year (March 1993); permit- ted other accounting innovations that delayed or concealed the impact of stock and land price declines on reported assets; allowed banks, in special cases where a loan default would have adverse social effects, not to report interest concessions as taxable income (Packer 1994); directed the Postal Life Insurance System to support the stock market through funds provided to trust banks; postponed sales of government held shares of Nippon Tele- graph and Telephone and Japan National Railways; used administrative guidance to encourage institLtional purchases of equities and discourage institutional sales of equities; and provided less than candid estimates of Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 183 the magnitude of the nonperforming loan problem.4 These actions had little impact on the downward trend in asset prices and the deterioration of the financial system. More specific and aggressive actions followed. These will be considered more or less in chronological order. Establishment of the Cooperative Credit Purchasing Company The Cooperative Credit Purchasing Company (CCPC) was the first obvi- ous effort to deal with the nonperforming loan problem. The government established the CCPC in late 1992, and it commenced operations in Janu- ary 1993. The CCPC uses pooled funds from 162 banks and cooperatives with the active encouragement and involvement of the Ministry of Fi- nance. The CCPC purchases real estate loans at prices determined by a panel of experts, with the institution selling the loan providing the fi- nancing. The CCPC then sells the real estate, and any difference between what the CCPC paid for the loan and the selling price is charged to the institution that originally sold the loan. From the outset the market viewed the CCPC with skepticism (Choy 1992). Concerns were voiced that the planned loan purchases would represent only a small percentage of the nonperforming loans; that the self-financing require- ment limited access to only the strongest banks; that only the best of the nonperforniing loans would be sold to the CCPC, thereby delaying the inevi- table adjustment; and that as the original borrowers continued to manage the properties, their incentives to cooperate were unclear. These concerns persist despite several years of operation during which the CCPC purchased Y 8.7 trillion of loans at face value with an average loss ratio from March 1993 to May 1995 of 55.4 percent. The critical issue, however, is to what purpose the loans have been purchased and how the 4. Until March 1993 the Ministry of Finance reported the sum of nonperforming loans only for the 21 majorbanks combined. The ministry's definition of a nonperforming loan is narrow: those loans for which the principal is not likely to be collected and those for which interest has not been collected in the past six months. This leaves consider- able room to e clude loans that are clearly in serious default, for example, restructured loans in which the interest rate has been drastically reduced, or even set to zero, are not regarded as nonperforming. Loans that receive only a fraction of an interest payment can also be excluded from the nonperforming category. Moreover, evidence indicates that banks made additional loans to enable borrowers to pay interest on previous loans, so the nonperforming loan amount is understated. This type of activity on the part of private depositories was widespread in the savings and loan industry in the 1980s, with disastrous consequences. Based on the loss ratios reported by the Cooperative Credit Purchasing Company and experience with insolvent institutions, the ministry's estimates of recoverable loans also appears to be conservative. 184 The Banking Crisis in Japan CCPC plans to dispose of the loans. Sales of CCPC loans have been only a fraction of total holdings, and the concern is that revenues from loan sales and rental income from properties will be insufficient to repay interest to the funding banks at market rates. Establishment of the Resolution and Collection Bank The Tokyo Kyodou Bank was established in March 1995 to assume the as- sets of several failed credit cooperatives. The bank was reorganized in Sep- tember 1996 into the Resolution and Collection Bank, which was loosely modeled after the Resolution and Trust Corporation, established in 1989 to dispose of the assets of failed savings and loan institutions. The CCPC's track record suggests that the Resolution and Collection Bank will have a difficult time liquidating bad loans. Many of these loans are tied to a real estate market that remained depressed during 1997. The fact that at the time of writing the CCPC had sold only a fraction of the loans assumed, which were the "best" of nonperforming loans, is not an impressive indicator of the likely success of the Resolution and Collec- tion Bank. The depressed real estate market, the slow pace of economic recovery, the lack of well-developed foreclosure and bankruptcy proce- dures, and even the influence of organized crime in some real estate trans- actions make disposing of nonperforming loans difficult. A possibility is that the CCPC, the Housing Loan Administration Corporation, and the Resolution and Collection Bank will end up being a form of forbearance, that is, merely new warehouses for bad loans, when what the Ministry of Finance really needs is a garage sale. Provision of Assisted Mergers Using the Deposit Insurance Corporation In 1991, for the first time in the postwar period, the regulatory authori- ties officially assisted in the mergers of insolvent depository institutions with stronger institutions using the DIC, the largest of Japan's two de- posit insurance agencies. Before 1991 the DIC had never had to pay de- positors, although some claim that the Ministry of Finance arranged several unpublicized rescue mergers without DIC assistance. Since 1991, however, the DIC has publicly assisted a small number of problem in- stitutions, and two of the institutions it recently assisted exhausted the DIC's reserves.5 The details of these assisted mergers, examined in 5. This is not surprising, because the DIC was grossly underfunded. To gain an appreciation of the degree of underfunding one can compare the DIC to the Federal Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 185 Cargill, Hutchison, and Ito (1997), reveal a disturbing pattern about the ability of Japanese regulatory authorities to administer deposit guaran- tees and limit moral hazard effectively. During its short history the DIC has had a turbulent record. The first 20 years were uneventful and few regarded the DIC as an important compo- nent of Japan's deposit guarantee system, although a number of commen- tators pointed to problems that could arise as a result of a poorly funded deposit insurance agency (see, for example, Cargill 1995; Ito and Ueda 1993). Since 1991, however, the DIC has become a focal point of concern for two reasons: iEirst, bailouts in late 1995 rendered the DIC insolvent; and second, the approach to assisted mergers imposed few penalties on shareholders and none on depositors. Resolution of the Jusen Problem and the Use of Public Money In December 1995, the government proposed a resolution plan for the seven insolvent jusen companies that the Diet passed in early 1996. Of the Y 13 trillion in assets of the seven jusen companies, Y 6.4 trillion was an immediate (primary) loss, Y 1.2 trillion was a possible (secondary) loss, Y 2.1 trillion was nonperforming but possibly recoverable, and Y 2.5 trillion was a normal or performing asset. The primary loss of Y 6.4 tril- lion was tc be borne by banks and life insurance companies, by agricul- tural cooperatives as a "gift" (the agricultural cooperatives insisted that any funds they lent would be repaid in full and that any burden they assumed would be in the form of a gift they would make to the bank that assumed the bad jusen assets), and public spending by the government. The Jusen Resolution Corporation, now known as the Housing Loan Administration, assumed the remaining assets. The secondary losses will be dealt witlh in the future using special accounts. The plan has two problems. First, burden sharing was inequitable and reflected the political strengths of the agricultural sector. The strength of farming interests was revealed, for example, during the de- bate about sharing the cost burden of the financial crisis when it be- came public that the Ministry of Finance's Banking Bureau director gen- eral had sec;retly signed a memorandum of understanding with his Deposit Insurance Corporation in the United States. While close to 80 percent of de- posits in both co antries are insured, the reserve to deposit ratios are significantly lower in Japan. In 1994, for example, the ratio of insurance fund reserves to insured deposits was 0.16 in Japan and 1.15 in the United States (Deposit Insurance Corporation of Japan 1995; Federal Deposit Insurance Corporation 1994). 186 The Banking Crisis in Japan counterpart at the Ministry of Agriculture, Forestry, and Fisheries that promised the Ministry of Finance's full backing of agricultural coop- eratives lending to jusen companies. The relatively heavier burden on banks also reflected the public perception that the banks that had set up jusen as subsidiaries had a direct involvement injusen operations. These founder banks provided staff to manage operations and maintained business relationships with the jusen in the form of referring potential borrowers, who turned out to be high risk customers. Second, although the proposed amount of public funding was small compared to the magnitude of the nonperforming loan problem, the politi- cal opposition to using taxpayer funding was strong and surprised some observers. The resolution plan will spend only Y 685 billion of public funds in filling the gap between the primary loss and private sector burdens.6 The opposition turned out to be so strong, however, that policymakers are likely to be reluctant to propose public funding as part of any solution to deal with financial problems in the future. Unfortunately, the problem and its resolution were poorly presented by the Ministry of Finance. The plan was actually a reasonable starting point for tackling the financial debacle, but its purpose was lost in the de- bate about burden sharing, policy errors on the part of the Minis try of Fi- nance, and protection of the agriculture sector. Lessons for Regulatory and Supervisory Policy The regulatory response to the nonperforming loan problem and the reso- lution of the jusen problem against the backdrop of widespread deposit guarantees offers several lessons. Any sweeping overhaul of Japan's regu- latory and supervisory structure should address each of the issues if Japan is to undertake a serious Big Bang deregulation of its financial system. The regulatory response was slow, and once regulatory action was ini- tiated, it often made the problem worse. This was due to improper admin- istration of the deposit guarantee system and failure to appreciate the moral hazard inherent in protecting depositors and assisting troubled institutions. 6. The amount of Y 685 billion figure was derived in a curious manner. During the negotiations between the Ministry of Finance and the Ministry of Agriculture, For- estry, and Fisheries, the Ministry of Finance originally proposed that the agricultural cooperatives share the burden of Y 1.2 trillion. Political opposition from the farmers lobby was so strong that this figure was bargained down to V 530 billion. The differ- ence between the estimate by the Ministry of Finance and the amount accepted by the Ministry of Agriculture, Forestry, and Fisheries, namely, V 685 billion, became the amount of fiscal spending proposed for resolving the jusen problem. Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 187 This fa.ilure is even more striking given that Japanese policymakers were familiar with the problems of and lessons from the U.S. experience. By 1995 Japan had come under intense international pressure to take decisive action to solve its nonperforming loan problem. An International Monetary Fund report (Folkerts-Landau, Ito, and others 1995), highly pub- licized in Japan, criticized the authorities for their delayed response. The market's judgment was also critical: the credit ratings of Japanese banks, previously at the top of the industry internationally, were downgraded sharply. The Daiwa Bank scandal and the Ministry of Finance's violation of international bank regulation rules through its failure to inform U.S. regu- latory authorities of the problem in a timely manner further weakened con- fidence in Japanese financial institutions and regulatory authorities. In the context of growing public awareness of the nonperforming loan and jusen problems, as well as intense international pressure, in Decem- ber 1995 the government initiated a series of reforms to cope with future financial problems. Legislators introduced six bills that the Diet passed in June 1996 that dealt with jusen resolution; established corrective ac- tions to limit the operations of weak financial institutions; and strength- ened deposit insurance by increasing the system's staff, increasing de- posit insurance premiums sevenfold to strengthen the system's reserves, and setting up a special deposit insurance reserve for credit cooperatives during the consolidation process. This round of actions omitted reform of the supervisory framework. In light of the Ministry of Finance's poor performance in dealing with the country's financial problems, some observers questioned the effectiveness of any reform as long as the Ministry of Finance remained the primary financial regulatory authority. These concerns are largely responsible for the government creating the FSA in mid-1997, which will gradually as- sume most of the ministry's supervisory and regulatory responsibilities. Delay in shutting down an insolvent financial institution increases losses that will have to be dealt with when resolving the problem. This is a classi- cal case of the moral hazard problem of deposit guarantees and the admin- istration of those guarantees. Permitting insolvent or almost insolvent in- stitutions to operate provides incentives for the institution to either gamble on high risk, high return projects or to engage in fraudulent behavior, such as lending to top management of related companies. The jusen industry had numerous incidents of excessive risk taking and fraudulent behavior. Indecision by the Ministry of Finance is primarily responsible for the delayed response, especially during 1992 and 1993, as by that time the jusen problem had been evident for several years, both to the ministry and to market insiders, before becoming public knowledge. Documents 188 The Banking Crisis in Japan submitted to the Diet in February 1996 show that the Ministry of Finance made the first on-site examinations of jlusen in 1991-92. Those examina- tions revealed that 67 percent of loans made to the largest 50 borrowers were already nonperforming; however, the ministry allowed the jusen companies to operate on the assumption that land prices would rise in the near future. Instead land prices continued to decline and the jusen problem increased in magnitude: in the subsequent four years nonperforming loans in the juseut increased by 75 percent. A number of factors contributed to the delay in regulatory response, namely: (a) the lack of political leadership; (b) the administration of de- posit guarantees; (c) the political power of the agriculture sector; (d) the existence of three separate, and in some cases competing, regulatory au- thorities (the Ministry of Finance; the Ministry of Agriculture, Forestry, and Fisheries; and the Ministry of Posts and Telecommunications); and (e) the general unwillingness to recognize that Japanese financial institutions are susceptible to moral hazard problems. Moral Hazard in Japan The moral hazard problem facing Japanese financial institutions is similar in magnitude to its counterparts abroad. The failure of two credit cooperatives in December 1994, Tokyo Kyowa and Anzen Credit Cooperative, offers some insights. The Tokyo Metropolitan Government, the direct supervisor in the case, reportedly became aware of the insolvency of the two credit coopera- tives in early 1993, when it conducted a special joint examination with the Ministry of Finance. The delay in closing the two institutions substantially increased the ultimate cost of the bailout. The two credit cooperatives suffered from a classic case of moral hazard in the last two years of operation (Asahi Shinbun February 16, 1995). Deposits at both institutions increased from Y 139 billion in March 1992 to Y 244 billion in November 1994, an annual rate of 32 percent, while lending increased from Y 137 billion to Y 225 billion, an annual rate of 22 percent, during the same pe- riod. Most new loans made during this time were ultimately classified as nonperforming. The total amount of nonperforming loans of the two credit cooperatives increased from Y 250 billion (out of total loans of Y 1,371 billion) in March 1992 to Y 1,769 billion (out of total loans of Y 1,990 billion) in March 1994. Moreover, unrecoverable losses increased from V 65 billion in March 1992 to Y 1,118 billion in March 1994. Not only did the two credit unions aggres- sively expand deposits by offering above market deposit rates to make new, higher risk loans during their decline into insolvency, but the amount of total Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 189 nonperforming loans increased sevenfold and of unrecoverable losses increased sixteenfold during the same period. The portfolio behavior of the two credit cooperatives illustrates the in- centives facing private depository institutions in the presence of govern- ment deposit guarantees when confronted with actual or impending insol- vency. Insolvent or close to insolvent institutions have an incentive to take on high risk investments in hopes of earning large returns that would re- store them to profitability. Depositors have little incentive to monitor the portfolio shift, because they are protected by the deposit guarantees. The supervisory agencies must assume the monitoring role, and in this respect the Ministry of Finance responded inadequately. The U.S. deposit insur- ance failure in the late 1980s and the early 1990s and the current difficulties in Japan illustrate the universality of the problem.7 The 1997 Legislative Agenda The collapse of asset prices in the 1990s, the recession, the nonperforming loan situation, and the continued deterioration of the financial system combined with the Ministry of Finance's failed policy response provides the background for the recent flurry of legislative efforts. In March 1997 the government introduced several pieces of legislation to initiate the Big Bang reforms (Choy 1997). The Diet passed part of this legislative package by mid-year, namely: I The foreign exchange market will be deregulated by eliminating the need to have a license to conduct foreign exchange operations, abolish- ing tile requirement to obtain advance approval for foreign direct in- vestments and overseas capital transfers from the Ministry of Finance, and eliminating the daily limits on positions in any one currency. * The Ministry of Finance's regulatory responsibilities will be reduced in two areas. First, its current monitoring and supervision of the fi- nancial sector will be gradually transferred to the FSA, which reports directly to the prime minister. The FSA will also have the power to monitor and supervise agricultural cooperatives, labor cooperatives, and a wide range of finance and leasing companies that the Ministry of Agriculture, Forestry, and Fisheries; the Ministry of Labor; and the 7. Before the series of credit cooperatives failures in late 1994 and 1995, several authors (Cargill and Todd 1993; Kane 1993) had argued that Japan was missing a chance to leam from the U.S. deposit insurance failure when legislators were discussing the Financial Systern Reform Law prior to its passage in June 1992. 190 The Banking Crisis in Japan Ministry of International Trade and Industry, respectively, currently monitor and supervise. Second, the operations of the Securities and Exchange Surveillance Commission, now part of the Ministry of Fi- nance, will be shifted to an agency outside the ministry. * The 1942 Bank of Japan Law will be revised to provide the Bank of Japan, one of the world's most formally dependent central banks, with enhanced formal independence. This is to be accomplished by limiting the Ministry of Finance's role over the bank's Policy Board, restricting the ministry's influence on the bank's budget and personnel matters, and improving the transparency of Bank of Japan policymaking by making Policy Board discussions and decisions public and by requir- ing the bank to provide the Diet with biannual reports. * The long-standing prohibition against a holding company structure for industrial organizations will be removed. * The corporation funded by the government will be required to disclose financial information much like their publicly traded counterparts. * The Nippon Telegraph and Telephone Corporation will be reor- ganized along holding company lines to enhance domestic com- petition in telecommunications. In addition to these specific proposals, the Liberal Democratic Party's policy committee is introducing a number of reform proposals to effect the government's financial market reforms and to move up the dead- line from 2001 to 1998. The flurry of legislative and reform efforts in the first half of 1997 raises two questions. First, how successful will these reforms be, both in terms of their content and their implementation; and second, do the reforms deal with the immediate problems in the financial system? Likelihood of Success The new legislation created the FSA, which is slated gradually to assume most of the regulatory and supervisory responsibilities currently held by the Minis- try of Finance. The ministry, however, will still be responsible for drawing up financial policies and creating financial market structures. The FSA must also confer with the finance minister if the FSA's actions are likely to affect the stability of the financial system. Whether this requires the FSA simply to con- fer with the ministry, or whether the ministry or other supervisory agencies must give their approval before a financial institution is be closed is unclear. Moreover, FSA decisions may not be completely independent of other miunis- tries: its initial staff of 300 people will include around 210 transferred from the Th7omas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 191 inspection offices of the Ministry of Finance and the Ministry of International Trade and Industry, and the others will come from the Securities and Exchange Surveillance Commission. Finally, the FSA needs to petition the Ministry of Finance if it needs public funds to liquidate troubled financial institutions. Choy (1997) notes that these limitations may change the institutional form of supervisory and regulatory policy, but not its substance. However, other pressures might lead the FSA to take a new and inde- pendent approach to regulatory policy given that the old model based on insularity, mutual support, and restraint between the Ministry of Finance, the finance industry, and politicians has not adapted to the new demands and challenges presented by the technological and economic environment of the 1990s. In addition, the FSA is entirely focused on regulatory and supervisory policy, and in principle will not have to deal with the multi- tude of often conflicting objectives the Ministry of Finance faced. Resolution of Current Problems The Big Bang and recent legislative efforts, if they succeed, will move Ja- pan toward a more open and competitive financial structure that will ben- efit both Japan and the rest of the world. Unfortunately, they do little to confront the current problems. To date Japan has shown little willingness to depart from its extensive system of deposit guarantees. Other concerns also remain, namely: the manner in which depositors and shareholders have faired in the closing of a number of institutions; the continued reliance on improved economic conditions to solve the nonperforming loan problem; the transparent role of politics in dealing with the jusen problem; and, most important, the government's inability to convince the public that public funds will be needed to deal with current and forthcoming financial problems. Conclusion Whatever the ultimate division of responsibilities among the Japanese government bureaucracy, a central element of any successful reform would be a new approach based on a more explicit and transparent ac- counting, supervisory, and regulatory framework. While the FSA may facilitate introducing and implementing a new policy, the process itself, as opposed tc the specific institution taking responsibility for it, is the most important for successful reform. Successful regulatory reform must resolve the current system's fundamental problems: delays in regulatory responses, moral hazard, and inadequate funding to deal with problem financial institutions quickly. 192 The Banking Crisis in Japan Reform dealing with these issues would naturally entail greater openness in the sense of adopting standard accounting procedures to value the assets of financial institutions and making this information public. It would also entail less informal administrative guidance and greater reliance on explicit rules in the supervision of financial institutions. This approach would reduce the un- certainty inherent in the present system and allow more rapid resolution of financial problems, such as the current nonperforming loan issue, as they arise. Financial institutions would likely benefit from regulatory reforms of this na- ture, and the overall economy -would also benefit, for instance, sectors such as the real estate market would be revived. Foreign financial institutions would also benefit from a more transparent and explicit regulatory and supervisory structure in Japan. Policy changes along these lines would stimulate growth and development in the financial system and help stem Tokyo's recent decline relative to other leading financial centers. This paper has focused on events through July 1997. Recent develop- ments, however, make even our cautionary tale for Japan seem optimistic. The currency collapses and banking crises in Indonesia, Korea, Malaysia, the Philippines, and Thailand that emerged in the second half of 1997 fur- ther weakened Japan's financial system. On January 12, 1998, the Ministry of Finance officially acknowledged what critics have maintained for years: the nonperforming loan problem is far larger than previous official esti- mates had indicated. The ministry announced that in the six months to September 30, 1997, problem bank loans totaled Y 76.7 trillion (US$592 bil- lion at Y 130 to the dollar), and acknowledged that this estimate does not include problem loans at credit cooperatives, insurance companies, and other institutions. Nor does the figure reflect the impact on Japanese institutions from the financial crises in other Asian countries since September 30. The government announced in early 1998 its intent to push for signifi- cant infusions of public money to support the financial system. The Minis- try of Finance's policy position is tenuous, however, in light of recent scan- dals. Several ministry officials have been arrested, and in early March 1998 prosecutors looking into allegations of influence peddling and corruption raided the ministry for a second time. The argument for public infusion of funds to bail out a scandal ridden financial system is likely to continue to face political opposition. References Borio, C. E. V. N. Kennedy, and S. D. Prowse. 1994. Exploring Aggregate Asset Price Fluctuations across Countries. Economic Papers no. 40. Geneva: Bank for Interna- tional Settlements. Thomas F. Cargill, Michael M. Hutchison, and Takatoshi Ito 193 Cargill, Thomas F. 1995. "A U.S. Perspective on Japanese Financial Liberalization." Moretanr and Economic Studies (Bank of Japan, Institute for Monetary and Eco- norric Studies) (May 3): 115-61. Cargill, Thomas F., and Shoichi Royama. 1988. The Transition of Finance in Japan and the United States: A Comparative Perspective. Stanford, California: Hoover Insti- tution Press. Cargill, Thomas F., and Gregory F. W. Todd. 1993. "Japan's Financial System Re- form Law: Progress toward Financial Liberalization?" Brooklyn Journal of Inter- national Law 19: 47-84. Cargill, Thomas F., Michael M. Hutchison, and Takatoshi Ito. 1997. Political Economy of Japanese Monetary Policy. Cambridge, Massachusetts: MIT Press. Choy, Jon. 1992. "Japanese Banks' Self-Help Questioned." Japan Economic Institute Report November 6, no. 42B: 1-3. _ . 1997. "Hashimoto Fills Diet's Docket With Reforms." Japan Economic Insti- tute Report March 21, no. lIB: 4-7. DIC (Deposit Insurance Corporation of Japan). 1995. Annual Report. Tokyo. Economist. 1997. "Banking in Emerging Markets." April 12. Federal Deposit Insurance Corporation. 1994. Annual Report. Washington D. C. Folkerts-Landau, David, and Takatoshi Ito, with others. 1995. International Capital Markets: Developments, Prospects, and Policy Issues. Washington, D. C.: Interna- tional Monetary Fund. Ito, Takatoshi, and Kazuo Ueda. 1993. "Editors' Introduction." In Takatoshi Ito and Kazuo Ueda, eds., "Intemational Comparison of the Financial System and Regu- lations." Journal of the Japanese and International Economies 7 (December): 323-28. Iwamura, Mitsuru. 1993. "Deposit Insurance and Moral Hazard." Monetary and Economic Studies (Bank of Japan, Institute for Monetary and Economic Studies) (July 11): 63-85. Kane, Edward J. 1993. "What Lessons Should Japan Learn from the U.S. De- posit-Insurance Mess?" In Takatoshi Ito and Kazuo Ueda, eds., "Interna- tional Comparison of the Financial System and Regulations." Journal of the Japanese and International Economies 7 (December): 329-55. Lindgren, Carl-Johan, Gillian Garcia, and Matthew I. Saal. 1996. Bank Soundness and Macroeconomic Policy. Washington, D. C.: International Monetary Fund. Packer, Frank. 1994. "The Disposal of Bad Loans in Japan: A Review of Recent Policy Initiatives." Paper presented at the conference on Current Developments in Japa- nese Financial Markets, June 9-10, University of Southern California. 11. Bank Failures in Scandinavia Sigbj0rn Atle Berg During the 1980s and early 1990s each of the Scandinavian countries faced severe banking crises that have had a substantial impact on their econo- mies. The experience could be described as a watershed event, and has been analyzed in many papers and books (see the list at the end of this chapter). The interpretation of events differs somewhat from author to au- thor, and there is no consensus on the relative importance of the large num- ber of factors that contributed to the crises. The Scandinavian countries proper include Denmark, Norway, and Sweden. 'This chapter also treats Finland as part of Scandinavia, but ex- cludes the fifth Nordic country, Iceland, which is too much of a special case to fit into the common framework discussed here. In the four countries discussed here, nearly all the major banks got into serious difficulties as a result of heavy loan losses. Only two of the larger banks, one in Denmark and one in Sweden, escaped serious problems alto- gether. This does not mean that most major banks failed in the strict mean- ing of the word. In reality few outright failures occurred. Most problem banks were either merged with other banks or enabled to continue as independent units through financial support from insurance funds or governments. All large Danish banks and the three largest Swedish banks also managed to get through the crises without receiving direct external assistance. The four countries differed in the nature of their banking crises. Moller and Nielsen (1995) point out some major differences, in particular, be- tween Denirmark and the three other countries. For instance, they stress that mark-to-market accounting rules in Denmark left Danish bankers 195 196 Bank Failures in Scandinavia with fewer opportunities to conceal their problems. While it is true that the Danish banking crises differed in some important respects from the developments in the three other countries, I believe that the similarities between the four countries are more striking. The basic structures of the crises were more or less the same, and proceeded in accordance with well- known theories of financial cycles. One of the special features of the Scandinavian banking crises was that so many of the largest banks incurred losses of a size that could not be dealt with without government intervention (table 11.1). In Finland the five largest banks and in Norway the four largest all received capi- tal injections from the government. In Sweden the three largest banks came through without direct support, but not the next three. Only in Denmark did all large banks manage without government capital injec- tions. However, even most of the large banks that did not receive gov- ernment capital sustained heavy losses. Table 11.2 reports financial support to banks from both governments and bank insurance funds until the end of 1993, when the crises had peaked in all four countries The banking industries are taken to include both com- mercial and savings banks, along with the many small Finnish cooperative banks. Since the distinction between the activities of the groups are becom- ing increasingly blurred, they will be treated as one group. Note that table 11.2 includes only bank-specific guarantees, and not the general guaran- tees issued by the Finnish and Swedish governments. Again, Denmark is a special case, where few banks received support from the government. The Dynamics of the Crises The common structure of the four Scandinavian crises included an initial stage with rapid increases in bank lending (table 11.3). This occurred first in Norway, and a couple of years later in the other three countries. The tax subsidy implicit in the deduction of interest payments from taxable income stimulated the growth in all four countries. While the average growth rates hide much higher rates in a number of individual banks, even the averages were sufficiently high that they should have been dis- turbing to bank managers and regulators. To some extent people did worry about the rapid extension of new loans. However, in Finland, Norway, and Sweden the growth in lending occurred relatively soon after the abolition of important pieces of bank regulation. The details of the deregulatory processes differed, but in each country the authorities felt that banks needed to adjust to an unregulated Sigbjorn Atle Berg 197 Table 11.1. Government Capital Injections, Guarantees, and Ownership of the Six Largest Banks in Finland, Norway, and Sweden, End of 1993 Government ownership Government Government (percentage Country and bank capital guarantees of shares) Finland, KOP 1,700 1,800 14.8 Unitas 1,700 1,000 5.8 PSP 900 0 100.0 SKOP 15,700 800 52.9 OKO 400 0 0 Aktia Savings Bank 0 0 0 Norway' Den nors.ke Bank 7,000 0 87.5 Christiania Bank 8,800 0 68.9 Union Banak 1,000 0 48.0 Fokus Bank 1,900 0 97.9 BN-Bank 0 0 0 Savings Bank of Nord-Norge 700 0 0 Sweden' SE-Banken 0 0 0 Handelsbanken 0 0 0 Savings Bank of Sweden 1,000 0 0 Nordbanken (including Securum) 50,200 0 100.0 F6reningsbanken 0 2,500 0 Gota Bank (including Retriva) 23,800 0 100.0 a. Millions of Fik. b. Milions of NKr. c. Millions of SKr. Source: Bank of Finland, Norges Bank, and Sveriges Riksbank. equilibrium, and that this equilibrium entailed a substantially higher level of bank lending. Initially rapid growth in lending could be rationalized as normal adjustments to a new regime. In Denmark most of the banking regulations had been abolished ear- lier, and sorne of the adjustments to a new equilibrium had been made during an economic recession. Only when economic conditions improved did bank lending really take off. In the other three countries the deregula- tion of banking coincided with high levels of economic activity, which acted as a further spur to bank lending. X198 Bank Failures in Scandinazvia Table 11.2. Capital Injections or Disbursements to Banksfrom Insurance Funds or Governments, 1988-93, and Bank Structure before the Crises Total equity Total Total Number of capital of all nutmber of disburse- receiving banks, end banrks, end ments, banks, Country of 1988a of 1988 1988-93 1988-93 Denmark (millions of DKr) 65,098 206 5,092 7 Finland (millions of Fmk) 22,760 590 54,278 278 Norway (millions of NKr) 22,831 187 24,912 25 Sweden (millions of SKr) 19,414 525 65,000 3 a. Excludes subordinated debt. Source: Scandinavian central banks. One important component of deregulation that permitted the growth in bank lending, but which has not drawn much attention, was the withdrawal of some foreign exchange regulations. For most of the postwar period all the Scandinavian countries except Denmark had heavily regulated capital flows. In Finland and Norway the remaining regulations that prevented banks from borrowing funds on international capital markets were lifted before the strong growth in bank lending took place.' This deregulation did not have much immediate effect, but international markets proved to be an important source of funding when bank lending took off. Without this access to foreign fund- ing the rapid growth in lending would have been impossible. The banking industries in Finland, Norway, and Sweden became heavily dependent on foreign funding. The ratio between domestic deposits and borrowing and domestic lending also declined in Denmark, and while the Danish banking industry did not become dependent on net foreign funding, the Danish bond market did. While free capital flows are important for long-term economic efficiency, the Scandinavian banking crises illustrate that they have the po- tential to increase the volatility of small, open economies. In all four countries the structure of liabilities on balance sheets was changed during the credit expansion, with relatively less weight given to ordinary deposits and more reliance placed on money market borrowing. 1. In Norway banks were free to obtain foreign funding from 1978 and in Finland from 1986. Sweden did not lift formal restrictions until 1989, but banks were in prac- tice free to obtain foreign funding before that time. Sigbjorn Atle Berg 199 Table 11.3. Growth in Lending Rates by Parent Banks, 1980-96 (percent per year) Year Denmark Finland Norway Sweden 1980 - 18.0 12.7 - 1981 - 14.9 16.7 10.8 1982 - 18.8 17.3 14.8 1983 - 14.5 18.4 9.3 1984 - 14.1 27.5 9.7 1985 - 17.6 31.8 3.4 1986 - 11.9 33.0 15.4 1987 15.0 19.1 18.7 15.9 1988 8.1 31.1 6.1 31.5 1989 10.0 15.2 9.3 25.7 1990 11.5 11.2 3.9 16.6 1991 0.8 1.4 -4.4 -1.5 1992 -7.0 -3.0 1.0 1.7 1993 -1.3 -13.4 0.5 -19.1 1994 -12.6 -8.3 4.1 -3.1 1995 3.5 -6.5 9.1 -3.7 1996 12.0 -0.7 13.9a 3.9 - Not available. a. The growth rate shown is artificially high because during the year some nonbanks were transformed into banks Source: Scandinavian central banks. In this sense the banks' funding base became less robust. In particular, the availability of foreign funding proved sensitive to macroeconomic devel- opments anid other systemic factors. Most of the foreign funding evapo- rated when the signs of an imminent crisis appeared. The regulatory regimes included restrictions on interest rates and lend- ing volumes. These regulations had a cementing effect on the market shares of individual banks. Deregulation was thus also seen as a unique opportu- nity to captLre a higher market share, especially by the larger banks. This perception certainly contributed to the growth in overall lending, and is the only possible explanation for the fact that some banks extended loans at rates below their marginal funding costs. Furthermore, an unusually large number of b