; > Econn iJc Development Institute I;;7 of T!l. ; ; eiurk1 D ank< cf _/9 andLiA V 100 ' i er ' 0 T H i i ~ ~~~~~~~~~~~~~~11" anJ" tit, Edited SRy j Shaki1Fl A EDI SEMINAR SERIES. EDI SEMINAR SERUS Financal Sector Reforms, Economiic Growth, and Stability Expeiences in Selected Asian and Latin American Countries Edlited by Shakil Faruqi Contributos Suman K. Bery Ross Leine Binbadi Cados Mad Chon Tai Clung Adrianus Mooy Michael Dooley Rainaldo Penn Mobammad Asbra Janua Alan P. Roe Takashi Kanlci J. B. Sumadin Ekamol Kiriwat Edgardo P. Zialcit Guillermo Le-Fort The World Bank Washing D. C 01994 The lternational Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433. USA All rights reserved Manufactured in the United States of America First printing October 1994 The Economic Development lnstitute (EDI) was established by theWorld Bank in 1955 to train officials concernedwith development planning, policymaking, investment analysis, and project implementation in member devdoping countries. At present the substance of the EDI's work emphasizes macroeconomic and sectoral economic policy analysis. Thmugh 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 expericnce 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 represenL Because of the informality ofthis series and to make the publication availablewith the leastpossible delay, the manuscript has notbecnedited as fully as would be the casewitha more formal document, and theWorld Bank accepts no responsibility for enrms. Some sources cited in this paper may be informal documents that are not readily available. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. T.he World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposcs, without asing a fee. Permission to photocopy portions for classroom use is granted through the Copyright Cleance Center, Inc, Suite 910,222 Rosewood Drive, Danvers, Massachusetts 01923, U.SA Thecompletebacklistofpublications from the WorldBank isshown in the annualIndeaofPnblctons,which contains an alphabetical title list (with full ordering information) and indexes of subjects, authors, and countries and regions. The latestedition is availablefrc of chargefrom the Distribution Unit, Office ofthe Publisher, The World Bank, 1818 HStreet, NW. Washington, D.C 20433, USA, or from Publications, Banque mondiale, 66, avenue d'Iena, 75116 Paris, France Shakdl Faruqi is principal financial economist in the Fmance and Private Sector Development Division of the World Bank's Economic Development Institute.- Library of Congres Cataloging-iu-Publieation Dawn Financial sector reforms, econmic growth, and stability: experiences in selected Asian and Latin American countries e edited by Shalol Faruqi ; contrbutors Suman Bey . . . [et al-]. p. cm.-(EDI seminar scries) Indudes bibliographical references. ISBN 04213-3D13-6 1. Fmancial institutions-Government policy-Asia-Congresses. 2. Fmanial institutions-Govcrmment policy-Latin America- Congrsses. 3. Banks and banking-Asia-State supervision- Congresses. 4. Banks and bankcing-Latin America-State supervision-Congrsses. 5. Monetay policy-Asia-Congresses. 6. Monetary policy-Latin America-Congresses. 1. Faruqi Shakil. U. Bery, Suman IL IlL Economic Development Institute (Washington, D.C) IV. Series. HG187.A2F573 1994 3321'095-dc2O 94-33523 CEP CONTENTS Foreword v Abbreviations vu 1. Opening Address I Adriarms Mooy 2. Seminar Proceedings: A Summary Report 9 Alan R. Roe 3. Financial Reforms, Opening Up, and Stability: Policy Issues and Sequencing Considerations 61 Carlos Massad 4. Financial Deregulation and Bank Supervision: The Case of Indonesia 81 B i3adi 5. Globalization, Speculative Bubbles, and Central Banking 103 Michael Dooley 6. The Financial System and Macroeconomic StabilitY: The Chilean Experience 113 Guillermo Le-Fort 7. Securities Market Regulations and Reforms in Thailand 139 Ekamol Kiriwat 8. Pakistan's Liberalization of the Extenal Sector 145 Mohammad Ashraf Jaua 9. Financial Liberalization in an Agrarian Econmy: The Case of Paraguay 153 Reinaldo Penner iu iv 10. Japanese Banks, Speculative Bubble, and Bank Supervision 179 Takashi Kanzaki 11. Government Insurance and Financial Intermediaries: Issues of Regulation, Evaluation, and Monitoring 193 Ross Leine 12. Capital Account Liberalization: The Philippines Experience 225 Edgardo P. Zialcita 13. India: Commercial Bank Reform 243 Suman K. Bery 14. Financial Sector Reforms and Liberalization in The Republic of Korea: Current Status and Prospects 261 Choon TaLk Chung 15. Closing Address 271 J.B. Swuarlin FOREWORD This publication consists of the papers prepared for a senior policy seminar held in February 1993 at Bali, Indonesia. The seminar was organized by the EDI in collaboration with Bank Indonesia and with the support of the Government of Japan. Shakil Faruqi of the EDI was the seminar director. This was the second seminar in a two part mini-series designed to share comparative experiences with financial sector reforms and issues of growth and stability in selected Asian and Latin American countries. The countries represented at this seminar were: Bangladesh, Indonesia, India, Japan, Malaysia, Pakistan, the Philippines, Thailand and Viet Nam from Asia; and Chile, Colombia, Mexico and Paraguay from Latin America. A distinguished group of ministers, central bank governors, their deputies and senior officials from these countries participated in this seminar The first seminar in this series was held in May 1992 at Economic Commission for Latin America and Caribbean (ECLAC), Santiago Chile and was organized in collaboration with Central Bank of Chile and with the support of the Government of Japan. The proceedings of this seminar were published earlier as Shaldl Faruqi (editor), Fiancial Sector Reforms in Asian and Latin Anerican Countries: Lessons of Comparative Experience, EDI Seminar Series, No. 340/073, Washigton D.C., 1993. These seminars were a part of EDI's ongoing program in the financial sector articulated around cycles of regional and worldwide roundtable conferences and seminars. The program is managed by Xavier Simon, Chief, Finance and Private Sector Development Division of the EDI. I would like to thank Nicholas Hope, and Andrew Sheng of the World Bank, Syabril Sabirin and senior officials of Bank Indonesia for their support of the seminar. The manuscript was edited and prepared for publication by Shakil Faruqi. The papers were copyedited by Jeannie Massie, Alice Dowsett and associates. The task of preparing and processing the text for publication was done by Elizabeth Crespo with skill and patience. Anmon Golan Director Economic Development Institute vp ABBREVIATIONS AAB authorized agent banks AFP pension fund agencies ALADI Latin American Integration Association ARF asset reconstruction fund ATM automatic teller machine BIS Bank for International Settlements BLR base lending rate BNF National Development Bank, Paraguay BNM Bank Negara, Malaysia BNT National Workers Bank, Paraguay CAMEL capital, asset, management, earnings, and liquidity CB central bank CD certificate of deposit CPI consumer price index CRR cash reserve requirement DBC dollar bearer certificate DRI dollar rate of interest DTC deposit taking cooperatives EDI Economic Development Institute EFF Extended Fund Facility FCA foreign currency account FCD foreign currency deposit FCDU foreign currency deposit units FDI foreign direct investnent FEBC foreign exchange bearer certificate FMC foreign manufacturing company PSR financial sector reform GATT General Agreement on Tariffs and Trade vii viu Abbreviations GDP gross domestic product GNP gross national product HCI heavy and chemical industries IFS international financial statistics IMF Intemational Monetary Fund IPO initial public offering LRM Letras de Regulacion Monetaria MAERS market average exchange rate system NAFTA North American Free Trade Agreement NBFI nonbank financial intermediaries NCI net capital inflow NIT National Investment Trust PDS Philippine Dealing System QR quantitative restrictions RBI Reserve Bank of India SAPV Savings and Loan Association of Paraguay SLA savings and loans associations SRI Securities Bank Indonesia SBP State Bank of Pakistan SBPU Money Market Securities, Indonesia SEC Securities and Exchange Commission SET Stock Exchange of Thailand SLR statutory liquidity ratio WPI wholesale price index OPENING ADDRESS Adrianus Mooy, Governor, Bank Indonesia On behalf of the Government of Indonesia, let me extend to you a warm welcome to Bali. It is our hope that you will find the setting for this seminar conducive to a constructive discussion and a fruitful outcome. Your agenda for this seminar promises to be demanding for the next four days, but I trust you will be able to spare some time to enjoy the natural beauty of this island and experience the richness of its unique culture. First I wish to express my appreciation to the Economic Development Institute of the World Bank for taking the lead in organizing this seminar and to the Government of Japan for its support. We at Bank Indonesia feel honored to be associated with this endeavor and we can only hope that the experience gained from this seminar will encourage further collaboration among developing countries on this and other topics of interest. Perhaps it was not accidental that Indonesia was chosen as the host for the second seminar on Financial Systems Reforms in Asia and Latin America. Like our friends in Chile, who hosted the first seminar, we in Indonesia have undertaken major fnancial reforms over the past decade. I have had the personal satisfaction in my capacity as governor of Bank Indonesia over the past five years to have been direcdy involved in the formulation of policy and its implementation. It should come as no surprise if I say that carrying out financial reforms does not come easy or without pain. Let me cite an example from our recent experience. Following the 1988 financial liberalization, there was an explosion in banking sector activity in Indonesia; the number of banks and branches mushroomed in the euphoria of the time, but banks had not prepared the necessary infrastructure to cope with the volume of credit that followed. The consequence was an overheated economy during 1989-90, and the government responded by tightening financial policies. These policies, together with the introduction of prudential regulations to improve the soundness of banks, exposed the full extent of the weaknesses of banks that had indulged in excessive lending in earlier years. In inaugurating this seminar, I am mindful of the progress achieved at the proceedings of the Santiago seminar. Those proceedings demonstrated a wealth of information available for analysis on financial system reforms undertaken over the past 2 Adrlanut Mooy decade in a number of Asian and Latin American countries. The specific reform packages vary from one country to another, as do the-policy responses of individual countries. This diversity of experiences in financial reform provides lessons from which the countries engaged in reform effort could benefit, and which could be especially useful as models for those countries, that may be contemplating such reform. The task for this seminar is to build on the useful work already completed in Santiago, and .1' feel confident this seminar will contribute significantly to further our understanding of the issues involved and lessons to be drawn. Before I discuss the reforms in the financial sector in Indonesia, allow me to touch briefly on some aspects of adjustment in the developing countries. Since financial reforms in developing countries generally constitute an integral part of an adjustment strategy, a proper understanding of the adjustment strategy adopted for each country provides a useful framework for assessing the substance of the financial reform itself and analyzing its impact on real economic activities. - The adjustment policies undertaken by developing countries over the last decade have been motivated by a need to restore domestic economic stability and to strengthen the capacity of the economy to deal with external shocks, be they -favorable or otherwise. For Indonesia, external shocks came mainly from.two sources: changes in global demand associated with variations in world economic activity and swings in the international price of oil. Indonesia's vulnerability to these shocks were reflected in sharp fluctuations in its foreign exchange earnings and government revenues, with implications for policy responses to ensure macroeconomic stability. Heavy reliance on a single commodity. (oil), red tape, and overregulation, created impediments to achieve development objectives. = For Indonesia, the strategy embodies three national development objectives (the Trilogy Pembangunan) comprising equitable development, a sufficiently high level of economic growth, and a dynamically stable environment. In turn, our adjustment strategy consists of four intermediate targets: first, increased grass root participation of the public in the development process; second, acceleration of export growth, especially of non-oil and -gas, through outward-lookdng strategies to tap export opporunities in world markets; third, a more diversified, broad-based and balanced economic structure to enhance our capacity to deal with external shocks; and fourth, improvement in the quality of development, including economic efficiency and the- quality of human resources with a view to upgrading the quality of life. In developing countries, policy adjustments covering- various sectors are generally required to be undertaken simultaneously. This is especially the case in such- areas as pricing and-tariff policy, as well as exchange rate policy. In Indonesia, adjustment was initially directed at liberalizing prices and the exchange rate as well as promoting non-oil and gas exports.' This was subsequently followed by banldng deregulation (1983 and 1988), tax reforms (1984 and 1985), and at various times and in stages other deregulations, including trade, industry, and investment. OpewMg Addrmss 3 A closely related aspect of adjustment is how to minimize any adverse consequences on thle vulnerable segments of the population. Experience has varied from one country xo another. Some countries that initially resisted going through adjustment to avoid the negative short-term consequences found themselves in greater difficulties that ultimately forced them to take even more drastic measures later. The consequences of a delay in needed adjustment resulted in disruptive social unrest and disorder. In Indonesia the government has pursued a policy of responding promptly when the need for adjustment arises. In my view, this policy has served us well by helping to promote long-term growth and social harmony. Let me now turn to financial sector reform. Past experience has shown that financial deepening is important in improving and sustaining economic growth in developing countries, and that poor economic resilience against extemal shocks is basically due to shallow finance, or fiancial repression, characterized by excessi-:e regulation and control of interest rates and exchange rates. Interest rates fixed below equilbrium levels inhibit savings mobilizauon, if allowcd to reflect market forces, higher rates will create incentives for depositors to save. In this case the adjustnent policies in ft financial sector will bring about a more efficient allocation of fumds, whereby increased demand for i.cstment can be satisfied by the higher savings generated. A regime of artificially controlled low intest rates closely characterized-what existed in Indonesr s before 1983. But in June of that year the government embarked on a reform procs designed to create a climate conducive to healthy competition among banks. The aim was to improve market mechanisms in ways that could lead to higher efficiency in savings mobilization. The liberalization involved not only the determition of interest rates but also the sucture of the financial institutions themselves. Formerly, intest rates on both the deposit and lending sides were detmined by Bank Indonesia by fixing rates for state banks, which accounted for the bulk of bank assets (60 percent). After June 1983 the determination of interest rates was left to each bank's discretion, relying more on market fres. The results were soon to follow with a threefold increase in saving over the next five years. Although the market mechanism had improved, strucural weaknesses stdll existed that inhibited the growth of a sound and efficient financial system. For example, the distrbution of bank services was very limited, the efficiency and soundness of banks needed improvement, and nonbank financial institons were -not well developed. Therefore, in October 1988 we introduced fiurter reforms aimed particularly at widening and improving the instituonal srucre of the fincial system. This was primarily achieved by pernitting entry of pnvate banks, including foreign banks in a joint venture with national banks, and of rural banks, and by allowing the establishme of foreign-exchange banks as well as the openng of branch offices. An increase in the number of banks can induce efficiency thgh enhaned comptition, thereby enabling the attainment of interest rates at a level that encourages savings and business activities. Widening distnbution of banking services throughout 4 Adrlanus Mooy the country can promote an active role for banks to accelerate economic growth and its regional distribution. Furthermore, opening up foreign and joint venture banks as well as foreign branches of domestic banks can promote exports, especially non-oil comnodities. On January 29, 1990, we introduced further financial reforms to improve the credit system as well as enhance Bank Indonesia's role in monetary control. Bank Indondsia's liquidity credits were restricted to only four priority areas, that is, food production, food procurement, cooperatives, and selective long-term investment. By limiting their scope, Bank Indonesia has been able to minimize the impact of liquidity credits on domestic liquidity and inflation. The credit structure of banks has also been improved as banks diversified their activities to finance projects previously funded by liquidity credits. Furthermore, because interest on liquidity credits was subsidized, their diminished role strengthened the market mechanism in interest rate determination. The rapid development of the banking industry in Indonesia has exposed a need for modernizing banking techniques, with emphasis on human resource development. In addition, the Indonesian banking sector also must adapt to the internationally prevailing regulations to face the rapid trend toward globalization of bani.ng services. Therefore, in February 1991 we issued a package of prudential regulations for banks. Inportant elements of the package include the following. * First, banking supervision was to improve early detection of problems and the disclosure of banks' financial condition based on objective criteria. * Second, banks were to observe prudential practices of international banking standards, such as compliance with the minimum capital requirement as stipulated by the Bank for International Settlements (BIS), provisions concerning legal lending limits, and maintenance of adequate reserves to cover risk assets. * Third, certain provisions were imposed for owners, boards of directors, and boards of commissioners of banks, concerning moral standards and banking experience. * Fourth, banks were to undertake human resource development by providing a training fund of at least 5 percent of their personnel budget. Finally, in March 1992 the government launched the new Bank Act, which provides a framework for the operation of the banking industry. It also set the general objectives of the national banlcing system, namely that its activities should support national development. Bank categories are simplified into commercial banks and rural banks, and the scope and activities of each category are clearly defined. The law also Openng Addrs 5 stipulates basic requirements for the establishnent of a new bank in greater detail and reterates the mandate of Bank Indonesia over bank examination and supervision. The thrust of the series of policy packages to reform the banking sector in Indonesia is directed to bring about higher efftciency in the mobilization and channeling of funds and to support the bottom-up development strategy based on equity, growth, and stability. Given this responsibility, the healthy development of a sound and reliable 'banldng system needs to be supported by higher professionalism through training and strengthened bank supervision. This brief elaboration is to stress that refonms in the baning sector in Indonesia constitute an integral part of our adjustment strategy. For us, adjustment is an ongoing process cautiously designed and introduced in a timely manner. The next issue relates to the sequencing of adjustment policies in the various sectors, especially foreign exchange system liberalization in relation to domestic financial sector reforms. On this matter, it is generally recommended that the liberalization of domestic markets should precede the liberalization of the external sector, because the benefits of reforms in product markets cannot be realized if factr mobility is significantly impaired. Furthermore, it is also argued that the hiberalization of the current account should precede the liberalization of fte capital accounts, because asset markets generally adjust much faster than commodity markets. Thus, the general view is that capital account liberalization should follow current account liberalizton, and both should be preceded 1v domestic financial sector reforms. In this regard it is interesting to note that the sequencing of liealizaton in Indonesia appeared to have deviated significantly from the normal route. Indonesia's liberalization of the foreign exchange system was introduced in 1970, long before deregulation in the trade area. With the buoyancy of oil revenue, the trade policies during most of the 1970s were still restrictive. It was only in the second half of the 1980s that significant deregulation of trade began. Furthermore, the movement toward a free foreign exchange system also came much earlier than the deregulation in the financial and bankdng sectors. As indicated earlier, it was only in June 1983 that the ceilings on the deposit and lending interest rates of the state-owned banks were lifted. Wher or not the sequence of policy lralization applied in Indonesia is an approprate model for others to follow is of course open for disussion. Our experience indicated that this sequencing contributed significanty to our economic development, especially in opening up the foreign sector. An alternative route may have resulted in less satisfactory result for us. However, applying our experience to other counties may have to be qualified for the following reasons. First, our liberalization of the foreign exchange system in 1970 was part of a comprehensive package of economic policy, of which adjustmen in the domestic interest rate was an ingral part The consequent increase in the domestic interest rates supported by a successful art.,:-inflation policy during much of the 1970s helped prevent large capital outflows. Second, soon after the new order goverment took office, a comprehensive economic policy package significantly improved public confidence in economic 6 Adriaiws Movy management. This effort helped to avoid destabilizing Specuion and served as an important factor in the success of the liberalization of our foreign exchange system, which encompassed liberalization of both the current account as well as the capital account. Essentially, the sequencing issue of capital account in relation to the current account in Indonesia was not as clear-cut as it may have appeared. Our experience in Indonesia illustrates that although the broad elements of the adjustdient policies in developing countries may appear to be similar, the appropriate speed and sequencing of the liberalization process may vary from one country to another. The magnitude of the initial imbalances, the comprehensiveness and baance of the policy package, and the political and social circumstances in each country serve to fashion the specific actions that are suited to each country's circmstan. The experience of developing countries suggests that the intaction among adjustment policies is complex, thus we should be very cautious m making broad generalizations. The Indonesian economy is now reaping the benefits frm the strucual adjusmts that have been implemented in the course of the past several years. The economy has been expanding at an impressive rate amid price stability. The rate of growth was 5.7 percent in 1988, 7.4 percent in 1989, and 7.3 percent in 1990, while the rate of inflation was 5.47 percent in 1988, 5.97 percent in 1989, and 9.53 percent in 1990. Moreover, the economy has become progressively diversified and has moved toward a more balanced structure. It has achieved impressive growth in the manufacuring sector, which now accounts for nearly four-ffths of our non-oil and gas exports. Overall, economic development has been supported by a strong growth in non-oil exports through private investments, ample funds from banking credits, and equity participation through the stock market. I should underscore, however, the success of our economic adjustments has not been achieved without problems. In 1989, unconstained demand for goods and services, primarily for investment and consumtion, rapidly exceeded the supply capacity of domestic production and imports. The inflation rate rose sharply, almost reaching double-digit figures in 1990. Significanty higher investment activity, especially in 1989 and 1990, also widened the savings gap, which was also reflected in a deterioraton of the current account deficit in 1990. The situation was aggravated by ivestments in costly projects with long gestation periods and by bottlenecks in the supply of public utilties and other infrastructures. The widening inteal and extral imbalances ultimately became a threat that required prompt response. Growing foreign borrowing t finance the current account deficit exerted pressure on our balances of payments with implications for our capacity to continue to service our extemal debt I would descnbe these developments as the problems of success that are to be found in one degree or another in any economy adjustng to cycles associated with waves of investment following strucural adjustment policies. To cope with these problems, in 1990 the government took measures to cool down the overheated economy and restore internal and extera imbalance. We had to move swiftly to curb the expansion of domestc demand by introducing tight financial Opening Address 7 policies. Monetary policy was tightened in mid-1990, followed by a rise in interest rates. These measures were reinforced by strict management of government finances as reflected in the establislunent of a development budget reserve to help neutralize liquidity expansion from increased earnings from oil during the last two years. Furthermore, in September 1991 the government announced a ceiling on foreign commercial borrowing and set up a committee to coordinate and supervise commercial borrowing activities, especially by state enterprises and banking insdtutions. - These measures succeeded in containing inflation- in 1991 at 9.52 percent. Although real GDP rose by an estimated 6.6 percent for 1991, the growth of real domestic demand fell to 3.4 percent from 12.2 percent in 1990. In US dollar terms, import growth slowed markedly to about 15.7 percent from 31.6 percent in the previous year. This was accompanied by a rebound of non-oil export growth to 22.2 percent The resulting increase in foreign sector activity has helped maintain economic growth in 1991 at the relatively high rate of 6.6 percent. We are proud to tell you that as a result of contined prudent financial policies, the Indonesian economy is in the "soft-landing' mode. In 1992 non-oil exports grew by 21.1 percent, which not only helped improve the balance of payments but also sustained economic growth at about 6.0 percent Import growth fell further to 7.1 percent in 1992 from 15.7 percent in 1991. These developments helped reduce the current account deficit from US$4.4 billion in 1991 to US$3.8 billion in 1992. The inflation rate was more than halved; it fell from 9.52 percent in 1991 to 4.94 percent in 1992. Durig our twenty-five years of development, we were able to keep inflation rates below this level on only two occasions. This is a brief review of our experience in formulating and implementiog adjustment policies, with particular emphasis on the reform of our financial system. We are proud that these adjustment policies have served us well by sustaining a growth momentum that has left a visible impact on the welfare of our people. We are aware, of course, that adjustment policies are an ongoing process, but I think it is fair to say that our successful experience has bolstered our confidence and strenghtened our resolve to stay "the course' as we face the challenges of the -future. Thank you. 2 SEMINAR PROCEEDINGS: A SUMMARY REPORT * Alan R. Roe, University of Warwick Financial sector reform and the subsequent management and regulation of the sector is an extremely important aspect of economic policy in many of the countries supported by World Bank loans and technical assistance. The distillation of the lessons from such reforms has become an important element in the work of the Economic Development Institute (EDI) in recent years. This work, in turn, has shown that although the component problems ofien differ from country to country, there is a great deal of commonality across countries and even continents. This paper addresses the latest initiative of the EDI in this field of work and summarizes the papers and associated discussions presented at the Senior Policy Seminar held in Bali, Indonesia from February 8 to 11, 1993. The seminar was the second one organized by EDI on the general theme of comparative financial reform experiences in Asia and Latin America; the initial seminar held on this topic was in Santiago, Chile, in March 1992.1 Participants included ministers, central bank governors, and senior officials from a total of hrteen Latin American and Asian economies. The counties represented were Bangladesh, India, Indonesia, Japan, the Republic of Korea, Malaysia, the Philppines, Thailand, and Viet Nam from Asia; and Chile, Colombia, Mexico, and Paraguay from Latin America. The seminar was supported by the Goverment of Japan, organized by the Economic Development Institute of the World Bank in collaboration with the Bank of Indonesia, and directed by Shakil Faruqi of the Economic Development Ibstitute. As m the earlier seminar, the main purpose was to assess comparatively the wide range of country experiences with financial reform in the two regions and to attempt to identify any general lessons. On this occasion, however, a second objective was to assess the o-verall consequence for economic stability of reform measures in somewhat greater depth than had been possible in Santiago, where a broader topic had been adopted. Thus, a good deal of attention was directed toward the question of the sustinability of large capital inflows now being experienced by several participating countries and the consequences and management of financial bubbles. The topics of 1. The proceedings of the earlier semiinar bave now been published as Shakil Faruqi (editor), Fruancial Sector Refonns in Asia and Latin American Countries: Lessons of Comparatfve Experene, EDI Seminar Series, No. 3401073 Washington DC., 1993. 9 10 Aan R. Roe bank regulation and supervision were also accorded greater prominence than in the previous seminar. Various speakers presented papers that supported discussions on a variety of sub-themes, including the problems and possibilides created by the liberalization of capital accounts; the issues arising from financial deepening and the regulation of securities markets; the impact on bank stability and the management of real estate and other financial bubbles; and the increasingly important role of strong bank supervision as a support structure for the overall reform of financial sectors. The presentations during the four days included papers on experiences in some or all of these areas in eight of the countries represented, as well as a number of papers of a more general and synoptic nature. Opening Address Several of the main themes of the seminar were identified and elaborated on in the opening address by Adrianus Mooy, Governor of the Bank Indonesia, the Indonesian central bank. Mooy noted that financial reform in his country, as in many others, had not been achieved without some pain. In particular, the explosion of banldng sector activity that followed the financial liberalization of 1988 had occurred without the proper establishment of a supporting infrastructure of regulation. Hence, one consequence had been a seriously overheated economy by 1990 and difficulties for banks that needed to be addressed eventually through a major strengthening of the country's system of prudential regulation and supervision. This was now underway. (See also paper 4 on Indonesia.) Mooy pointed out that the adjustnent policies undertaken by many developing countries in recent years had been motivated by an anxiety to build up greater resilience to external shocks. Indonesia was no exception. There the most damaging shocks had normally been associated with general swmgs in the levels of world economic activity but more importantly with large movements in the internatonal price of oil. Indonesia's vulnerability to these shocks had often resulted in large and disruptive swings in foreign exchange earnings and government revenues. Hence, a major component of the adjustment effort had involved attempts to achieve a more diversified, broad based, and, above all, outward looking productive structure to ensure that this vulnerability could be lessened significantly. This, however, had to be achieved in the context of the country's other development objectives, which included increased participation of the general public in the development process and a general improvement of the quality of development, such as giving greater attention to improved human resource capabilities. Mooy felt that his country's decision generally to respond promptly to adjustment needs as they arose had helped both the overal development of the country as well as the promotion and mainenance of equity and social harmony during the process. Seminar Proceedigs: A wmwy Rrep 11 As for the content of his own country's financial reform efforts, Mooy pointed out that prior to 1983 Indonesia's situation closely mirrored the McKinnon-Shaw representation of shallow finance caused by excessive regulation associated especially with controls on interest rates and exchange rates and, more generally, with financial intervention. The liberalization of interest rates and the fostering of a more competitive banking system began in Indonesia in June 1983 and led quicldy to a substantial improvement in the country's savings performance and in the efficiency of use of available savings. But the financial system remained narrow. In October 1988 the government enacted further reforms to widen the system and generally improve its structure. This was done mainly by permitting the entry of new banks, especially through joint venture arrangements with foreign banks; establishing foreign exchange banks; and expanding bank branching. These measures were extended further in the early 1990s when steps were taken to limit Bank Indonesia's direct involvement in the provision of liquidity credits to priority uses, thereby limiting the central bank's impact on the expansion of domestic liquidity and according it greater control over monetary and inflationary pressures. The rapid changes in the size and the roles of the Indonesian banking system that followed these reforms, together with the chalenges associated with the parallel globalization of many banlkng activities and regulations, exposed the need for the modernization of banking techniques, including the control methods used by the monetary authorities. These matters and others were addressed in the frther reform measures of February 1991, which redefined the prudential framework of banks in severa main areas. First, it substantally strengthened bank disclosure and bank supervision to ensure that the authorities were better equipped to detect problems earlier. Second, it introduced internationally accepted standards in the areas of capital adequacy, legal lending limits, and reserve cover for risk assets. Third, it defined new and tighter standards for the owners, directors, and managers of banls. Fourth, it recognized the human resource implications of the new environment by introducing a mandatory training obligation linked to a training fund equal to 5 percent of each bank's personnel budget. Finaly, in March 1992, the overall framework for banidng activity in the country was defined in the new Bank Act. Cumulatively through these various reforms, Mooy felt that the Indonesian banking industry had been put in a position to support the country's bottom-up development strategy by ensuring greater efficiency in both the mobilization and channeling of fumds to all sectors of economic activity and all segments of the popuion. Certainly the reforms that had been introduced had confirmed the authorities' belief that financial reform and modernization were integral parts of the country's overall adjustment effort. On the matter of sequencing the various component reforms, Mooy noted that Indonesia appeared to have deviated significandy from the conventional approach to this matter. The general view is that capital account liberalizaton should follow 12 Aln R. Roe current account and trade liberalization, and that both should be preceded by domestic financial sector reforms. Indonesia, however, had significantly liberalized its capital account in 1970, many years before deregulation began in the area of foreign trade, which began only in the second half of the 1980s. Furthermore, the deregulation of the foreign exchange system had come much earlier than the deregulation of the domestic banking and financial sectors, which, as just noted, remained relatively restricted untl about 1983. Although this unusual pattern of reform worked well for Indonesia, Mooy did not expect that the experience could carry over easily in other country contexts. There were several reasons for this. First. foreign exchange liberalization in the 1970s was part of a comprehensive policy that had also included significant upward movements of interest rates. This, together with a generally successful set of and-inflation policies at that time, had both prevented large capital outflows Cm spite of the still-restricted domestic financial system) and had snmulated much greater public confidence in the government's overall management of the economy. In addition, the 1970 hberalization of the capital account had actually involved a significant relaxation of current account controls. Thus, the apparent separation of the current and the capital account reforms was not nearly as clear-cut as it may have appeared- More generally, Mooy argued, the interaction among different aspects of an overall adjustment program was extremely complex and often country specific. Hence, it was somewhat hazardous to make broad generalizations about issues, such as appropriate sequencing. Indonesia has certinly reaped considerable benefits from the numerous reforms implemented during the past few yeara. Growth rates m the recent past, for example, have been between 5.7 and 7.4 percent, and inflation has been contained at the single digit level. Even more significant, the diversification necessary to provide better insulation from global shocks has certainly been achieved: the manufacturing sector now accounts for almost as large a proportion of total exports as does oil and gas. This diversification has been driven by private investment and the availability of ample credit from the domestic banldng system. The overheated economy in the late 1980s, already rererred to, had created considerable difficulties for the authorties. Although associated with the country's success in stimulating new investment, it was nonetheless manifested in unsustainable external deficits and an accelerating upward movement of inflation. The government, however, had moved quickly to correct the imbalances. This had been done by enacting a tighter monetary policy in mid-1990 that included higher interest rates and a significant tightening of the government's own expenditures, especially on development projects. Although this caused a major slowdown in the growth of total domestic demand, especially in 1990, Mooy noted that the good performance of export activity meant that overall growth in both 1991 and 1992 continued to be high, at rates of 6.6 and 6.0 percnt, respectively. This was achieved with a large fall in the external current account deficit and a decline in inflation to less than 5 by 1992. S&rminar Proceedings A Sxrnary Report 13 In replying to Governor Mooy's introductory remarks, Guillermo Ortiz from Mexico addressed four main points. First, he agreed with Mooy that financial reform needed to be analyzed within the general context of a country's overall adjustment effort. Mexico, like Indonesia, had needed to diversify from an excessive dependence on oil exports. Subsequently, it was very successful in this regard, having reduced the share of oil in total exports from over 70 percent in the early 1980s to only 25 at the present,time. Second, the traditional price stability of Mexico through the 1950s, 1960s, and part of the 1970s had been seriously interrupted until the end of the 1980s. An important consequence had been low or negative real rates of interest, which resulted in a considerable decline in the importance of the country's financial sector relative to GDP. This was reversed only by the reforms of the late 1980s. Third, Ortiz argued that the experiences of Indonesia, Mexico, and other countries clearly demonstrated the importance of fiscal correction, including a far more disciplined use of central bank rediscounting as a major precondition for the successful liberalization of interest rates. Ortiz felt that a further imnportant step was the simplification of systems of reserve requirements and, for Mexico, the elimination of the multiple reserve requirements and discount rates formerly applied to different sectors. Finally, Ortiz agreed with Governor Mooy that institutional reforms linked to a more competitive banking system constitute a very important second stage of overall financial sector reform. In the case of Mexico, strengthening competition, privatizing several banks, and generally openiLng up the financial system were associated partly with treaty obligations under the North American Free Trade Agreement (NAFTA), which played an important role in this process. Nonetheless, interest rate spreads remained too high pardy because of the high ratio of overdue obligations, possibly associated with an excessively rapid growth of credit. Mexican reforms, like those in Indonesia, had been followed by some over- heating of the economy and had also generated the response from the authorities of a very substantial tightening and improvement of arrangements for prudential regulation and bank supervision Ortiz noted more generally that traditional control systems would mevitably become outdated as domestic financial markets evolved in response to reforms and to financial sector globalization. Although both Lnrdonesian and Mexican authorities had responded positively to these situations, many unanswered questions still remain about how these inherent dangers can be avoided. Overview of the Paper The remainder of this paper is organized into seven sections. Discussions are in the following sequence: 14 Akin R. Roe * The topic of capital inflow and financial bubbles, based mainly around the presentation given by Michael Dooley. e The issue of capital account liberalization in the context of the reforms undertaken by Mexico and the Philippines, based around presentations by Guillermo Ortiz and Edgardo Zialcita. * Certain institutional aspects of the fmancial reform process addressed in two sections. The first one discusses the issue of securities market reforms, with special reference to the experiences of Thailand as presented by Ekamol Kiriwat. The second one focuses on reforms in India and Palistan, especially on the various aspects of the resolution of bad debt problems of Indian banks. This section is developed around presentations by Suman Bery and Mohmed Janjua. * Certain aspects of the link between financial system performance and reform on the one hand and macroeconomic stability on the other, based on presentations by GCilllermo Le-Fort, Lin See Yan, and Takasui Kanzald. * The issues that confront the tasks of bank regulation and supervision in a post- liberalization era, based on the presentation by Binhadi from Bank Indonesia. * A few conclusions and contrasts from the four days of discussion. Financia Liberalization, Capital Flows and Bubbles Financial reforms in recent years in many of the counties represented at the semina involved a significant opening to capital account movements. In several cases, for example Chile, Malaysia, Mexico, and Indonesia, this hleralization was followed by large- scale inflows of foreign capital. The most obvious and appealing explanation of these large inward flows is that they represent the response of international capital markets to the improved fundamentals in the reforming countries: that is the genuine pursuit by scarce international capital of the highest rates of return available. Participants generally acknowledged, however, that the true nature and motivation of these flows was often unclear to them; thus, the correct policy responses to large capital inflows remained a matter of considerable difficulty and disagreement. In this general context, it was the role of the first main area of discussion at the seminar to try to clarify the possible causes of large capital inflows and to assess the implications of this for domestic policy management. The discussion was led by Michael Dooley from the University of California. Seninar Proceedings: A Srunmary Report 15 Dooley began by noting the short-life of the prediction made in the darkest days of 1982 and the international debt crisis that debt distressed developing countrinc- .- Sild be unable to regain access to international capital markets for a generation. }n fact, large new capital inflows from abroad and associated boom conditions in emerging stock markets have become widespread since 1989. The Western Hemisphere countries, for example, attracted total inflows of US$24 billion, US$39 billion, and US$43 billion in the years 1990, 1991, and 1992, respectively, and had been able to build their international reserves by US$15 billion or more in each of those three years. The Asian countries, which had suffered fewer negative effects from the debt crisis in the early-1980s, had attracted private inflows in excess of US$40 billion in both 1991 and 1992 and had built foreign reserves by an average of US$32 billion in each of those years. In the Westem Hemisphere the private inflows have been divided approximately equally between the buildup of international reserves and the accumulation of current account balance of payments deficits, while in Asia the impact on reserves has been proportionately higher. In both cases, however, it is clear that policy decisions to sterilize the private inflows have been an important feature of recent experience. The magnitude of the inflows has varied considerably among the indvidual counties. Most dramatically, the inflows into Mexico in 1991 were the equivalent of 8 percent of GDP, which is a larger inflow than that recorded before the onset of the debt crisis. In most other Latin American economies, the recent inflows, though impressive, are smaller relative to GDP than the levels achieved before the debt crisis. Among the Asian economies the recent inflows to Indonesa and Thailand are proportionately larger than before the debt crisis, while the opposite tendency is evident for Singapore and, to a lesser extent, Malaysia. Dooley suggested three possible competing explanations for these large inflows. The first is the one already mentioned above: the fundamentals of inflation and growth rates in the recipient countries may have changed for the better during the past two to three years. Together with low real yields on investment opportunities in the industial countries, this could be a cause of the current large flows. In this case, though, the trend could obviously reverse when competig yields in the industrial world begin to recover. Second, the inflows might be associated with a speculative bubble, in which relatively poorly informed private sector investors are merely playing a follow the leader game based peripherally, if at all, on fundamental factors. The anxieties to which this explanation would give rise if it turned out to be correct is that speculative bubbles evenually burst, giving rise to large-scale reverse flows. This second explanation appears to be consistent with the fact that many actual and potential foreign investors in the reforming economies have litde or no experience in how such economies behave. Thus, as in other market situations, it is plausible to argue that market participants can get caught up in the excitement of the speculative move in 16 AIon R.Roe prices of some assets and can carry these asset prices even further upward and well beyond the levels that are consistent with the fundamentals. Because some market participants made money by buying the assets yesterday, others go along for the ride to make money tomorrow. For this to continue, someone tomorrow must be willing to pay a higher price for the asset than was paid today. The third possible explanation, and the one that Dooley himself regards as the most plausible currently, is that the policies of the monetary authorities themselves may be creatng arbitrage opportunities that the foreign investors are exploitng when they push large sums of money into some of the reforming economies. As in the case of speculative bubbles, these inflows may have little connection with the fundamentals in the economy. Unlike ill-informed speculative inflows, however, these inflows depend on the implicit guarantees that governments seem prepared to offer. In essence, the inflows will continue or reverse depending on the scale of the resources that seem likely to be available to support the implicit government guarantee. The analogy with the 1980s debt crisis is usefully spelled out to explain this proposition. In the buildup to that crisis, the govermnents of debtor countries offered guarantees to foreign creditors on the dollar value of their own or domestic residents' liabilities to foreign creditors. Thus as debt accumulated during the 1970s, the government matched this actual accumulaton with the buildup of large contingent liabilities. These became a little noticed but eventually important part of the govemrments' financial positions. Arguably, to some extent the offer of the guarantee may even have become the cause of many of the inflows. By contrast, the more recent inflows have taken the form of foreign purchases of domestic currency instruments, and explicit government guarantees have been largely absent. Thus, as Dooley noted, if mistakes are being made, they are certinly different from those made in the 1970s and early 1980s. Instead, he argued, the mplicit guarantees now bemg provided by many governments are of two main forms. First, by operating exchange rate policies linked to a fixed or managed exchange rate, many governments, in effect, provide investors in domestic currency-denominated assets with exchange rate guarantes. Second, monetary authonties are limiting the decline in yields on domestic assets. In principle, large-scale infows of capital should sooner or later give rise to iblis in the rates of return on domestic assets, which will then discourage further inflows. This process was short-circuited in the early 1980s when large inflows into many countries were matched by large unrecorded private outflows. By contrast, in the early 1990s governments in many countries are showing a strong tendency to limit the monetary and inflationary consequences of large inflows by issuing securities to finance the purchase of large foreign exchange flows. In effect, the monetary authorites intervene in the foreign exchange markets to prevent too much appreciation of the domestic curency (fearful of the scale of the depreciation that would follow from a turnaround). They thereby recycle these funds back into the international Seminr Proceedings: A Summaiy Report 17 circuits. The authorities also intervene in the domestic money markets to prevent a large fall in interest rates. Dooley contended that, unlike a speculatively motivated inflow, this process can go on for a long time because of the demonstrated willingness of central banks in recipient countries to pay a higher interest yield on its domestic currency liabilities than the return it earns on its accumulating foreign currency reserves. The implied losses to the central bank need not become constraining for some time. Also by contrast with speculatively based inflows, the process, whereby investors feed off implicit guarantees of governments, means that the turnaround, when it comes, imposes losses not on private speculators but on the government itself as its contingent liabilities are called. Ihis arises because the turnaround involves both a rundown of foreign currency reserves in the central bank as well as probable assistance to the domestic banking industry to meet its domestic currency withdrawals. The policy implications of all this are somewhat disturbing. It is unlikely that most governments, benefiting from large capital inflows, will be able to persuade investors that the implicit guarantee accorded to bank deposits will be withdrawn, because investors know that in most countries bank bailouts do and will happen. Therefore, the onus for exposing investors to more gemnine risks (that is, limiting the one-way nature of their bets) has to lie with signals coming from exchange rate policy. More explicitly, Dooley argued, that governments should be prepared to allow capital inflows to have a more depressing effect on domestic interest rates by not steriling the inflows, or, alternatively, governments should allow the nominal exchange rate to adjust more to market forces (that is, appreciate more when there are large inflows). In Dooley's view neither policy would be likely to seriously damage those capital inflows genuinely associated with improved investment climates in the reforming countries. It would, however, help to limit the "round-tripping" of foreign exchange that, in his view, has increasingly dominated fixed exchange rate systems since the onset of globalization in capital markets. The implication of Dooley's suggestion is that countries adopt more flexlble exchange rate policies and develop some alternative target to control inflationary expectations. Discussion from the floor generally confirmed the lack of consensus about how to deal with large inflows. It was noted, for example, that countifes such as Mexico, Indonesia, and Malaysia, which have been explicitly working for a more diversified production and export base, cannot be entirely indifferent to the way in which the exchange rate is moving. However, it was agreed that complete sterilization will certainly cause interest rates to rise, which will intensify the attractiveness of investment in a country and so give rise to more inflows. The other extreme policy of doing nothing in response to the inflows was certain to result in serious exchange rate appreciation and damage to export prospects. It might also harm the prospect for attracting foreign direct investment in export-based industries. The implication seemed 18 Alan R. Roe to be that the correct policy was somewhere between these two extremes, but participants differed on how to determine the exact stance. A number of people argued that an exchange rate target, preferably based on a degree of undervaluation, was necessary for reform. Associated with this, another participant suggested that a third policy instrument, namely fiscal policy, might be introduced for this purpose. Specifically, if there was some reluctance to abandon a fixed exchange rate target, then the dogmatism of near balance in fiscal accounts might be abandoned with the size of the fiscal balance being used explicitly as a countcr- cvclical device to compensate for the appreciating effects of large inflows. An opposing view was that among the faster growing of the reforming economies it was inappropriate to try to maintain the same nominal exchange rate in the face of that fast growth. Indeed, the rate had to be allowed to appreciate to reflect the changes in the necessary structure of production associated with rising real wage levels. Another aspect of the discussion related to Dooley's point about the implicit guarantees accorded to domestic currency liabilities in the banks. Participants generally agreed that such guarantees are indeed provided by the countries represented at the seminar and necessarily so. Bank failure in many of these countries can threaten the entire countries' payments systems. Because this must be avoided at all costs, the banldng system itself has to be protected. At the same time, participants argued, blanket commitments to bail out all banks in most circumstances were unnecessary and even damaging, as the earlier experiences in Malaysia had indicated. The current procedures for dealing wist bank failures in Malaysia are more considered and they include arrangements to penalize shareholders and others for such failures. The practice in the United States of relying heavily on deposit insurance has been rejected, because it is taken to imply a lack of credibility in the central bank's overarching responsibility for the stability of the banking system. Nonetheless, depositors normally receive priority treatment in the event of failure; thus, Dooley's basic point about an implicit guarantee has validity. An important practical implication that the discussion drew out related to the prudential regulation of banks. If banks are indeed able to attract large liabilities from foreign sources, then regulatory arrangements need to reflect the risks of this. The approach adopted in Mexico had been to require the dollar-denominated liabilities of the banks to be matched by loans and other assets by organizations able to denominate their eamings in dollar terms. While this approach did not completely remove the dangers of a reversal of capital inflows to the banks, it did provide the banks and the authorities with some risk limitation against such an evenuality. The final point that was debated concerned the practical problems for central banks in managing large inflows, an issue that recurred in subsequent sessions of the seminar. Participants noted that the sterilization process was far more difficult when fiscal accounts were in balance or in surplus and there was no government issued paper to sell. In these situations the central banks themselves were having to issue paper as the medium for conducting the sterilization process. This was expensive given the &minor Proceedlhgs: A Smnmary Report 19 lrge volumes that had to be sold. Currently, returns on dollar-denominated reserves were only about 3 percent; far lower than the yields typically required to sell domestic paper. Thus, central banks were indeed incurring significant losses on these transactions. In an environment of steady currency appreciation, there is also a tremendous demand to sell export earnings forward. This can also expose central banks . - substantial foreign exchange losses by providing the forward cover. Overall, the fmancial situation of central banks, as well as their ability to operate independently, are compromised by a situation of sustained capital inflows. Capital Account Liberalization in Mexico and the Philippines Following this general discussion of the possible determints and dangers of capital account inflows, the proceedings of the seminar moved on to review and discuss the experiences of capital account liberalization in two specific cases, namely, Mexico and the Philippines. Mexico now has a record of sustained progress on economic reform spanning several years and including most major areas of stabilization and adjustment. By contrast, financial reforms in the Philippines date back only three years to 1991. Hence, wnile in both cases it was possible in the discussion to consider the content of some of the relevant reforms, there was a rather stronger basis for assessing the results of reforms, including capital account liberalization, for Mexico than for the Philippines. Mexico In introducing the topic, Guillermo Ortiz, Undersecretary for Finance and Public Credit, began by confirming that in the past few years Mexico has witnessed a considerable increase of capital inflows. From a net flow of minus US$1.5 billion in 1988, the figure has grown rapidly to a total of about US$20 billion by 1991. In 1992 private flows amounted to US$20.9 billion. In addition, the public sector reduced its own external debt by some US$2.8 billion. From 1991 to 1993 a total of some US$70 billion of inflows is anticipated. This is already having, and will contiue to have, a profound effect on the country's ability to finance its investment programs. The reasons for the inflow represent a combination of the three factors indicated in the presenation by Professor Dooley, that is fundamentals of inflation and growth rates, speculative bubbles, and policies creating opportunities for foreign investors, but Ortiz stssed the important nicroecononic reforms that have greatly improved the producfivity of capital in Mexico's economy. Ortiz stressed that, like Indonesia, Mexico had operated with a reasonably open capital account since the end of World War II. Although dual exchange rate 20 Alan R. Roe arrangements continued for a time, representing a tax on capital inflows, few formal restrictions existed on capital movements. Ortiz agreed, however, that the return of the country to a situation of voluntary lending in 1989-90 after the completion of the Brady debt reduction program has been equivalent in many ways to the opening of the capital account. In Ortiz's opinion, the increased diversity of the financial and real productive assets available has motivated the recent large inflows. These assets now offer investors considerable scope for achieving various different combinations of risks and returns. The enviromnent is now one in which fiscal deficits (excluding privatization revenues) have been eliminated and a variety of attractive new capital market instruments have been established. The flows have accrued mainly to the private sector. In both 1990 and 1991 significant inflows were channeled through the Mexican commercial banks, but in 1991 and 1992 the authorities imposed limits on commercial bank borrowing from abroad. This meant that in 1992 all the inflows went to the nonbank private sector. Indeed there was even a small reduction of foreign currency liabilities for commercial banks in that year. This evolution in policy was motivated by the anxiety of the authorities to avoid the lender of last resort obligations to foreigners that Dooley's analysis had indicated. Ortiz noted that the inflows of the past two years (1991 and 1992) emphasized stock market instruments, including some government securities investment. Foreign direct investment (EDI) amounted to 25 percent of the total in that period against 40 percent for corporate stock market securities. This latter influence has resulted in a near quadrupling of the capitalization value of quoted securities on the Mexican stock exchange since early 1990. The value of securities now stands at about US$140 billion, including a US$25 billion element attributable to foreign investment. A foreign stake of approximately 25 percent in the outstanding issues of government securities also exists. Banking sector liabilities to foreigners grew by almost US$5 billion per year cumulatively in 1990 and 1991 but have fallenf subsequently because of the recent restrictions on bank offshore borrowing. Aspects of Mexico's Suiccess. Ortiz attributed Mexico's considerable success in attracting increased foreign investment to a combination of three factors: the government's consistency in maintaining sound macroeconomic policies; the liberalization of the country's financial markets; and the country's abilitv to introduce a wide range of new and varied financial instruments able to tap the interest of a number of different types of investors. Regarding macroeconomic aspects, confidence had returned largely because of Mexico's ability to make deep fiscal adjustments, to reduce inflation dramatically, to move toward external account balance, and to achieve higher rates of economic growth. Some of the institutional changes behind these adjustments have also been significant, not least the liberalization of trade through the NAFTA agreement and the general retreat of government from many areas of economic activity where it had previously been a significant player. Seminar Proceedings: A Swnriay Report 21 Concerning new financial instruments, Ortiz noted that the private stock instruments that had been used had included American depository receipts, free subscription stock, Mexican funds, and Eurobonds. Government securities have included cetes (with maturities ranging from 28 to 364 days), bonds (floating rate notes with one and two year maturities), ajustabonos (yields indexed to the consumer pnce index), pagafes (yields indexed to the controlled exchange rate), and tesobonos (yields indexed to the free exchange rate). Because of the generally favorable international perceptions of Mexico's reforms, the risk ratings attached to the country's financial paper have also improved considerably, which also helped stimulate the inflows. For example, the discount on secondary market pricing of Mexico's external debt fell from 35 percent to only about 16 percent from the beginning of 1990 to the end of 1992. Mexican Eurobond issues bad become the actual benchmark against which other Latn American issues were judged and in 1991 and 1992 had accounted for almost 50 percent of al Latin American bond issues. Furthermore, the interest rates that the government needed to offer on its own paper had fallen considerably since the late 1980s. The average nominal rate on 28 day treasury certificates (cetes), for example, fell from over 100 percent at the end of 1987 to less than 20 percent by 1992. Matuities also lengthened, with about 59 percent of overseas holdings of cetes, having a maturity of 364 days by December 1992. Some Macroeconomic Consequencee. Ortiz applied to the Mexico sitation some of Dooley's earlier analysis about the macroeconomic difficulties caused by large inflows. He reiterated that sudden and large inflows will make it far more difficult for the authorities to control inflation. Additionally, the authorities always need to consider that large inflows in one period could become large outflows at some later stage and will thus destabilize the economy. Relative to the two extreme responses considered in the earlier session (full sterilization giving rise to large central bank losses or no sterilization giving rise to economic overheating), Mexico tried to steer a middle course. This involved, above all, limiting external borrowing of domestic banls and adhering to structural reform policies that, among other things, sought to slat an increased flow of funds into productive investment. Ortiz emphasized that a great deal of the industrial restructuring currently taking place in Mexico required large-scale imvestment funds and high levels of imported equipment. The restuctring, however, did not necessarily give rise to instant improvements in overall growth rates of productive activity, partly because some unproductive industries were declining. This resulted in a large current balance of payments deficits. Ortiz pointed out that while the authorities took measures to restrain inflows, the inflows continued on a substantial scale and had caused a very significant appreciation of the Mexican real exchange rate since 1987 (from an index value of 165 22 Alon Roe in 1987 to 108 by 1992: 1970 = 100). Although, officials decided late in 1992 to widen the bands within which the nominal exchange rate was free to move, thus far it has not made it any easier to avoid the appreciation of the real exchange rate. Thus, microecononie reforms need to be sustained in an effort to generate average levels of improvement in productivity sufficient to validate the new and still rising level of the exchange rate. This, he noted, is an extremely demanding treadmill. Much of the discussion following the Ortiz presentation focused on the sustainability of the successes and the large inflows so far achieved by Mexico. Several participants raised questions about the extent of the adverse effects that the appreciation of the real exchange rate has had on the country's export performance, especially in the machialadores sector, which had traditionally relied on its large wage cost advantages relative to the United States. More generally, participants noted that the tendency for productivity performance to adjust to reconcile the exchange rate with the level of capital inflows was very much a long-term phenomenon. In the short run it was more likely that wage rates and nontraded pnces generally would rise as a consequence of large inflows and that this would undermine i onal competitiveness. Ortiz agreed but pointed out that in spite of this factor Mexican exports had risen quite rapidly in 1991 (13 percent) and reasonably fast (7 to 8 percent) in 1992 in the wake of the U.S. recession. Also concerning susmainability is that in the long run the current account of the balance of payments must balance, which implies zero capital inflows (net) over the long term. Thus, the domestic savings rate must rise eventually to provide for an exportble surplus, which Mexico currently is not achievng. Ortiz agreed that domestic savigs rates had dropped, although rates for the public sector, where interest payments were formerly much larger, had improved considerably. Possibly the most controversial issue was whether the precautions Mexico had taken to limit capital inflows would insulate the stability of the domestic financial sector from the more serious potential consequences of an eventual reversal of flows. Ortiz argued that because so much of the corporate borrowing of the country was now done using the companies' own paper rather d borrowing through the banks, some of the earlier risks of external borrowing had been lessened. More specifically, he saw no particular reason why the commercial banks should bail out industrial companies that found themselves in difficulties because of international funds being withdrawn or drying up. Thus, the government too was insulated to some extent from the likelihood of turning implicit guarantees to banks into explicit payments if ddficulties emerge in the indusil sector. Other participants, however, argued that if loans to nonbanlks were called, it might be dfficult for domestic banks to avoid serious involvement because the companies in difficulties would trn to domestic banks for funds to replace those being wihdrawn. This point, although unresolved, is clearly a significant concem in his area of policy debate. Ortiz's contention, if correct, implies that the form in which international capital flows accrue to a country can be important in diluting the problems Seminar Proceedings: A wmwy Report 23 of the implicit government guarantees as explained by Dooley. Countries such as Mexico, in other words, can securely locate the risks of international capital transfers with the foreign lender and the domestic private corporate borrower. If that contention is incorrect it implies that some of the efforts being made by participants at the seminar to influence the types of capital inflows they attract may be less effective than they anticipate in limiting the risks of those inflows to the government and the macroeconomy in general. The Philippines The experiences of the Philippines were iimroduced by Edgardo Zialcita, Deputy Governor of the Central Bank of the Philippines. Zialcita began by briefly reviewing the history of exchange control arrangements in his country. Exchange control had been introduced in 1949 in an effort to conserve scarce foreign exchange for the reconstruction of the country's war-damaged economy. The government finally achieved full convertibility of the current account in 1991-92 at the same time that it took significant steps to liberalize the capital account Although there had been an earlier and initially successful liberlization of the current account m the 1960s, both an import surge and a deteriorating terms of trade soon forced the authorities to again introduce controls. Thereafter, until the early 1990s, reforms had been largely of a fine-tmning nature. Capital account restictions had remained pervasive. For capital inflows, officials estabLished criteria to channel both borrowings and direct investmens by specified priority area borrowers. Registration was a strict conditon for borrowers gaining approval to service debts and to make dividend payments. A few small windows of opportunity for foreign funds were nonetheless created in this generally restictive environment. For example, in 1976 an offshore banking system was established to allow foreign banks a small base for operating in the Philippines in spite of a general block on the entry of foreign banks. Zialcita explained that, in spite of this restrictive regime, the economy still proved susceptible to a series of balance of payments crises. These culminated in the crisis of the early 1980s, which eventually led to the debt moratorium declared in October 1983 and to an emergency package of firther exchange controls. The Interational Monetary Fund (NM), supported reforms that followed this crisis resulted in a floating exchange rate for the Philippine peso based on a freely determined interbank rate. TIhen between April 1986 and April 1988, the government undertook phase I of an import liberalization program, followed by a second phase. These two phases achieved a high degree of liberalization of exchange controls on trade. Little action, however, was taken at that stage on nontrade items, and the conditions under which the interbank foreign exchange market functioned continued to be restrictive. 24 Alan R Roe Further economic difficulties in 1989, partly attributable to political factors, also confirmed the view of many that the economy was still failing to achieve its true potential for growth, especially in the areas of investment and exports. Thus, a new programn with the RMF introduced a finther effort to reform exchange controls, as well as to boost exports and foreign investment These reforms were staggered through 1991 and 1992. Recent Measures. Zialcita explained several of the main components of the reforms implemented since 1991. First, exporters were given ful freedom to dispose of their own foreign exchange receipts. Second, restrictions on all methods of imaport and export payment were lifted, except payments through letters of credit. The central bank also lifted all prior approval requirements on export ransactions. Third, restrictions on access to foreign exchange for service payments were lifted and free trade in gold was allowed. Fourth, exporters were allowed to hold foreign currecy deposits in domestic banks and were given access to short-term foreign currency loans, mncluding trade faciRity credits from domeszic sources without central bank approval. This reform for exporters should produce much lower costs of trade finance with minimal foreign exchange risk. Fifth, most restrictions on foreign direct investment were removed and investors were granted full rights to make profit and capital repatrations, except when these arc financed by borrowing from the domestic banking system. Equaly, outward foreign direct investments were allowed without central bank approval, provided these come from foreign currency deposits or from nonbank sources. This reform is seen as a route to improve export performance by faciliating overseas export-related infrastructure and marketing alliances with overseas companies. Fmally, the Pilippines commercial banks are now allowed to hold short or long foreign exchange positions, subject only to limits imposed for prudexial reasons. In addition, the earlier trading restrictions on the interbank foreign excbange market were lifted. Although some of the components of this reform package appear similar to those attempted unsuccessfully on earlier occasions, there is now one disfinctive strategic difference. Efforts ftiis time focused on those aspects of exchange liberalization that can genuinely enhance the supply of foreign exchange, through, for example, lower transaction and financing costs, greater flexibility in capital expenditure decisions, and more efficient foreign exchange pricing and management arangmts. Some restrictions still remain. For example, outward direct investment is stll subject to a US$1 million limiit per investor per year. These arrangements, however, are subject to review and are likely to be removed in part if the current high level of capital inflows to the economy continue. In particular, the government is seriously considering liberaizing the entry of foreign financial institutions to futher improve the efficiency gains arising from the reforms, so far. Seminar Proceedings: A Srwnmy Repon 25 Thze Early Results. Zialcita explained that the reforms in his country are very recent, with some of them being completed only by the second half of 1992. Thus, it is too early for a full judgment about their effects. A few initial indicators of performance, however, are available already. First, flows of invisible receipts, such as travel and workers remittances, have surged since 1991, suggesting that the greater freedom to acquire foreign exchange combined with a market-determined exchange rate is making people more relaxed about bringing money into the country. Second, preliminary data indicate a portfolio shift away from overseas deposits by residents and, to a lesser extent, a shift from local foreign currency deposits by residents. If confirmed, this would indicate that reform is improving the attractiveness of domestic relative to foreign financial instruments, a tendency clearly demonstrated in recent experiences in Mexico. Third, there has been a large increase in the volume of interbank foreign exchange trading, from only about US$1 million per quarter in 1990 to almost US$1.5 billior. in some recent quarters. Within this total the central bank has been very active, particularly in buying foreign exchange into official reserves, which now amount to almost US$10 billion (including commercial bank holdings). Finally, the surge in import and service payments, one of the fears expressed before liberalization began, has not materialized, and 1991 and 1992 have shown no obvious deviation from past trends in payments either for merchandise imports or for services. Overall, it is a justifiable initial conclusion that reform has enhanced rather than diminished the availability of foreign exchange to the country. Unfortunately, this has not yet been matched by obvious success in attaining one of the two primary objectives, namely, that of enhanced export performance. Lie Mexico, it seems possible that the appreciation of the real exchange rate combined with the early restructring bias in productive sector investnents have jointly contributed to a slower growth in this area than was hoped for. By contrast, investment has performed well with both portfolio and direct investments from abroad manifesting strong improvement through 1992. Policy Dilemmas. The results for the Philippines are very similar to those experienced by Mexico. Specifically, the real exchange rate appreciation has resulted in pressures from exporters for a better deal. Although the central bank has responded to some extent by intervening in the interbank market, it is highly constrained in this area because of its monetary commitments to the IMF. Hence, interventions have largely been sterilized. This process is fiscally damaging, especially since the Philippines began with such a high level of domestic debt. Two solutions have been debated in the country to try to resolve these problems. The first is to impose some administrative restrictions on capital inflows, an action that could be ineffective given the liberalization of current account transactions. A second is to strengthen the stance of fiscal policy. This, howeve-, is very difficult because of the enormous need to 26 At&nIL Rue replace and modernize much of the country's infastuctre. Possibly, the best approach, Zialcita argued, is to look for greater flexibility in monetary policy, especiarly now that inflation is tending downward (8.2 percent in December 1992 compared to over 20 percent in mid-1991). Commentary from the floor focused first on the crucial need to get export growth under way. It was noted that the Philippines' export sector was handicapped not only by the high exchange rate but also by the world recession and the poor state of the country's public capital. This last point was certainly an argument against an excessively tight fiscal stance in the immediate fture. Participants also pointed out that from sometime in 1991 the general perception had been that a currency devaluation was unlikely, and this had resulted in a big increase in capital inflows from abroad. Other participants referred to the difficulties the authorities have had in relinqishing some of their traditional controls. Four main fears seem to underly this: the far of losing control of monetary policy, the fear of outflows leading to a complete loss of foreign reserves, unpredictabilities associated with imperfect markets, and fears associated with the inherent uncertines of the global environment Some countries, such as Indonea, had begun to liberalize when their reserves were extremely low; tus, it could be argued they had nothing to lose. Looking at Malaysia's experiences, participants argued that it was largely unnecessary to retain controls on payments and that it was in any case a good principle to allow people to do whatver they liked with their own money. -A number of critical controls on receipts, however, had proved useful for Malaysia. These were that all export proceeds come back to the country, that offshore borrowing be subjected to some linitations, and that regulons be maintained to deter the fly-by-night type of inward investor. Finally, participants noted that liberalization has an extremely important informational content. In particular, in highly controlled environments the informatioa that the authorities accnie is likely to be seriously incomplete and subject to major error. The liberalization of formally controlled acdvities is likely to improve the willingness of the public to divulge information about most aspects of its transactions and balance sheets. Hence, the necessary residual regulations of the post-liberalization period are likely to be far easier to monitor and enforce than those of the earlier era! Securties Market Reforms: The Experience of Thailand The seminar moved on to debate a variety of themes around the general topic of instittional development or deepening, and reform. The first of these themes, discussed mainly in the context of the Thai economy, related to securities market developments and explicit attempts to dilute the relative importance of commercia banks within the overall financial system. The second addressed the more general question of how a country such as India that is still in the relatively early stages of its structural reforms should approach financial instiutional reform given that such a large Seminar Proceedaqsi A Sumary Report 27 part of its overall financial system still operates on a socialized basis and is plagued by high ratios of non-performing loans. This discussion continued into a discussion of a rather broader set of reform issues in the context of Pakistan's reform measures. This section focuses on the security market issues; the broader issues pertaining to countr>s such as India and Paidstan are set out in the next section. The issues of the reform and management of domestic securities markets were seen as important from the viewpoint of the reform programs of many of the countries represented at the seminar. However, these issues were discussed centrally in only one paper, namely, that presented by Ekamol Kiriwat, Deputy Govemor of the Bank of hailand. Kiriwat noted that the extremely high growth rates in his country through part of the 1980s had been achieved in spite of a number of significant weaknesses in the financial sector These had included a very minor role for money and secuity market instrments in the process of corporate finag. He also pointed out that, by contrast with the Philippmes, where financial reform took place against a background of macroeconomic crisis, the financial reforms implemented in Thailand were undertaken more in anticipation of difficulties to come rather than in response to immediate problems. Specifically, the government designed a comprehensive reform program at the end of te 1980s to cover the tree-year period from 1990 through 1992. This was after the authorities bad already achieved a balanced fiscal position and a strong export onentation in trade policies. The first step toward reform was an important prerequisite for the liberaliation that the authorities introduced. This first step began in June 1989 when the financial authorities freed interest rates on time deposits with a maturit of more than one year, then abolished interest rate ceilings on all time deposits in March 1990, and, in Jamary 1992, removed ceilings on savings deposit rates. Although the heavy dependence of Thai business on finance from a small number of domestic banks represented a potential break on the pace of possible development, the authorities nonetheless decided not to expand the financial system by allowing the entry of new banks. This is in sharp contrast to the sitation in many other reforming economies where such an opening up has been seen to be an important influence on the competitiveness of banking insttutions. Instead, the authorities were reasonably satisfied that the old style of family managerial control in the bans had been replaced by more modern approaches and that the capital adequacy and portfolio quality had also significanty improved. On this basis the authorities then encouraged a significant expansion of new bank branches. Thus, the number rose from some 600 in the 1970s to over 2,000 by the end of the 1980s. The capitalizton in ban increased significantly in support of this expansion when the authorities required banks to access the stock market to increase their capital. This further diluted the ownership influence of the largest and most powerful owners, and injected some good quality new paper into the markets. 28 Alan R. Roe The share of bank capital attributable to the largest owners is now reduced to only about 20 percent of the total. In most cases, too, the tradition of owners also being presidents of most banks has given way to a high incidence of well-qualified professional managers in the top jobs. Although Thailand has had some distressed banks, Kiriwat explained that the small number of banks in total had made it reladvely easy for the regulatory authorities to deal with these problems on a case-by-case basis rather than through any systemic approach. Concerning securities markets, Kiriwat explained that before the new Securities and Exchange Act of 1992 there had been numerous weaknesses in the Thai legal system regarding the issuing of securities and their wading. Numerous laws emanated from different official bodies. Participants in securities markets needed to observe each of these laws, which were often poorly coordinated and inconsisent with one another. Furthermore, a number of different supervisory agencies oversaw different aspects of the securties market business. This led to considerable inefficiency and confusion. Above all, the role of the stock market in corporate financing had remained marginal, with only public companies authorized to makre public offerings of their shares, and only some thirty-tbree public companies actually in operation in the country ten years after the enactment of the Public Company Act of 1978. The Thai authorities felt that securities market development was crucial to (a) provide some counterweight to the power of the large banks; (b) provide managers in a more deregulated enviromnent with more and better mstruments to enable them to manage their cash flows and risk; and (c) above all provide a more secure and transparent medium for investment financing and the channeling of foreign as wefl as domestic sav ngs into corporate investments. The Securities and Exchange Commission (SEC) set up under the 1992 Act represents a radical improvement on previous arrangements. Kiriwat explained that this Act now unifies under SEC control most of the regulatory and supervisory functions pertaining to the securies business that had previously been dispersed over a number of different agencies. He explained that the SEC comprises eleven members appointed from the various bodies that used to participate in securities market regulation. Because these include the governor of the cental bank and the Mnister of Finance, the SEC has clearly been constituted as a body that is potetally able tO achieve consensus among all the top people involved in the running of the Thai fimancial system. Its work is supported by a secretariat of full-time professional lawyers and accountants. The SEC is now required to supervise all types of public offerngs, thereby closing the earlier loophole in which sales through existing shareholders of a company could easily bypass stock exchange scrutiny. The stock exchange itself remains responsible for the opertional work of the exchange, but regulation and enforcement now fall under on the SEC. The 1992 Act also includes many requirements to improve investor protection. These include the requirement to obtain SEC approval prior to any issue;. the requirement to disclose correct and fUll information prior to any public offering; severe Seminar Promcedings: A Swnay Report 29 penalties for insider trading; and a very strong emphasis on the importance of capital adequacy of the participating dealers. An early priority under the new arrangements, Kiriwat explained, is to try to achieve higher levels of activity in issuing and trading domestic debt instruments, which in Thailand have been used even less than have equity issues. To this end the Act stipulates that debt instruments can be issued by both public and limited companies, subject to SEC approval (equity issues are still restricted to public companies). The Act has also provided a comprehensive definiton of the term "debenture" to include all ps of debt instruments, including long-term and most short-term debt. This is to avoid the problem of borrowers circumventing the regulatory rules by calling basic debt instruments by other names. The discussion focused on a number of issues brought out by Kiriwat's presentation. Some participants wondered whether, in a banldng system that had been so dominant in the provision of finance, there had been significant problems of banks moving into securities market development, and whether there had been significant problems of bank distress. Kiriwat explained that although some banks had gained control of finance companies, some of which were active in the securities markets, the significance of nonbank financial institations in the economy was still relatively minor (only 7 percent of the overall system). He also confimed that some three or four of the country's sixteen banks had been in difficulty from time to time. These problems, however, had been resolved on a case-by-case basis with some occasional recourse to the country's fund development institution, which was set up originally to support finncial sector development and still operates with a degree of independence from government. Thailand did not operate a deposit insurance system because it was assumed that the government would not allow a bank to fail. The one case in recent years where the diffculties of a bank proved to be serious had resulted in actions to effrctively close that bank thomugh a merger process. Kiriwat emphasized that in the recent past the minimum capital required of banks has doubled and that their provisions for doubtfil debts have increased. He felt that with these two prudential measures strengthened, all other aspects of bank stability were reasonably well assured. On the question of fiscal and other incentives to stimulate securities market development, Kiriwat explained that no such incentives had been used in Thaiand in spite of the clear objective to achieve a more important role for the securities markets. Kiriwat felt that most companies in the country now understood the need to achieve better debt to equity ratios and that fiscal incentives in that context would be unnecessary. Indeed, the curremt arrangements whereby equity issues are only permitted to public companies who in turn face very onerous disclosure requiements means in practice that the incentives to list are somewhat negative. 30 Alan R Roe Institutional Reform: The Cases of India and Pakistan India Suman Bery from the Reserve Bank of India explained that hndia is currently engaged in a wide-ranging program of reform and deregulation, not only in the financial sector but in many other areas, such as trade. Although these reforms began in the early 1980s, they had been intensified significantly since July 1991 after the political crisis that followed the assassination of Rajiv Ghandi. With a number of senously negative external factors-the Gulf Crisis and the turmoil in the former Soviet Union-to contend with, it became apparent that the economy could no longer cope with the severe inefficiencies of its old protective arrangements. Although financial reform was not given top priority in the early stages, a series of events since the middle of 1991 have brought it to the forefront. These include the report of the high-level Narasinham Committee that is higbly cntical of the state of the country's depository institutions; the international pressures for improved capital adequacy, at least in Indian banks operatng abroad; and the evidence emerging in April 1992 of major iregularies in the behav-or of some Indian banks and their systematic violations of Reserve Bank regulations. As a result of these various factors, Berry explained that now there is a broad-based recognition in the country of the need to change the role of the public banks from one of bureaucratic channeler of funds to public sector acdvities, to one mvolving full commercial assessment of private sector loans and nsks. Likce the Philippines, India's reforms are set in the context of broadbased sutbilization and structural adjustment programs supported by the IMP and the World Bank. Bery explained that the main components of these have been fairly standard and have fcused on two major elements. The first has been fiscal retenchment with the target being to reduce the overall public sector deficit from the equi-alent of 10 percen of GDP (6.5 percent for the central govemment alone) in 1991-92 to 5 percent for the central government in 1992-93. The second has been a set of strucural measures to estblish more tansparent and less distortng trade and foreign exchange arrangements. The program includes the elimination of quantitative restrictions on trade and the replacement of a highly complex administrative allocation of foreign exchange by a much simpler dual exchange rate system for current account transacdons. In the financial sector, much of the emphasis has quite naturally been on the bankig system, which accounts for some 70 percent of the overall financial system. As a result of the 1969 nationaliztions, about 90 percent of the assets of the banks fall under direct state control with some limited competition coming from a number of foreign-owned banks. For at least twenty years, Bery explained, the banks have been regulated more by the achievement of quantitative targets for lendig to pnority sectors, including the government itself, than by normal standards of financial soundness. This has resulted in some deepening of the fimncial sector, for example, a Seminar Proceedings: A Swnmmwy Report 31 better branch network and the marginalization of money lenders, but, has led to a very poor record on allocation performance and financial soundness. Diagnosis. Against this background, the first and obvious step toward institutional strengthening has involved a diagnostic exercise to establish the true fnancial condition of the Indian banks. In April 1992 more stringent standards of income recording and asset classification were introduced for incorporation in the accounts for the financial year 1992-93. For example, a non-performing asset is now defined as one in which interest has remained unpaid for a period of four quarters after the date it became past due. Banks are now instructed not to book interest on non- performing assets. Assets now have to be classified into four categories, namely, stadard, substandard, doubtful, and loss assets. Banks are also required to make provisions against substandard assets and those in the lower classifications according to clearly stated rules. Concurrently, capital adequacy requirements linked to risk- weighted assets now are: 4 percent by March 1993 and 8 percent by March 196. Foreign banks operating in India needed to hit the 8 percent ratio by March 1993. Preliminary estimates suggest that the total of the domestic non-performing assets in the banks could constitute about 7.7 percent of total domestic deposis. Relative to the new capital adequacy and provisioning requirements, there is probably a provisioning gap of some Rs. 100 billion (US$3.3 billion) and a capital gap to achieve even the 4 percent trget, of some Rs. 40 billion (US$1.1 billion). Bery felt this indicated an aggregate situation that is serious but not unmanageable. Suggested Soltions. Bery's discussion of the likely resolution of this problem of bank restructuring covered five main areas: capital injections (mcluding bad loan recovery), the restoration of bank profitability, interest rate policies, the regulation and supervision of banks, and bank management Bery noted that direct capital injections from the budget were largely ruled out by the fiscal constraints referred to earlier. Hence, the Narasimham Committee recommended that the stronger banks be given access to the capital markets. While this is a technically realistic suggestion, as the Thai experience indicates, in the Indian case it would require existing legislation governing bank nationalization to be amended. In response to questions on this particular point, Bery stated that India is now witnessing a definite move from state-led to market-led development, but that the dismantig of the old state system still constitutes an enormous problem. The Narasimham Committee also proposed an asset reconstruction fund (ARF) along U.S. lines that would capitalize the banks with its own bonds and then collect the bad loan component of bank portfolios. Bery noted that this idea had met little favor because it was recognized that in a country the size of India it would be inpractical to expect a new ortion to be effective in collecting loans on behalf of numerous and 32 Aln R. Roe dispersed banks. Additionally, it was unclear how selling bank assets at a considerable discount through the ARF could serve to inject sufficient funds into distressed banks. Regarding bank profitability, Bery noted that restructuring the Indian banks would not be sustainable unless their profitability could be restored to reasonable levels. The Narasimham report, in its examination of the reasons for unsatisfactory rates of profit, had ascribed most of the blame to the massive preemption of bank funds by the government through the statutory liquidity ratio and the cash reserve requirements imposed on the banks at generally increasing rates through most of the 1970s and much of the 1980s; and to the losses associated with much of the directed lending to priority sectors. Action so far taken has reduced the combined burden of liquidity and cash reserve requirements from 63.5 percent of incremental deposits to about 25 percent by early in 1993. While the large number of different concessional rates for priority sectors has begun to be rationalized to some extent, no reduction in the 40 percent of total loans targeted to these sectors has yet been approved. Thus, with more funds available to the banks there is some danger that the absolute amount of lending to priority sectors could rise rather than fall. As regards interest rate policy, Bery noted the considerable anxiety about allowing too much freedom in determining interest rates until the fiscal deficit is somewhat reduced and the situation of distessed banks is closer to resolution. Pending this, the government continues to control the situation mainly by setting maximum interest rates on longer term deposits and minimum lending rates on loans. This system is partly motivated by the government's desire to keep its own funding costs low. However, it aiso reflects the situation of distress in many banks and is thought to be likely to deter deposit mobilization at very high interest rates for the purposes of meeting the needs of distressed borrowers. For the moment, this system of interest rate control has resulted in wide spreads between deposit and lending rates, mainly because of the contnuing high implicit taxation of the banks. Although the Narasimhan Committee recognized the fundamental need to shift the regulation and supervision system of banks from one based on quantitative lending targets to one based on the financial soundness of banks, no decisions have yet been made to implement this suggestion. The Reserve Bank, however, has announced new guidelines for establishing new private sector banks designed to increase the overall competitiveness of the sector. These new banks will be required to achieve the 8 percent capital requirement from the outset, but they are also subject to the same priority scctor lending ratios as the established banks. On the topic of bank management, the Narasimham Committee clearly recognized the need to upgrade the organization and management of the banks to make them more globally competitive and to move them away from the bureaucratic banking, which merely channeled funds to public sector institutions and highly protected private sector clients. Although the committee argued that upgrading would be best achieved by leaving full operational autonomy to individual banks, Bery doubted whether this approach could work unaided, especially for weaker banks. Instead Bery wondered Seminar Proceedings: A Summary Rqkrt 33 whether there was not a basic role for public policy to help organize and implement such a major organizational change. Comments from the floor further emphasized the basic difficulties in building a strong financial system when the commercial banks themselves are forced to be so heavily involved in subsidizing public sector activities. For example, someone referred to the basic difficulty, of building a secondary market in government securities when so much of the primary issue of those securities is sold at below market prices on a captive basis. One participant pointed out that it was inherently difficult in such an environment to have any real idea of the basic technical efficiency of the banks because this was obviously not indicated properly by their profitability performance. Other participants questioned whether the technical reforms to which Bery referred could really be effective while the statist culture of the Indian banks remained in place. Someone drew the analogy with the one large Thai bank specialized in mobilizing savings deposits. These deposits had routinely been channeled to the government, but now that the government's own financing needs were much lower, the staff of the bank were having considerable difficulty managing its asset position. Few staff members had any real experience in evaluating commercial lending prospects and hence, they were tking the safe route of channeling large volumes of resources through the interbank money market. Similarly, the agricultural bank in Thailand was fairly good at evaluating loans to fanners but was really rather poor at mobilizing deposits. In short, the types of interventions that Thailand and, to a greater extent, India had faced were likely to have resulted in the emergence over time of one-sided banks that were poorly equipped to operate in a market environment. Some participants expressed surprise that the Indian reforms had not embraced the privatization option more centrally since this might be seen as the basic precondition for achieving change in the attitudes and culture of banks. Bery's own view was in accord with that of the Narasimham Committee, namely, that private ownership was not an absolutely necessary precondition for successful bank restructuring. Bery observed that some of the Indian public banks were moderately efficient. He pointed out too that India had always had a vibrant private sector and that it was only in the last twenty years that this aspect of its culture had been repressed. In this sense India was more like Latin America than the former Soviet Union: its problem was largely one of reactivating the market-based instincts of its economic agents. Bery pointed out that while there was in India a basic consensus in favor of more private ownership, especially in the area of state manufacturing enterprises, this did not extend necessarily to support the outright sale of institutions, especially those in the financial sector. Concerning capital adequacy, it was also suggested that in this climate of continuing state involvement there was possibly less of a need rigidly to enforce capital adequacy requirements because the government presumably stands as the guarantor for the loans extended to public sector bodies. Bery rejected this viewpoint, explaining 34 Aan R. Roe that because bank loans to state enterprise did not carry formal guarantee', they involve full commercial risks. Thus, loans need to be fully accounted for in the determination of the capital adequacy of banks. In any event, a soft line on capital adequacy could not be extended to Indian banks that have a significant part of their operations overseas. 'Pakistan The comparable case of Pakistan, where reforms in the fmancial sector had started slightly earlier an in India, was explained by Mohammed Janjua, from the State Bank of Pakistan. In Pakistan, lhe serious efforts to liberalize the country's traditionally highly regulated system began in earnest with a comprehensive medium- term program of trade liberalization and tariff reduction and rationalization initiated in 1988. This set the scene for the financial sector reform (FSR) program that had commenced one year later. The foreign exchange liberaization, in turn, began by slimming down the list of restricted imports and replacing many quantitative barriers to imports with still high import tariffs. In February 1991 these reforms were further extended to cover several important aspects of the country's capital account transactions. This new and more liberal regime included the following: Allowed direct foreign investment in most Pakistani businesses * Allowed dividend transfers outside the country without prior permission * Enabled foreign residents to hold special convertible rupee a' counts for the purchase of shares on the stock exchange * Allowed unlimited domestic borrowing for worldng capital purposes by foreign- controlled maufacturing companivs * Allowed overseas borrowing on unrestricted terms for business setup and expansion where no govenmment guarantees are involved * Permitted foreign currency accounts (FCAs) by residents of Pakistan on the same basis as nonresidents * Introdaucw dollar bearer certificates with a one-year maturity, and five-year foreign currency bearer certificates denominated in dollars, marks, pounds, or yen, which supplemented the rupee-denominated foreign exchange bearer certificates (FEBC) first issued in 1985. Semnar Proceedings: A Smmry Report 35 Janjua noted that these reforms, although relatively recent, had already begun to affects aspects of Pakistan's balance of payments performance. For example, foreign direct investment in 1991-92 amounted to US$335 million as against US$246 million during the previous year. The new freedom on portfolio investment attracted inflows of some US$219 million during 1991-92. The international borrowing by Pakistani companies increased -to US$559 million in 1991-92, compared to US$158 million in the pre#ious year. The new dollar and foreign currency bearer cerdficates attracted sums of US$53.1 and US$62.7 million, respectively, by December 1992. Finally, the widening scope of the foreign currency accounts system since February 1991 that included accounts of resident Pakistanis resulted in a very sharp rise in the total amounts outstanding. From a figure of some US$2,346 million at the end of March 1991, the total rose to over US$3,800 million by the end of December 1992. Janjua noted a sharp contrast between some aspects of the starting conditions for financial sector reform in Pakistan relative to, for example, Mexico and Malaysia. In particular, Pakistan started out with a large fiscal deficit and was having great difficulty in reducing it from the equivalent of 7 percent of GDP to 5 percent. The country's large debt overhang as well as its substantial external and savings, and investment gaps were also unfavorable conditions for reform. The FSR program had five main components. First, improvements had been made in domestic debt management with greater reliance placed on market-determined intrest rates. This, however, had exerted a significant negative impact on the govermnent's own difficult fiscal position, because the resulting increase in interest charges raised the deficit by about 1 percentage point of GDP. Second, bank credit allocated by the governent was significantly reduced. However, as one participant noted, because some of the directed funds still go to finance the deficit, the reforms so far introduced have failed to eliminate the lidden subsidies organized through the financial system. Hence other borrowers were undoubtedly having to pay unnecessarily high interest rates. This is merely one example of why it is advisable to achieve a serious fiscal adjustment before underking a major liberalization of finance. Tbird, the State Bank had shifted its method of monetary control from one based on quantitative limits to one dependent on indirect methods using market mechanisms. To this end an open market policy had been put in place. The problems so far experienced with this in Pakistan were those associated with the thin money markets in the country and especially with the very smal size of secondary market activity for treasury bills and other short-term secunties. As in many other counties that had long been dependent on direct methods of monetary control and the captive sale of government securities, banks and oDher institutions were too wedded to the practice of buying securities and holding them ftrough to maturity. It was difficilt for an active market to emerge while ftese behavior patterns persisted. One participant 36 Alan R. Roe noted that the thin secondary markets are likely to result in undesirably large movements in the prices of the underlying securities. Fourth, in Palastan many banks had been nationalized about twenty years ago. The country now recognized some advantages of greater private ownership and had already moved to privatize two state banks. In response to questions, Janjua confirmed that the three main banks still under state control accounted for about 50 percent of all bank deposits. Two of these, however, are expected to be privatized. Finally, as in most other countries at the seminar, the liberalization process had been accompanied by serious efforts to strengthen the regulatory and supervisory framework governing banks and other financial institutions. A particular featr of this had been to bring nonbank fmancial institutions under explicit central bank control for the purposes of regulation and supervision. Many of the comments from the floor focused on the knock-on implications for ficial reform of the continuing large fiscal deficits in Palistan. Participants noted, and Janjua agreed, that these deficits were caused in large part by high military expenditures and high debt service payments: both difficult elements to cut. Others observed that the import liberalization program had also led to a significant loss of revenue in spite of some surge of imports. More generally, it was pointed out that before liberalization, in Pakistan and elsewhere, external creditors had often succeeded in receiving the payments due to them but only because domestic creditors were required to forego the payments due to them altogether or accept very low rates of interest. With the end of financial repression, this implicit seniority of creditors must change but, unless properly managed, can result in a new type of risk to external creditors. Looked at from another perspective, very large fals in the real interest costs of the government are a possibility if fiscal reform is put fmly in place. Mexico in recent years is a leading example of a country that has derived substantial benefits in this way. Pakistan, by contrast, has left itself with substantial difficulties for successful financial reform because it has not established this preconditon of fiscal adjusinent. Finandal Systems and Macroeconomic Stability: Experiences from Chile, Malaysia, and Japan Various topics involving the interdependencies between macroeconomic policies and financial sector reform emerged at some point in most sessions of the seminar. For example, the successful opening of the capital account in countries such as Mexico and Malaysia has clearly been possible because of the generally sound macroeconomic policies pursued in these countries in recent years. By contast, some of the difficulties with financial reform , for example, Paldstan relate very directdy to certain weaknesses in macroeconomic policies and especially to the problems of reducing the fiscal deficit. In this section of the paper, we summarize three presentations and Seminar Proeedings: A &mmary Report 37 associated discussions that focused more directly on the topic of the links between macroeconomics on the one hand and financial reform policies on the other. First, Guillermo Le-Fort, Research Manager of the Central Bank of Chile, presented a broad-based analysis of the links between financial systems and macroeconomic stability in Chile. Then, Ln See Yan, Deputy Governor of Bank Negara, Malaysia (BNM), discussed a number of key aspects on the same broad topic from the perspective of the Malaysian experiences. Finally, on a somewhat narrower front, Takashi Kanzald, Chief of the Bank Supervision Department of the Bank of Japan, considered the problems created in the banking sector by Japan's speculative bubble in asset prices and the eventual collapse of that bubble. Chil Le-Fort's remarks focused on three main aspects of recent Chilean experiences: * The effects of the aborted mid-1970s liberalization in creating the crisis of the early 1980s; * The impact of the crisis on banks and the post-crisis financial reforms; and * The general lessons from this concerning the lnkages between macroeconomic stability and finanCial sector reform.2 Liberaizoon and Crisis. Mr. Le-Fort noted that the Chilean financial system in 1973 was highly repressed, with most banks under state control, negative real rates of interest, and ubiquitous quantitative controls on credit. The 1974 reforms involved the privatization of banks, liberalized interest rates, and the licensing of new financial institutions. Capital controls, however, were retined -until mid-1979 and were not completely removed until April 1980. The financial liberalization program in trn was designed as a part of a broad program of economic deregulation that encountered a variety of extremely senous and ultimately fatal problems. First, in the period 1975- 78, dte system sustained the collapse of the whole savings and loan system, several unregulated financial institutions, and one medium-sized bank, namely Banco Osorno. In November 1981 the authorties had to begin rescuing banks through takeovers, 2. Le-Fort's paper as cirulated w participants also included a substatial section on various aspects of the post-reform fmancial system. This, however, was not discussed during the seminar and is not iuwded in this paper. 38 Ala R. RoE accounting for about 8 percent of private bank deposits. This process reached its peak by January 1983, by which time all remaining private sector banks had become insolvent and had been taken over by the authorities. In this early and disastrous experiment with financial liberalization, the comptroler of banks in effect failed to exercise any real regulation of the financial system. In what amounted to a chronic misunderstanding of the true meaning of liberalization, many influential people were prepared to argue that regulation was no longer necessary. Thus, some banks operated with litde or no capital, and the incidence of connected lending within financial-industrial groups was very high. The rescue of Banco Osorno in 1977 made matters worse by creating the false impression among bank depositors that their funds were insured by the government and they had no need to concern themselves with the solvency of banks and the types of uses to which their funds were being put. This event reinforced the conditions for a substantial moral hazard problem and is critical, in Le-Fort's view, i undertanding the crisis. The serious but manageable difficlties of the pre-1981 siution were intensified by the recession of the early 1980s and especially by a serious deterioration in Chile's extemal terms of trade. This rendered an already large current external deficit, fueled by large capital inflows, quite unsustainable, and made some form of macroeconomic adjustment unavoidable. As this became clear to almost everyone and capital flows began to reverse, the bad debts in the banking system accelerated rapidly, especially because of the losses sustained by clients heavily exposed to interest rate and currency risks. Le-Fort noted some of the microeconomic incentives that had allowed the initial basic problems of that period to become so severe. From the viewpoint of many debtors, the early elimination of most of their equity meant that they had nothing to lose by borrowing more and hoping for some recovery. From the viewpoint of the banks, their strong interlinling with industrial conglomerates put them under strong pressure to grant risky loans to benefit the conglomerate as a whole. Bank regulators, as noted earlier, had neither the weapons nor the authority to stem the tide. Thus, by September 1982, loans in arrears net of provisions amounted to no less than 181 percent of the total banking system's capital and reserves. For a while, real interest rates were extremely high because of the reversal of foreign capital flows, the high demands from distressed borrowers, tighter monetary policy, and strong expectations of a peso devaluation. When that devaluation actually occurred in mid-1982, the risk premium built into interest rates was replaced by realized foreign exchange losses incurred by many of the clients of the banks. This, further compounded the extent of the banks' financial distress. Le-Fort argued that three factors in particular had created the conditions for the crisis: * The disassociation between the risks attached to bank loans and the yields paid on bank liabilities; Seminar Proceedings: A Sumnmay Report 39 * The excessive optimism about the economy's growth prospects including overinflated views about the real value of assets and the real burden of accumulated debt; and * The strong interlinking of banks and the large conglomerates that rendered the capital of most banks virtually marginal witiin the context of decisions made by the conglomerates. Cleanup and Reform. Le-Fort explained that in dealing with the crisis, the authorities had four possible options: The liberal option of allowing widespread bank failure; the inflationary option that would, in effect, melt the debt; a monetary reform designed to write down the value of depositors' claims on banks; and the socializaton of losses through a debt restrucuring program financed by the central bank. This last option was the one actually chosen. The core of the cleanup operation was the purchase by the central bank of non- performing bank loans at par in exchange for promissory notes of the central bank. This required banks to eventually buy back some part of these bad loans using their own profits. Le-Fort indicated that the extent of such a buy-back was likely to be small. Additionally, the central bank subsidized the interest rates on the promissory notes by offering a higher rate than its normal borrowing rate. The approach that was eventually chosen avoided the collapse of asset prices that would have been associated with any attempt to simultaneously sell a large part of the assets of the bad debtors. It also avoided a run on the bands. This approach, however, had the serious disadvantage of puttg litde immediate pressure on debtors to settle their obligations, and it may have created some expectations of debt forgiveness. It also involved a substantial short-term transfer from taxpayers to banks and debtors and allowed many firms to remain in business even though they had negative capital. The major change in macroeconomic policy that accompanied this rescue of the banks involved a sustained real currency depreciation that more than doubled the relative price of tradables in the years between 1982 and 1988. There was also a considerable number of fiscal and quasi-fiscal implications. Most directly, there were two implicit subsidies involved in the portfolio purchase program. The first arose, because the new promissory notes carried a higher interest cost to the cental bank than its average cost of funds at the time when the notes were issued. The second arose from the gap between the interest rate on the commercial banks' obligation to repurchase their portfolios and the discount rate applicable to bank profits. For banks under govermnent control, the central bank had to absorb these losses, and it also incurred some subsidies in the process through its efforts to encourage new shareholders in the process of privatizing some of the banks. Additionally, when the real currency depreciation started in 1982, the government decided to help foreign currency debtors through a system of exchange 40 Alan R. Roe subsidies (that is, sales of dollars at an exchange value substantially lower than the prevailing official rate). This arrangement, operated through the central bank, resulted in substantial quasi-fiscal losses on its books. These losses were further increased by the central bank provision of foreign exchange arrangements that linked sales of foreign exchange to central bank repurchase agreements at the prevailing exchange rate. which, at the time, was depreciating strongly. ' The estimates of the cumulative costs of these various interventions amounted to somewhere between US$7 and US$9 billion. The ongoing annual deficit of the central bank is about 1.5 percent of GDP. It is financed through the issuing of new debt and through inflation tax collections- With falling rates of inflation in the recent past, the contribution of the latter has been declining. It is estimated that the domestic debt to GDP ratio is currently about 40 percent, and that an annual GDP growth of about 4 percent is sufficient to keep that ratio stable. Because Chile's growth prospects are somewhat better than this, the debt ratio can be expected to decline. Because of the clear diagnosis that inadequate banking regulation was strongly implicated in the Chilean crisis, Le-Fort emphasized that the establishment of a new framework of banldng legislation, especially the Banking Law of 1986, has been a very importat part of the reform package. A central and unusual feature of these new arrangements is that the Comptroller of Banks must publish in the press at least three times a year its opinion on the state of affairs of each banking instituion. In their traditional activities of intermediating between deposits and loans, the commercial banks are now subject to regulation in a number of aspects. These include limits on loans to individual debtors, limits on loans in particular currencies, and limits on connected lending. Provisioning arrangements have also been gready tightened- Banks are now forced to write down non-performing loans that have higher than normal risks and to inject additional capital to compensate for such provisions. Capital adequacy is now assessed relative to the economic rat.er than the book value of the banks' assets and in that sense is sensitive to the differential riskiness of different types of portfolios- A parallel limit on bank debt to equity is used as a further indicator of bank viability. The new law is also very explicit about the basis on which banks can participate m nontraditional activities such as stock trading, mutual and investnent fund management, and leasing. Banks can perform these activities but must be operated through subsidiary companies using segregated capital. For example, any bank using part of its capital to set up such a subsidiary must deduct this from its own capital when calculating capital adequacy and debt to equity ratios. Lessons in Macroeconomics and Financial Reform Linkages. The Chilean experience provides very important lessons: The reforms of the 1970s eventually led to the worst recession in the country since the Great Depression, as well as to the insolvency of the whole banking system. By contrast, the reforms of the 1980s are generally regarded as successful. Le-Fort described the crisis of 1981-82 as being created through the interaction of inappropriate macroeconomic policies on the one Seminar Proceedings: A Summay Report 41 hand and significant structural weaknesses in the financial system on the other. It follows that actions in both areas were necessary to ensure the improved performance of the more recent reforms. Regarding the macroeconomic dimensions of the problem, Le-Fort noted that the reforms introduced in the mid-1970s generated three types of macroeconomic disturbances and inconsistencies. First, there was the shock to productive sectors when the current account of the balance of payments was opened and competitive imports became more available. Second, but somewhat later in time, was the shock associated with the liberalization of the capital account and the greatly expanded availability of foreign borrowing. Third, there were the problems arising from the inconsistencies of some aspects of macroeconomic policies. In particular, a continued rapid expansion of aggregate demand coexisted with high domestic interest rates. With very liberal capital movements, high domestic real interest rates did little to discourage expenditures, and monetary policy interventions were largely impotent. Equally, although the narrow fiscal accounts were in surplus after 1978, the contingent fiscal transfers to the private sector were perceived to be large; hence, many borrowers believed that they would be able to defer debt service or be bailed out. Finally, the combination of an appreciating real exchange rate and a rapid growth of aggregate demand produced a large current account deficit that was unsustinable on a long-term basis. The financial sector origins of the problem can be traced to the particularly risky behavior of financial market participants (borrowers and lenders), much of which was derived from problems of moral hazard and agency in the financial system. While the explicit deposit insurance was perceived to be available, Le-Fort noted that the moral hazard problems associated with this could have been avoided if there had been' adequate bank regulation and supervision to defend the solvency of banks. Hence, deposit insurance does not need to be eliminated. Significantly, the later Chilean reforms preserved an arrangement for explicit deposit protection. The final link between the macroeconomic and the structural dimensions involved the macroeconomic problems associated with excessive growth of aggregate demand that can be blamed partly on structural weaknesses in the financial sector. In particular, while economic agents believed that the government was insuring their banks deposits as well as their foreign currency exposures, the implicit transfers of wealth related to this insurance, helped to both foster excessive aggregate demand and weaken the effects of high interest rates. An important issue raised in the discussion that followed Le-Fort's presentation concerned the scope of bank supervision in dealing with fundamental sources of fnancial instability of the types experienced in Chile Participants argued that in the laissez-faire atmosphere that prevailed in Chile at the begmining of the 1980s, even a good system of bank supervision would have been powerless to stop what occurred. The microeconomic decisions that were made at that time were entirely rationale given the then-prevailing set of macroeconomic policies. Even a good system of bank 42 Alan R. Roe supervision could have succeeded in arresting the crisis only if it had been able to, for example, end the policy of fixing the exchange rate something that is clearly beyond its powers. More generally, participants noted that the domain and influence of even a good regulatory system is limited to that which is consistent with the prevailing policies of the country. Such a system certainly cannot go against, or compensate for, serious errors in macroeconomic policies. In relation to the specific issues raised by Le-Fort on the fiscal deficit, some participants were uneasy about the notion that the contingent inuance provided through various acts and statements of the government could be construed as some form of hidden fiscal deficit. For at least one participant, it seemed that the operational significance of this concept was very limited because it was only after the event (and maybe tautologically) that one could establish that such insurances were actually available! Others noted the dilemma that exists for fiscal authorities in a country that is attracting large capital inflows in the way Chile did, and indeed still is. Specifically, while rapidly growing aggregate demand indicates an overheating economy and suggests the need for a fiscal tightening, the rapid accretion of international reserves creates very considerable pressures to increase spending. A further point on bank supervision, which was also made earlier in the sinar, related to the increased problems of both banks and supervisors in a newly liberalized enviromnent. It was pointed out that in Chile before 1974 real interest rates were generally negative and, with credit rationing as the norm, bankers did not have to do much work to allocate credit. At the same time as in India, bank supervisors could concentrate on monitoring certain quantitative targets. With liberalization, all this changes. The bankers have to become much more alert to the commercial returns and risks of competing loans, and supervisors have to focus much more on the quality and risks of bank portfolios if the solvency of banks is to be protected. The reforms in Chile during the 1970s obviously failed to consider this point. However, even today this message may not be fully appreciated and one participant noted that his country was still being advised to forego banking regulation and supervision during the process of fnancial system reform! Malaysia A further perspective on some of the issues discussed by Le-Fort was provided in the presentation given by Lin See Yan, Deputy Governor of Bank Negara, Malaysia (BNM). Lin See Yan commented that by contrast with most other countries Malaysian economic and financial policies were now guided by a clear long-term (twenty-year) vision of what the country wished to achieve. At the basis of this vision was the objective of sustained long-term growth at an average anmual rate of about 7 percent, with inflation maitained at about 3 to 4 percent. Furthermore, the BNM, since its Seminar Proceedings: A S&emar Report 43 establishment in 1959, took the view that long-term macroeconomic stability in an environment of considerable economic change required a financial system that was both diverse and stable. Because only a very simple financial sector existed in the early 1960s, Lin See Yan explained that it had been a fimdamental task of the BNM to establish and nurture almost all the new institutions now operating in the sector.3 He then considered two particular aspects of the BNM management of the financial systm. The first was the reorientation of the sector toward more support for the private sector. The second was the measures required to respond to the economic and banldng crisis of the mid-1980s. Aspects of Financial Sector Reonns. Lin See Yan noted that the BNM's pro- active role in financial sector development had involved problems that other counties shared. Specifically, for about twenty five years the system had the strong need to mobilize nonnflationaxy funds for government use Only after the serious economic deflation of the mid-1980s did it become fully apparent that the government could no longer be the engine of growth and that a far greater private sector involvement would be needed to achieve the economy's ambitious growth targts. Since then, authorities have made many attempts to downsize the government and leam some of the lessons from Eastern neighboring countries, especially Japan. In particular, the concept of -Malaysia Limited" had emerged with the government in effect benefiting as a shareholder in most economic activities by extracting approximately 30 cents in taxes from each US$1 of profit realized. Seen narrowly from the perspective of the financial sector, this change generated the need for far greater diversity in the range of financial institutions and instruments and, above all, for the development of more acive and efficient capital markets. The Malaysian reforms to achieve these broad objectives had had some of the same stop-go characteristics of the Chilean experience, although the intnsity of the drama was somewhat lower. At a relatively early stage, officials recognized that the objective of capital market development would require the end to many of the traditional govermnent controls, especially the controls on interest rates. These were freed in October 1978, but because many of the preconditions for effective free markets in financial instruments were absent at the time, the liberalization proved to be short- lived. Controls were reintroduced early in the 1980s mainly because the reforms at that stage did nothing to address the problems of limited competition in the banking sector. Thus, interest rate movements tended to be asymmetrical: rates rose freely in 3. For details see Lin See Yan, 'The institutional Perspective of Fmancial Market Refimr MTh Malaysian Experience." Shakil Faruqi (editor), Fznania Sector Refonn in Asx and Latn Ameican Cowitries: Lessons of Conpaaive Experience,) EDI Seminar Series, No. 340/073 Washington D.C.. 1993. 44 Alan R Roe response to tighter monetary conditions but did not fall correspondingly when conditions were relaxed. Eventually the government introduced the base lending rate (BLR) system in November 1983. This anchored the lending rate of each bank or finance company to a declared BLR, which was calculated by reference to the cost of funds to the institution concered, plus a markLp that covered the costs of statutory reserves and overheads but not bad debt provisions. Lin See Yan explained that, following some experimentation with the manaeent of the BLR system, all instituons are free to set their own BLR and their own lending rate, but they must adhere to an unchanged formula. Thus the central bank retains some influence over rates even though instittions are decontrolled. In Lin See Yan's view, it was a good example of a flexible form of intervention, capable of redressing the limited enthusiasm of banks to indulge in real price competition. Over time it has certainly led to a reduction in the margins of the Malaysian banks. Resolving Banking Sector Distress. The second main topic discussed by Lin See Yan was the serious banking sector crisis experenced in Malaysia in the mid- 1980s. He noted that this could be traced to the collapse of the prices of most of Malysia's main commodity exports and the resulting sharp decline in the economy's overall growth rate. The banidng system, as in the case of Japan, was seriously affected by the collapse of property and stock market prices and by the abrupt end to a long period of very rapid deposit growth. Specifically, from a rate of growth of 20 perent in 1984, the growth of deposits declined to less than 4 percent on an anmal basis by the end of 1986. Loan growth, however, did not contract commensurably so the loan to deposit ratio in the banks rose sharply to almost 100 percent and liquidity became extremely tight Banks also susained a rising burden of non-performing loans and filling profitability as their customers faced the triple shock of lower asset prices, higher itrest burdens, and sluggish or negative growth in their income flows. Bad debt provisions, equal to 3.5 percent of loans in 1984, increased to almost 13 percent by 1987 even before taking account of the significant failure of some deposit taug insqtions to fuly provision for bad debts. AltIough the incidence of these fnancial sector problems in 1987 was not high, public confidence in the system was seriously eroded when some illegal deposit taking institutions filed and then large-scale withdrawals were made from the deposit taking cooperatives (DTCs), which with over one million depositors occupied an extremely importnt role in the overall deposit mobilization of the financial system. The run on the DTCs prompted the enactment of emergency legislation and the suspension of the activities of most of the thirty five DTCs. Twenty-one of these were subsequently shown to be insolvent. Lini explained that the BNM reaction to this crisis was prompt and multi- fceted. The cisis arose from a series of macroeconomic disturbances and its resolution also had significant macroeconomic feedbacks. For example, the bank acted Seninar Proceedings: A Summary Report 45 quickly to loosen monetary controls to ease the liquidity constraints. It also adopted or strengthened a whole range of regulatory and supervisory measures to further limit bank insolvency. Lin, however, concentrated his remarks on the measures that were taken to redress the accumulated problems of bank distress and to rescue at least some of the banks that found themselves in difficulty. The essential point on this topic was that the BNM adopted a wide variety of differenr interventions to deal with different aspects of the problem. These approaches, however, were tailored to the needs of different banks. The BNM needed to get very close both to the banks and to the private sector in general. Institutions that had preserved their capital were allowed a relatively free hand, but a tough line was taken witha banks that had lost capital. Although the resolution of the problem in Malaysia did not involve the wholesale state takeover of banks, as had been necessary in Chile, nonetheless the law was changed to allow BNM to hold bank shares. This in turn helped BNM expedite the amalgamations of some banks by becoming a temporary owner of banks on its own account. Since this might have involved a conflict of interest with the BNM's regulatory responsibilities, distinct organizational arrangements were set up to deal with the BNM's shareholdings. Six further examples of the range and flexibility of the BNM approach are as follows: * In the cases of the worst affected banks, the BNM required the replacement or existing managers and also organized the injection of new funds to cover accumulated losses; * The Bank called on existing shareholders to subscribe new capital through rights issues supplemented as necessary by direct injections of funds from the BNM, with options available to existing shareholders to buy back BNM shares at a later stage; * For DTCs that had relatively modest problems, the BNM provided loans on relatively soft terms; * For other DTCs, depositors were assured a US$1 for US$1 payment, but part of this was provided not as cash but as equity holdings in some other licensed financial institution(s); * In the cases of some DTCs, the BNM directly acquired the net assets and liabilities through a small holding company; * The BNM established an Enterprise Rehabilitation Fund to directly support the financial and economic restructuring of some of the productive sector clients of 46 Alan R. Roe the distressed banks, and established an Abandoned Housing Fund to undertake loan recovery for housing loans. Overall, the BNM needed to invest large volumes of funds to resolve the problems of distress but used a variety of approaches to achieve this. It was significant that through the vigorous pursuit of bad loans, the resale of its shareholdings, and in other ways the BNM succeeded over time in recovering a significant part of that investment. Thus, the burden of bad debts of the banks had not had the ongoing implications for fiscal and macroeconomic management that had been noted in the case of Chile. Japanese Banks and the Speculatve Bubble Some of the same issues discussed concerning Malaysia recurred, but in a somewhat different form in the presentation by Takashi Kanzald about the effects on Japanese banks of the recent collapse in that country's speculative asset-price bubble. Kanzaki was particularly concerned with assessing the manner in which the Bank of Japan responded to this crisis in order to defend the stability of the bankdng system and the macroeconomy. The contrast with the Chilean case in 1981 and 1982 is sharp. When the bubble burst in Chile it imposed considerable costs and distress on both the banking system and the real productive sectors. In Japan, by contrast, the authorities seem to have handled the bursting of the bubble in such a way as to achieve a soft landing for both financial and productive sectors as well as for the economy as a whole. What, if anything, does this say about the behavior and performance of the regulators? ln explaining the background to recent problems, Kanzakd noted that the rapid rise in asset prices (especially real estate and stock market securities) during the 1980s prompted intense competition among Japanese banks. The competition focused mainly on a scramble for quantitative expansion rather than portfolio quality. Kanzaki also took the view that the Japanese experiences in this area were part of a global phenomenon caused by the increased openingup of capital account transactions through te 1980s and the resulting globalizaion of many real estate and finacial sector transactions. Specifically, in France, the United States, the United Kingdom, as well as in Japan, the late 1980s saw common patterns of inflationary expectations about asset prices prompted by expansionary monetary policies but perpetuated in all cases by high levels of speculative transactions in these assets. In the case of Japan and with the benefit of hindsight, three types of risk were inherent in this behavior as far as the banks were concerned. The first of these was the credit risk, the risk of relying excessively on the asset values that were apparently securing most of the loans at the expense of proper attention to credit analysis and loan foUow-up and management. The consequences of this were manifest in the Seminar Proceedings: A Summary Report 47 considerable increase in non-performing loans when asset prices began to decline in the early 1990s. The second was the market risk. Interest rate liberalization, implemented gradually over the fourteen years since 1979, raised funding costs for the banks. Many small and medium-size banks increased their exposures to high risk-return securities in an attempt to cover these higher costs. Unfortunately, they frequently lacked the knowledge and experience to make these investments successful. Finally, the managethent risk was associated with the excessively fast expansion of many banks. This behavior pushed more responsibilities onto less experienced branch managements, weakened bank compliance with internal controls, and gave rise to a significant increase in financial irregularities. The very large falls in real estate and stock market prices that have occurred during the past few years have exposed the true nature and extent of these risks. Kanzaki, however, explained that many Japanese banks had a good cushion of revenue reserves to help them withstand the higher incidence of non-performing loans, and their excesses of the past have now been reined in by a series of reform measures guided closely by the on-site inspections of the bank supervisors. First, capital adequacy requirements, as laid down in the Basle Agreement, have been enforced. For many Japanese banks whose unrealized capital gains on security holdings fell sharply as a result of the stock market collapse, this has meant significant restraint on asset growth in an attempt to rebuild Tier-2 capital. Second, the banks have been encouraged to give much more attentioni to the analysis of creditworthiness and the financial condition of their borrowers. In some banks this has been done by giving credit analysis departments greater independence, including the ability to reject inappropriate loan applications. Third, the banks have significantly reinforced their internal audit and other internal controls, including the use of dual responsibility controls. A question from the floor inquired as to the extent to which internal auditors can ensure detection of problems if they themselves are integrated with the general management of te bank. Kanzald confirmed that internal auditors in Japan report to bank managements, but in most cases they provide bank managements with some independent advice about how internal controls are working. Kanzaki agreed that in the final analysis the onus for detecting blatant abuses would lie with Bank of Japan supervisors. Finally, to respond to the extremely rapid expansion of real estate loans by poorlycontrolled subsidiaries of banks in the late 1980s, the banks have been guided to substantially improve the control that they exert over such subsidiaries. Although the combined efforts of the banks and the supervisors had so far avoided any major financial collapse, Kanzald felt that there were still some further important steps that banks needed to take. They needed, for example, to pursue more vigorously the liquidation of collaterized real estate lending and the collection or write- off of other bad loans. They needed to review their own structures in order to conctrate ativities on core profitable banking business and eliminate unprofitable 48 Alan R. Roe activities. Kanzaki also felt that banks would benefit eventually from a fuller disclosure about their own affairs, including information that could communicate something about the quality of their underlying assets. He noted that under recent new arrangements, twenty-one of the big money center banks were now required to disclose loans where interest is past due for six months or more. Although the regional banks faced less demanding guidelines, the general move in the arrangements was toward more disclosure. In response to questions, Kanzaki, however, felt that it would not be appropriate in the Japanese case for the bank supervisors to adopt the very explicit publication of status reports on banks in the manner in which this is now done in Chile. He felt that this might precipitate too great an influence on banks in difficulty and cou .d compromise the confidence in the relationship between banks and their supervisors. Kanzalk then used the experiences that had followed the bursting of the bubble to draw out several main conclusions about the manner in which bank super isors need to conduct their work in Japan and elsewhere in tne 1990s. First, it was increasingly important in his view for supervisors to conduct regular and full scope on-site examinations. The heightened importance of portfolio quality could not be over emphasized, but this could not be monitored without regular on-site inspection. Kanzaki argued that such inspections ought to accord a great deal of discretion to the supervisors about the precise manner in which they should be conducted. Supervisors needed the freedom to make judgments that were both fair and fully cognizant of the prevailing macroeconomic situation as well as the circumstances of particular banks. Again this raised questions like those raised in the Chilean context about how far supervisors can go in discouraging banks from loans that are viable but reflect poor macroeconomic policies or speculative epidemics. Kanzaki's advocacy of an approach to supervision that includes some flexibility to changing macroeconomic conditions accords with the earlier conclusion about the severe limits on the scope of what supervision can actually take on. Finally, Kanzaki strongly emphasized the need for supervisors to conduct their work on a consolidated basis. More specifically, supervisors need to monitor the mnancial condition of all parts of a bank, including all its subsidiaries, because, whatever the organizational anrangements, a contagion of risks between different arms of the same bank is unavoidable. Furthermore, this consolidated approach certinly needs to include the overseas branches of a bank. The Bank of Japan itself has greatly irncreased its on-site supervision of the overseas branches of Japanese banks in recent vears- It sends examiners abroad to assess loan quality and other aspects of bank performance. In the case of medium-size banks, examination teams are sent four times 'ach year to conduct on-site examinations of several selected banks in each major financial market. The Bank has also been progressively increasing its exchange of supervisory information about individual banks, with overseas monetary authorities. In the case of the domestic nonbank subsidiaries of the banks, the Bank of Japan is not authorized to conduct on-site examinations directly. Thus, it relies on the Semninar Proceedings; A Snary Report 49 detailed information it receives about such subsidiaries when conducting its examination of parent banks. The discussion following Mr. Kanzaki's presentation did not definitively identify the reasons why the Japanese banking system had proved to be resilient to the shocks associated with the major fall in asset values of the early 1990s, while the. Chilean system had proved so vulnerable to the shocks that it had faced one decade earlier. It was noted, however, that the Chilean financial system itself had shown far more ability to withstand shocks, such as the large 1989 fall in the price of copper, after it had implemented the banldng reforms of the mid-1980s. This indicates the considerable importance of such reforms in building the resilience of the system. It is presumed also that the avoidance of chronic macroeconomic management errors in the Japanese case, as well as the absence of some of the Chilean strucural weaknesses in the financial system, have a large part to play in explaining the differences. Bank Supervision in an Era of Deregulation: The Indonesian Case The final substantive topic of the seminar extended the discussion of bank supervision by considering the role and processes of bank supervision in a deregulated environment. The topic was introduced in a comprebensive paper presented by Binhadi, Managing Director of Bank Indonesia. Binhadi organized his remarks around three main themes, namely, the nature of the Indonesian deregulation process, the impnact of this on the banks and the macro-economy, and the changing nature of bank supervision in the post-regulation era. Financial System Dereguton Binhadi explained the reforms that had been introduced both in the country's management of foreign exchange and in its operations of the monetary and banking systems. The first of these two elements of reform had been initiated in 1967 with the establishment of a trading center for foreign exchange transactions, namely, the Foreign Exchange Bourse. As early as 1970 the restrictions on the holding, selling, and purchasing of foreign exchange were removed, although for some years export proceeds still needed to be surrendered to a foreign exchange bank. From 1978 to 1982 the pegging of the rupiah to the dollar was replaced by a more flexible policy of pegging to a basket of currencies. From 1982 onward, exporters were no longer required to surrender export proceeds, and any tansactions using these proceeds or involving the purchase of imports could be arranged through foreign exchange banks. Then in 1989, the authorities abolished the Boarse and established new arrangements, whereby all foreign exchange transactions between Bank Indonesia and 50 Al4nR. RRo the foreign exchange banks were done through the foreign exchange dealing room. At about the same time, the authorities removed the restrictions on foreign borrowing by domestic banks but introduced prudential arrangements (a net foreign exchange open position providing daily squaring for the banks) designed to minimize bank risks. Officials also adopted a variation in the managed floating exchange rate system by defining the rate set by the central bank as an indicative rather than a mandatory one. h 1991 the authorities introduced various types of swap facilities and began to refine the central bank intervention procedures in the foreign exchange market. Intervention, however, is still organized around the guidepost of the indicative rare and the managed float. The monetary and banking reforms came somewhat later, starting in 1983 with the removal of interest rate controls on state banks and the elimination of the quantitative ceilings on bank credit. These reforms in turn made it possible for Bank Indonesia to begin using indirect methods of monetary control based on reserve requirements, open-market operations, and central bank discount facilities. Partly to support this development, a variety of money market instruments were developed m the mid-1980s. These included Bank Indonesia cerdficates (SBIs) and money market securities (SBPUs). As Governor Mooy noted at the beginning of the seminar, these various early reforms had been successful in stimulating the growth of the economy and also in assisting its diversification away from oil dependence. Partly because of this success, the ongoing development and improvement of the bankig industry has come to be regarded as one of the keys to the continued rapid growth of the economy. This was the background to the most substantial series of deregulation measures in the financial, monetary, and banking sectors that were introduced in October 1988 (that is, the measures referred to as PAKTO 27). The general aims of these further reforms were to improve resource mobilization, to firther improve the efficiency of bankg instiutions partly through enhanced competition, further develop non-oil export sectors to improve the effectiveness of monetary policy management, and deepen the domestic capital markets. Resource mobilization was pursued by expanding the opportunities to open new bank and nonbank financial intermediaries (NBFI) branches, by giving banks more freedom to develop new savings schemes, and by allowing NBFIs more freedom in deposit mobilization. Export diversification was pursued by opening opportunities for new foreign exchange banks, establishing improved swap facilities, removing the offshore borrowing limits for banks, and requiring joint banks and foreign banks to provide a substantial part of their total credits for export purposes. Improved bank efficiency was pursued by allowing banls to create new products for fund mobilization and credit expansion, and also by allowing state enterprises to deposit a substantial part of their available funds with the private banks. Monetary control was enhanred by unfying the reserve requirements for banks and NBFIs. Capital market development was fostered by equalizing the tax treatment of income from deposits with that from Seminar Proceedngs; A Summary Report 51 securities. Banks and NBFIs were also encouraged to issue new shares through the stock market. In December 1988 the government made what Binhadi called a giant step toward stronger capital market arrangements. At that time it established a private stock exchange, opened the exchanges outside Jakarta, and established finance companies and other institutional arrangements to support the market. Binhadi noted, however, that these arrangements had all gone ahead without any real changes in the Bank Indonesia policy of requiring the allocation of liquidity credits for particular purposes. The inherent inefficiencies associated with such arrangements, however, are recognized by the authorities as are the difficulties they cause for effective monetary control. Hence, in the PAKJAN 29 measures of January 1990, the authorities sought to make some improvements in national credit arrangements. Above all, the policy now is to gradually reduce the scope of liquidity credits and allow the commercial banks more influence in providing fumds and allocating credit. Finally, many of the reform measures just described were consolidated in legislation in the new Bankdng Act, Insurance Act, and Pension Fund Act, all of 1992. The Banldng Act defined the scope of banks in an essentially universal way and extended to foreigners the rights to buy commercial bank shares through the stock exchange. The Effects on the Bankcng Sector and the Macroeconomy This extremely comprehensive package of reforms has had both quantitative and qualitative lications for Indonesia's commercial banks. In terms of quantity, between the date of the main deregulation measures in 1988 and December 1992, the number of new banks in the country rose from 122 to 221. This included 144 pnvate national banks, 10 foreign bank branches, and 20 joint banks. Bank branching also expanded substantially with the 1988 number of 2,045 expanding by no fewer than 3,681 new branches in the same period. Together with some new smaller rural balks and village units, the country can now claim about one banking office per 10,000 people, a doubling of the coverage relative to the pre-reform period. Savings mobilization has also increased considerably with banks experiencing 44 percent and 51 percent increases in their deposits in 1989 and 1990, respectively, before the leveling off associated with tighter macroeconomic policies in 1991. In the same two-year period, total bank credits increased by 37 percent and 51 percent, repectively As for the qualitative dimension, there has been a major increase in the range of bank products on offer as each bank has tried to develop its own products to attrct customers. The number of different savings schemes, for example, has increased to a total of about 140. The products available in the foreign exchange markets have expanded to now include swaps, forward contracts, options, and margin trading. The 52 Alan R. Roe interbank money market, the short-term securities market, and the foreign exchange market have aLl seen very large increases in their trading volumes. As in Malaysia and Japan, this rapid expansion of the sector and its qualitative deepening has contributed to a significant shortage of trained personnel. One unfortunate consequence of this has been considerable poaching from bank to bank. In spite of an intensified human resource development effort initiated before deregulation began, shortages of qualified staff have remained a problem. The impact of the reforms on macroeconomic performance has also been considerable. As Governor Mooy noted earlier, the high growth rates achieved in 1989 and 1990 (more than 7 percent in both cases) were welcome but also symptomatic of an overheating of the economy which accelerated inflation. At the same tine, the diversification of the economy has been well served by the reforms, after slow growth in 1990, non-oil exports increased by 22 percent in 1991 and 21 percent in 1992. The greater openness of the Indonesian economy in general was indicated by the rise in the export to GDP ratio, from 18.2 percent in 1986 to 25.4 percent in 1991, and by increases of a similar order of magnitude in the ratio of imports to GDP. Finally, the confidence placed in the economy following the reforms is indicated by a rise in annual average investment approvals (foreign and domestic), from US$1,311 million in the five years from 1983-87 to no less than US$7,122 million in the next five years to 1992. liationsfor Bank Regulanon and Supervision Binhadi explained that the essential philosophy of bank supervision had remained unchanged after deregulation. The essential objecdves were to sustain a sound and efficient banking system that could retain the confidence of the public; grow at a satisfactory rate; and support economic development and the implementation of sound monetary policies. The scope of bank supervision, however, has become very much wider as a result of thre expansion of the numbers of institutions and the range of financial products. Bank Indonesia's strategy to deal with te new situation was intoduced in February 1991, and has since been consolidated in law through the new Banking Act of March 1992. The strategy incorporates what Binhadi referred to as six main areas, as follows: * Improved prudential regulation as a guide for bank operations. This is fundamental and covers rules about the licensing of new banks and branches, including the requirements for owners and managers; improvements in operational guidelines such as capital adequacy, asset qualky, and provisioning; and an improved rating system that complements the quantitative checks on banks based on the CAMEL system, with further qualitative checks on bank Smina Proceeins:. A SwMwy RAi 53 soundness.4 Under the terms of the 1992 law, licenses for new banks and branches are the responsibility of the Minister of Fmance on the recommendation of the central bank. The information flows to support bank supervision are ensured by a legal requirement that banks shall submit all inlformaton and clarifcations required by the cental banlk All such information, however, is to be kept confidential. * Supervisory monitoring as an early warning system. This relates essentially to improved information flows on matters such as bank finances, peer group analysis, and economic sector development, all designed to assume prompt adjustments to the rating of any bank and corrective actions as required. This has involved improved computerization and telecommunication systems linking the banks and the supervisors. e Improved proceduresfor bank amirnawioL This has included systems of asset evaluation, flexibility to anticipate the consequences of new banking products, and the quality and timeliness of reports. 3 Discussion th banks . Improvements here were sought to enhance the benefits of periodic discussions between banks and superviso. The objective has been to improve the openmess of such discussion by making full use of the early waming system and the results of on-site e tions to achieve the highest possible level of bank cooperation in developing proposed solutions to problems. This open attitude to dicssion is also seen as one way of amending the thinking and the attitudes of bank owners and mamgers in the interest of protecting bank soundness. * Enforcement of sanctions, including cease and deist orders. Sanctions available now include fines for breaches of regulations, but also various types of instructions to banks to change their organization or behavior. For example, banks can be asked to restrict the development of new brancbes and prducts and to make changes in their own managemet In serious cases, a bank can be instucted to merge with a second bank or to transfer all or a part of its bank shares to new shareholders. All these powers were legislated in the 1992 Act, which also gives the central bank the ultmate right to recommend the revocation of the license of a bank to the minister. Criminal sanctions are also available to deal with cases of fraud and other breaches of legal duties. 4. CAMEL means capital, asset quaity, management, earnis, and liquidity. 54 AlmnJR Roe Support to bank efficiency and effectiveness. This has maily involved certain institutional developments, including the establishment of a clearing house, a centralized credit information system, money and foreign exchange markets, and a code of conduct for bankers linked to the establishment of the Indonesian Bankers Institute. Binhadi recognized that the various changes in the strategies and techniques of supervision required an appropriate number and quality of bank supervision staff with high levels of skill, dedication, morality, and integrity. Additionally, the substantial widening of the geographical network of banks required that a greater decentralization of bank supervision was also required. The incmased numbers of supervisors needed had also necessitated attention to a good recrulument system, as well as to the selective use of other types of expertise to supplement the supervisory skills available. A final aspect of the necessary improvements is related to the greater international cooperation among the supervisors from different countries. Indonesia was very alert to this need and participated actively in a variety of intenational cooperative arrangements. Finally, Binhadi drew the attention of the seminar to the close relationship between effective bank supervision on the one hand and the conduct of sound macroeconomic policies on the other He noted that the tighter monetary policies followed in 1991 had worsened the condition of some banks and had also increased the burden of bank supervision. This outcome led to some criticisms directed at bank supervision. The enforcement of sound prudential arrangements would sometimes be construed as a restriction on the growth of credit at rates that were otherwise possible- To the extent possible, the supervisors had to ignore such criticisms, recognizing that over-rapid credit expansion in the short term often results in a deterioration in the quality of bank assets and problems for the banking system in the longer term. Concusions In the fina session, Alan Roe presented a summary of some of the main issues and contrasts that emerged during the four days of the seminar. He noted that the discussions indicated considerable similarities in the motivations behind the financial reform efforts of most of the counties represented. From the beginnig of the 1980s, if not before, the majority of the countries had recognized two basic realities: (a) sustained rapid growth in the future would imply a substantial dependence on exports and so a considerably enhanced need for flexibility and diversity in productive sectors; and (b) large inflows of inexpensive external fiance could no longer be relied upon to supplement domestic savings. Significantly, such points were stressed equally forcefully by representatives from some of the larger counries, such as Mexico and Indonesia, as well as by those from smaller economies. But in nearly all cases this line of reasoning had resulted in financial sector reforms Seminar Proceedings: A Swnmary Report 55 being embedded firmly within the broader programs of productive sector adjustment and restructuring and within the context of a significant general liberalization of commodity, labor, and other markets. The implications of this diagnosis for the types of financial reform undertaken also indicated considerable similarities across the countries represented. First, most countries accepted the fact that the preferential financial treatment of large public sectors had resulted in the past in overly rigid productive systems inadequately equipped for the speed of responses necessary for successful exporting. Hence, most countries represented at the seminar had taken steps to limit the extent of such preferences. Second, because domestic savings was going to assume greater relative importance, it was widely agreed that reforms had to ensure the more efficient mobilization and use of saviugs. The irony, of course, from the vantage point of countries such as Mexico, Chile, and Malaysia that have been most successful in improving the efficiency with which saving is mobilized and allocated, is that these countries now face capital inflows that are proving to be embarrassingly high. Third, the long-established practice of cheapening credit to a few favored or priority sectors was widely regarded as incompatible with the main objectives of reform because of the mcreased financial costs this normally imposed on the non-preferred sectors and the substantial rigidities and inefficiencies in production it encouraged. Though generally accepted, it is noteworthy that a few countries represented at the seminar, such as India, were still ambivalent about the merits of this particular point. Next, Roe drew attention to some small differences of opinion regarding the necessary preconditions for successful financial reform. The strongly articulated point of view from some participants, notably those from Mexico and Thailand, had been that a serious effort to bring fiscal deficits under control was one of the vital preconditions for liberalizing domestic financial markets. There were two man reasons for this. First, large deficits nornally result in an undesirable large preemption of available financial savings for the use of governments. Second, large deficits normally result more or less directly in high rates of inflation, which is iimical to the desired mprovements in productive sector performance. The operational difficulties in actually applying this precondition, however, had been noted, especially for the case of Chile where the fiscal accounts, narrowly defined, were roughly in balance when the unsuccessful reforms of the mid-1970s began. In that case, it was the quasi-fiscal deficit of the government, associated partly with the contingent liabilities the government was accumulating, that were one of the main factors responsible for the ultimate collapse of the reform program. Both India and, to a greater extent, Paldstan had embarked on their reforms at a tme when fiscal deficits were still high. This, however, did not necessarily represent a counter-example to the general proposition. Indeed, the representative from Pakistan had spoken explicitly of the considerable difficulties caused by interest rate reforms in his country because of the greatly enlarged interest burden this had imposed on the fiscal accounts. 56 AAn R. Roe In Mexico, by contrast, the greatly reduced interest charges in the fiscal accounts had been one of the beneficial results of the efforts made in that country to address the debt overhang and make deep cuts in the fiscal deficit. Turning to the main components of a reform program, the seminar had provided considerable similarities but also some important divergencies among countries. The similarities uniting the experiences of most of the countries included moves to liberalize access to and control of foreign exchange, shifts from quantitative to indirect methods of monetary control, the removal of controls on most interest rates, efforts to extend and diversify the range of fmancial institutions and instruments, and steps tD strengthen arrangements for bank regulation and supervision. While some countries had also sought to encourage greater competition in banling, mainly by means of freer entry to the sectr, others had regarded this step as unnecessary. Thailand, for example, explicitly sought to increase competition within the financial sector by encouraging the increased branching of existing banks and by encouraging a larger and more active securities market. Similarly, in some counties, such as Mexico and Chile, the pnvatization of banks had been a key element of overall reformn In others, such as India, radical changes in the ownership of banks had been either rejected or delayed. Roe noted that a central and recurring theme of the seminar had been the role of capital account liberalization and the associated difficulties of managing large inflows where they were occurring. The merits of such Liberalizations were well understood and included factors such as improved access to capital fumds, greater availability of the managerial and technological skills often associated with new foreign direct invsmnent, and enhanced financial sector competition associated with the entry of foreignL banks and other foreign financial instmtions. These advantages, however, accrued only in those cases where capital account liberalization led to increased inflows of funds. In countries such as India where there were still serious distortions in domestic financial mnarkets, the authorities were justified in their concern that liberalization might lead to large outflows. Roe felt that countries such as these ought to be able to draw lessons from several other countries at the seminar, such as Mexico and Malaysi, about the steps required before hberalization could be contemplated seriously. Other risks of full capital account liberalization bad also been articulated during the course of the seminar. Although few governments were now giving explicit foreign exchange guarantees to those borrowing in foreign currencies, Professor Dooley's presentation had indicated that some of the large inflows currently being experienced could be explained by various implicit guarantees available to lenders because of current govermnent policies and especially policies toward exchange rates. Together with the strong possibility that interest rates in the major industrial countries would sooner or later become much more attractive, this situation indicated a significant risk of a large reversal of inflows at some future date. For example, of the tbree hypotheses put forward to explain Mexico's recent success in attracting foreign finds, two would be likely to lead naturally to a subsequent reversal of such flows. In addition, the account of the Chilean financial crisis in the early 1980s had shown the Seminar Proceedings: A Swnmary Report 57 manner in which such a reversal might come about and what its consequences could be. Seminar participants, however, were divided as to the relevance of such experiences for the current situation. Some argued that the safeguards that had been put in place in terms of improved prudential management of bank risks, limits on the overses borrowing of banks, the avoidance of explicit government guarantees on overseas borrowings, and the greater relative importance of private stxtor borrowing should all contribute to the greater resilience of the economies concerned to unfavorable external developments. Others doubted this. Nor was there any obvious unanimity about the manner in which fiscal, exchange rate, and monetary policies might best be arranged to try to limit the dangers that Dooley had highlighted. All the countries currently benefiting from large inflows, however, were agreed about the substantial economic management problems these flows created. From the viewpoint of central banks, it was noted by almost everyone that the expansionary monetary effects of the inflows had to be avoided, either because of commitments to the IMF or because of national anti-inflationary policies. Additionally, the negative impact on the profitability of the central bank associated with the high interest costs of the sterilization operation was seen as a major difficulty in many countries. Opinion, however, was divided about the extent to which the appreciating exchange rates in such countries were inimical to the export diversification, which was one of the central objectives of overall economic reform. Some countries, such as Indonesia and Mexico, that had expenenced very large inflows and upward pressures on exchange rates had nonetheless been able to achieve extremely impressive shifts of production structues in the direction of manufactures for export. Many participants felt that upward pressures on exchange rates would be too large to be compensated by the productivity gains that were feasibly atainable in the countries concerned. This was partly because the exchange rate pressures come in the short term, whereas the productivity gains necessarily need lengthy time periods and large restructuring investments before they can be achieved. Overall, Roe commented, it was not entirely clear from the seminar whether capital account opening was aLways and reliably advantageous. The safeguards and restrictions, lie those now employed in Mexico and Chile, were yet to be fully tested. It was also difficult to articulate the full set of preconditions for hlberalization that countries still under monetary restrictions, such as India, needed to achieve before moving forward. On the subject of securities market development, the Thai paper had Mustrated the potential role of such developments for the diversification and improved competitiveness of the financial sector as a whole. It also provided an interesting illustraion of how the pressures on banks to build their capitalization had also served to dilute the power and influence of the large traditional family owners of the banks. In the case of Mexico in particular, emphasis was placed on the major role that an extensive privatization program can play in helping to develop the securities markets 58 Alan R. Roe and diversify the range of available financial instruments. In the course of several presentations, emphasis was given to the safeguards required to ensure that banldng operations in securities markets did not threaten the safety of bank payments and other operations. In Chile this is done by requiring banks to run their securities business through separate companies with segregated capital. More generally, it was observed that bank regulation and supervision needs to deal with institutions on a consolidated basis. This is because it is virtually impossible to prevent problems in one line of business from spreading contagiously to damage other lines, even though they may be organized in apparently separated companies. Finally, Roe noted the very considerable attention that the seminar had devoted to the topic of bank supervision and regulation and to the considerable advances in practice in this area that the seminar discussion had revealed. The analysis of the Chilean crisis of the early 1980s had drawn explicit attention to the need for financial reform to be accompanied by- tighter and better prudential regulation and had clearly exemplified the folly of the altemative viewpoint that liberalization should incorporate the removal of such regulation. Most countries represented at the seminar had taken this lesson fully on board, although the discussion had also revealed at least one case where contrary advice had.continued to be offered. There was also some danger that the current enthusiasm for tighter supervision arrangements would lead to overinflated expectations about the extent to which supervision alone could alleviate the dangers of fnancial sector instability. In reality, supervision also needed continued sound macroeconomic policies. Seminar participants emphasized several important aspects of good practie in this area. The first is that capital adequacy is the main foundation of prudential regulation, and this must be interpreted relative to the riskiness of bank asset portfolios. Hence, approaches such as those in Mexico and Chile where increased capital is required to match higher levels of portfolio risk are increasingly the norm. Second, integrated approaches to regulation and supervision are increasingly advocated implying standardized reserve and capital requirements across most financial initutions and the gradual downgrading of the distinction between banks and nonbanks. This also implies a consolidated approach to supervision, including the extension of the supervisors' influence to bank branches operating offshore, a practice that has already been refined in the practice of the Bank of Japan. Third, regular and detailed on-site inspection is increasingly regarded as vital to the early detection and correction of problems in banks. This point, in turn, goes back to the increased emphasis on the need for quality in bank portfolios and the impossibility of assessing this on the basis merely of statistical reurns from ffie banks and the off-site consideration of these. It is also linked in to the point stressed by Bank Indonesia about the importance of banks and supervisors worldng collaboratively to both idertify and then find solutions for banldng problems as they emerge. Some small differences of opinion were evident in relation to the matter of the disclosure of the findings of the supervisors. The Chilean example, where status Semia Proceedings: A Summey R4oon 59 reports on the banks are published regularly in the press, seems to be quite exceptional. Some participants felt strongly that such a bigh level of public disclosure was inimical to the preservation of the basis of confidence that must exist between banks and their regulators. It was generally agreed that effective regulation depends increasingly on a high level of information disclosure flowing from the banks to the supervisors if not always to the public at large. Roe concluded by considering briefly whether the seminar had revealed any obvious differences, as between Asian and Latin American practices and experiences in the area of financial sector reform. On this occasion it seemed reasonable to assert that the similarities in country experiences across the two continents had stood out more clearly than the differences. Thus, for example, there were very considerable similarities in the arrangements, the successes, and the problems experienced by Chile and Mexico on the one continent, as compared to Indonesia and Malaysia on the other. Some of the differences within coinents, however, were large. For example, the problems that India still encounters with bank distress and restictions on capital flows stand in marked contast to the more mature reforms in Thailand, Indonesia, and Malaysia. At the end of the Santiago seminar it had been suggested that the generally higher savings rates and stronger anti-inflation traditions of the Far Easten economies gave them somewhat better starting conditions for improved resource allocation based on more iiberl financial systems and, more iiportan, provided them with a somewhat better cushion for accommodating shocks and policy errors. This proposition had again been suggested in this seminar, especialy from the example of the very limited damage that the massive collapse of Japanese assets prices seems to bave inflicted on the stability of that country's financial sector. More generally, the high saving economies of the Far East still seemed able to combine broad-based reforms with continuing high levels of discretionary interventions in various aspects of financial sector behavior and to do so without doing obvious damage to the high quality of overall economic performance. This remained a puzzle that neither of the two seminars had been able to resolve. 3 FINANCLAL REFORMS, OPENING UP, AND STABILITY: POLICY ISSUES AND SEQUENCING CONSIDERATIONS Carlos Massad This paper was especially commissioned for this pubication after the completion of the Seminar Series to provide an overview from the Latn American perspectve. The author igratefid to Gamher Heldfor his valuble commns, and to Alfredo 7iraferri and Hector Vera for their statistical assistance. Editor There is ample literature on opening up policies and stability in Latin American countries. The literature covers a discussion of the nature and origin of both domestic and external shocks, real and monetary, and of the propagation of their effects in the economy. It also includes a critical evaluation of policy instruments and opening up sequence: effects of alternative policies on the rate of growth of output and employment, balance of payments, and investment. t The literature is based on a variety of experiences, ranging from the first-trade-then-capital strategy of Chile to the first-capital-tben-trade strategy of Argentina and Uruguay. However, there is still no agreement among economists on the natre of an "optimal" opening up strategy and policy or even on whether there is such an optimal approach. The differences of opinion are based not only on who 'pays the bill," that is, how the short-term costs and medium:-and long-term benefits of opening up are distnbuted. They are also based on differences implicit or explicit, regarding the analytical models used, the values assumed for relevant parameters, and the initial conditions before the opening up program is implemented. This paper examines some of the unresolved issues. The first section on over- sized capital movements explores the adjustment mechanism to oversized capital inflows under fixed and non-fixed exchange rate regimes, and the efficiency of monetary and fiscal policy in miimizing negative efflects. This section concludes that 1. See Bustelo (1987). Bouzas (1993), Corbo (1988), Dornbusch and Fisher (1993); Edwards (1988), Massad and Zabler (1986), Meller (1993), Rosales (1990), Solimano (1987), Velasco (1987). 61 62 Carlos Massad under a wide set of conditions monetary policy is substantially powerless, while fiscal policy is not flexible enough to substitute for it. The second section examines interest rates and saving in the context of an open economy and the role of monetary and fiscal policy to handle interest rate disturbances in the opening up process. A distinction between tradable and nontradable securities is drawn, and the consequence of such distinction on the behavior of interest rates in an opening economy are studied. The finding is that the strength of the domestic financial system is more important than the sequence of the reforms in trade and in the capital market to explain interest rates movements. The conclusion is that monetary policy is not efficient in a number of circumstances, while fiscal policy is not equipped for the task. The third section takes up the question of the limitations of fiscal poLicy at present, and suggests a particular form of fiscal policy that could contribute better to short-and medium-term stability. Oversized Capital Inflows: How to Cope? In the first opening up experience in recent Chilean economic history, 1975- 1981, net capital inflows reached peaks of over 10 percent of GDP. During the second erience, 1986 to the present, the peak reached almost 5 percent (table 3.1). A simil case of oversized capital inflows has registered in Argentina (Fanelli and Machinea 1993) and Mexico.2 As pointed out by Dooley, the nature of these movements has been different. During the first period, the inflow was essentially in te form of lending by foreign private banks to both goverments and the domestic private sector; during the second period, capital infi- has taken more the form of direct investment and acquisition of domestic currency securities. From the point of view of the macroeconomic impact of an eventual reversal of the flows, this difference is extremely important. In the case of foreign bank creditors lending to governments or to domestic banks, they could assume, rightly or wrongly, that there were explicit or implicit government guarantees on their loas. When the assumption proved to be wrong, as in the case of Chile in 1982, creditor banks had power enough to convince the authorities to extend an ex-post guarantee on Chilean private banks borrowigs, despite the fact that the Chilean authorities had insisted starting as early as the late 1970s, that no such guarantee existed and that it would not be granted. Mexico, in fact, extended such an ex-post guarantee through the nationalizaton of the banking system in September, 1982. In both cases, the pending danger of a complete collapse of the financial system prompted the authorities to provide the guarantees demanded. 2. See Michael Dooley, "Globalization, Speculative Bubbles, and Central Baning," paper 5 in this volume. Financial Reforms, Opening Up, and Stabilily: Policy Issues and Sequencing Considerations 63 Post-debt crisis financial reforms and the character of capital inflows after 1985 would make it very difficult for such guarantees to be demanded and even more difficult for them to be granted. There do not seem to be practical policies that could put expos, guarantees into effect in the case of direct investment. It takes time to sell real assets, and there is no way for governments to guarantee their prices short of nationalization, an approach completely out of fashion. Furffiermore, there does not seem to be an incentive for govermnents to do so. The selling of assets, including securities and shares of stock by foreign investors, would depress asset prices without endangering the stability of the financial or other markets. The reduction of asset prices would impose capital losses on the investor and iinit subsequent demands for foreign exchange, and the increase in the rate of return on assets as their price is pushed down would attract other, perhaps domestic, investors. Furthermore., under conditions of exchange rate flexibility, there would be a weakening of the domestic currency, thus increasing the capital losses faced by the withdrawing investor. In fact, the spread-over effects of an evenhtul withdrawal of foreign investments would -cushion the impact on any one of the several markets such as securities, real assets, and foreign exchange markets. Naturally, the withdrawal of foreign investors could create other problems related to expectations and confidence, but such problems would be in the economy as before the withdrawal, as the later would most likely be a consequence of the lack of confidence to start with. My conclusion is that the macroeconomic problems caused by an eventual paralysation and reversal of the tyve of foreign investment prevailing today would be far less damaging than those registered in the decade of the 1980s.3 However, the short term "digestion" problems caused by the inflow of capital in lirge amounts relative to GDP, investment, and exports, do not seem to depend closely on the nature of the inflows but rather on their relative size to the economy. Capital Inflows under Fixed Exchange Rates The process through which autonomous capital inflows are absorbed in an open economy crucially depend on the foreign exchange regime. Under fixed nominal exchange rates, a nonsterilized inflow will produce an increase in the money supply, asset pnces, perceived wealth and expenditures, generating excess demand in the markets for tradables and nontradables. Effects on assets prices and perceived wealth 3. This conclusion is to be qualified on dLe basis of the rapidly growing placement of bonds by official instilutions in international markets and the strong demand for securities expressed in domestic currency by foreign investors like the Mexican cetes. However, both cases are specific to a very smiall number of comntries, all of which have managed to strengthen their external sector and their fiscal accounts. 64 Carlos Massad will be larger, the larger the proportion of the capital inflows devoted to buying shares of stock and existing firms. Since prices of tradables are basically determined by " ,iominal exchange rate and foreign prices, they will not show substantial changes, and the excess demand will be reflected in an excess of imports over exports. In the market for nontradables, prices will increase, thus strengthening the domestic currency's exchange rate in real terms, which may lead to a deficit in the trade account. As long as there is no sterilization, this process will continue until the surplus in the capital account is matched by an identical deficit in the current account. At that point, foreign exchange reserves and the money supply will stop growing. The economy is adjusted to the increased capital inflow, until new changes in the flows start the process again. While adjustment is going on, prices of nontradables are increasing, which is reflected in domestic price indices. The economy is overheated. One readily available evidence that this process night be in operation is the fact that wholesale price indices with heavier content of tradables move up at a slower rate than consumer price indices, where nontradables have higher weight (table 3.1). If there is widespread indexation, increases in prices of nontradables are rapidly transmitted to the rest of the economy, accelerating the loss of competitiveness and generating some "overshooting" in the real revaluation of the domestic currency. If exports are essentially commodities with little or no demand in the domestic market and with relatively low short-nm elasticities of supply, the burden of the changes in the current accounts will be born mostly by imports. This, in turn, means that changes in income rather than in relative prices will bear the brunt of the adjustment Income growth will be associated with currency appreciation in real tcrms during the period of inflows, and fall in income will be associated with a real depreciation of the currency when the inflow comes to an end. The analysis above implies that, under these circumstances, revaluation or devaluation are not "caused' by changes in income, or vice versa, rather by autonomous capital movements reflecting structural changes in the financial and goods and services markets. Attempts to sterilize the capital inflow under the assumed conditions of fixed nominal exchange rates and open capital movements will be unsuccessful. Under such conditions, the supply of money is determined by its demand, and the authorities will not be able to affect domestic interest rates significantly to induce some effect on expenditure levels once the opening up process is fully established and credible. Interest rate differentials between the two markets may remain for a long time, as the experience of many countries show. However, the differential will eventually tend toward a level that could be explained mostly by higher operating costs and higher risk. As the oversized capital inflow comes to an end, so does the increased level of expenditures, and the deficit in the current account recedes accordingly. The process is reversed. Financial RJfornms, Opening Up. and Stability: Policy Issues and Sequencng Considerations 65 Capital Inflows under Floating Exchange Rates Floating rates introduce changes in the process described above. Assuming sterilization of exchange rate affects the govermnent budget, capital inflows would be reflected in a real revaluation mainly through a drop in the nominal exchange rate and a consequent drop in the nominal price of tradable goods rather than in an increase in the nominal price of nontradables. The initial inflationary effects would then be smaller, or nonexistent, and the overheating will be due to the increasing disequilibrium in the current account. Of course, there will be an excess demand for nontradables but the change in their relative price would come about mainly through the drop in nominal prices of tradables. Inflationary effects are mininiized. If the government accounts have a current surplus in foreign exchange, as in Chile or Venezuela, the effects of changes in the exchange rate on the economy will differ from those in countries with a deficit, like Argantina and Uruguay. The effects will depend on how governments face the new situation. These differential effects probably contribute to explain why Chile failed in its effort to fix the nominal exchange rate in the early 1980s, while Argentina succeeded in the early 1990s. Chile and Venezuela have continued applying some form of flexibility in their foreign exchange markets. In both fixed and floating rate cases, the capital inflow is absorbed in the economy through changes in income and relative prices. If capital inflows were somewhat erratic, or if they were large relative to the size of the economy, transitory changes in relative prices of tradables and nontradables may introduce undesired noise in the market. This is particularly so as the capital inflows are positively associated with opening up, at a time when most countries are vitally interested in the expansion of exports. Suppressing capital inflows would convey information contrary to the long- run objectives of the authorities and, under conditions of full opening of the current account, even limiting such inflows may prove impossible. Furthermore, direct investment, which is an important form of capital inflows in the present phase of the world economy, will increase the investment/GDP ratio, which is desirable from a long-term point of view. Policy Tools: Monetary and Fiscal Policies The question arises whether there are other policy tools that could facilitate absorption of foreign investment minimizing short-term negative effects. In my view, the real question is whether it is possible to affect domestic spending in the desired direction during the process of absorbing additional investment from abroad. Of course, the need to regulate expenditures arises mainly in cases where the economy is close to full employment. If it is working substantially below full employment, 66 Carlos Massad increases in expenditure would be welcome, and foreign investment would become an undoubted blessing. The traditional policy tools to affect spending are money and interest rate policies. However, under conditions of fixed nominal exchange rate and open capital and trade accounts, these tools are substantially powerless since interest rates are determined abroad and the supply of money is determined by demand, a well-known conclusion of the monetary approach to the balance of payments. Under floating rates, monetary and interest rate policy become meaningful, but changes in the exchange rates needed to maintain a stable level of foreign exchange reserves seem to be too large to swallow. Countries, in general, are willing to accept changes in reserve levels to avoid big changes in the exchange rate. The finding by Dooley, that cffects of capital inflows in recent years have been about equally spread over reserve changes and trade account changes, can be interpreted as an evidence of 'dirtiness' in floating or of limited flexibility of exchange rates. As a consequence, monetary tools can play only a limited role in accommodating foreign capital flows. And they would do so by affecting maimly private sector spending, as governments increasingly avoid recourse to the banking syslem. A growing number of Latin American central banks are independent from government and forbidden from lending to the government. It is mostly pnvate spending that is likely to be affected by changes in asset prices and perceived wealth brought about by capital inflows. Fiscal policy is the remaining macroeconomic tool to control spending. As it is well-known, acting on both the income and the expenditure side of the budget may take a long dme. Income tax changes, apart from changes in the prices of goods or services provided by state enterprises that are not important in modem states, usually require prior action from the legislature. Expenditre changes take time to be implemented, as governments commit their budgets well ahead of the time of disbursement I shall come back to this later. Interest Rates and Saving in Open Economies We have learned from economic theory that saving is related to interest rates. Equilibrium between present and future consumption at the microeconomic level is reached when the subjective rate of time preference equals the rate at which present and future resources can be exchanged in the markeL An increase in the market rate of interest would induce consumers to expand future consumption at the cost of present consumption, and consumers' saving will increase for any given level of income and wealth. Also, if consumers are faced with an increase in their income, saving will inease unless wealth increases enough for the difference between desired and existing F7nancial Reforms, Opening Up, and Stabiliy: Poicy Issues and Sequencing Considerations 67 wealth to go down sufficiently. Under several circumstances, interest rate changes will affect wealth and income, introducing some uncertainty in the final effect on saving. Business aing But consumers are not the only savers in the economy. Firms and governments are, by far, much more important as savers or dissavers in most developing economies. This is clearly so in Latin American countries. The relationship between interest rates and saving at the firm and at the government levels is by no means clear. At the firm's level, net saving equals undistributed profits. Current theory states that, in an ideal world, the firm's dividend policy is completely independent from investment pohcy, as investment can be fianced with funds exteral to the firm (Modigliani and Miller 1958, 1963). As personal income taxes are introduced, the above conclusion is revised. It is also revised when capital gains taxes are taken into account. But interest rates do not seem to play a central role since they are supposed to affect equally all forms of financing. The capital structure of the firm is undetermined (Modigliani and Miller 1958, 1963). This general theoretical conclusion is modified as the assumption of an ideal world is relaxed, but the result seems to be that taxes, risk and the cost of bankruptcy, and not interest rates, have an influence in the determination of a firm's capital structre. As a consequence, interest rates do not seem to affect the firm's policy regarding dividends. This conclusion, however, may not hold in recently lhberalized markets. As a matter of fact, undistrnbuted profits have played a large role in financing new projects undertaken by firms in Latin America. Longer-term financing in the domestic market is limited due to inflation. Foreign financing, except for a brief period between the late 1970s and the early 1980s, has been limited to funds provided to governments, or with government guarantees, by the World Bank and the Inter-American Development Banlk Commercial bank lending has reappeared on the scene only recently. On the other band, equity fimding has not been an easily available option. Stock transactions are still small relative to any scale variable, share prices are quite volatile, and firms are not commonly organized in the form of open corporations.4 (table 3.2.) At the firm's level, the behavior regarding undistnbuted profits is probably influenced very heavily by growth opportunities. As profitable investment opportunities arise, the firm is stimulated to save in order to finance new projects. Thus, saving becomes directly related to the availability of new investment options. If 4. However, direct buying of shares of stock in domestic exchanges by foreigners is growing rapidly as information imprroves. 68 Carlos Massad this is so, then saving at the firm's level is negatively related with interest rates in the financial market. If rates go up, financial costs go up and the present value of the expected futre income stream of available new projects declines. A consequence of this is a diminished incentive to retain profits within the firm, and saving at the firn's level will drop. However, firms are owned by consumers. It is reasonable to expect that, under perfeet information conditions, a change in business savings will be matched by a change in household savings in the opposite direction, as both should be perfect substitutes. If information is not costless to the household, substitution will not be perfect, and changes in interest rates will have, if any, an effect on household plus business saving running counter to the usual theoretical expectation. On balance, the effect of a change in interest rates on the sum of business and consumer savings is difficult to predict. Government Saving As regard govermments' saving behavior relative to changes in interest rates, the exploration can follow many leads. One, which has been explored mostly in Mexico and Argentina, is the effect of interest rates on government expenditures. As rates grow, debt-related expenditures of the govermuent go up and, other things being equal, government saving is negatively affected. Another lead is of a more macroeconomic cbaracter. As domestc spending increases, the monetary authorities intervene by tightening monetary policy, and domestic interest rates go up. The public sector's stabiizng reaction ought to be in the direction of reducing its own spending, with a consequential increase in saving, in order to provide support to the monetary policy stance with minimum effect on output. This implies that there is an indirect relationship between interest rates and government saving in the "right" direction, that is, increases in interest rates lead to increases in government saving. But, these two cases lead to opposite conclusions regarding the direction of the effect of interest rates on government saving. Therefore, the effects of changes in interest rates on private savings as a whole and on government saving are difficult to predict. Available evidence supports the view that there appears to be no discernible positive relationship between interest rate and saving. More than that, if there is any evidence, it shows a negative, rather than a positive, influence of interest rates on savings (Giovanni 1983). My interpretation of the evidence up to this point, at least for firms, relies on the assumption of imperfect and relatively closed financial markets. However, this assumption is not strictly necessary. Even with open capital and financial markets, it is useful to distnguish between tradable and nontradable securities following the same type of distinction as in the markets for goods and services. FPrnowal Reforms, Opening Up, and Srabilry: Poliy Issues md Sequencig Consitfrarions 69 Interest Rates in the Opening Proeess: Tradable and Nortradable Securities Tradable and nontradable securities differ on the basis of information costs, transaction costs, and risks. They are all partly related to the cost of obtaining information on the issuing firms, which in turn is related to the size and institational form of the firm (information on open corporations is usually easily available). It is, of course, .also related to the cost of information on the issuing country. Opening capital and financial markets will influence and stabilize interest rates in the market for tradable securities, but may have little influence on the market for nontradable securities, at least during the first stages of the opening process. Large- size firm's securities will probably become tradable soon after opening, while those of medium- and small-size firms, and of households (consumer borrowing), may remain nontradable, perhaps for a long period of time. Domestic and foreign rates of interest may differ by more than the expected rate of change of the exchange rate plus "country risks." In principle, even if there is a segment of the financial market where nontadable securities are trmnsacted involving lending mainly to consumers and medium-and small-size firms, increases in rates in this market should drive financial intermediaries to find additional financing abroad in a sort of induced capital inflow, thereby defeating the purposes of the authorities. Even wit relatively close connections between the tradable and nontradable sections of the domestic financial market, experience in Latin America shows that differences between domestic and foreign rates may remain for long periods of time (table 3.3). T-he above discussion is relevant in the context of the opening up process. A reduction in trade barriers will change domestic relative prices, generating new investment opportwnites as well as economic obsolescence. Both will push up demand for credit Furthermore, the change in pohlcy and in relative prices will increase systematic risk. As a consequence, domestic interest rates will shoot up. If the capital account has been open, the rate will go up more in the nontradable sector of the market. If it is still closed, it will go up in the market as a whole. But rates will increase in both cases, without generating an increased flow of saving domestically while attracting foreign financing to the tradable financial sector. This discussion leads to the conclusion that it does not seem to be too important to decide on whether the capital market or the goods and services market are to be opened first. In both cases, interest rates relevant to a wide sector of the economy will tend to go up. And domestic saving will not respond positively, even when facmg huge increases in real interest rates (table 3.4). Under the conditions generated by strong changes in relative prices, it is most likely that real interest rates go up very sharply. Firms, borrowing in distess or in search of new oppornitdes, will face a low short-run elasticity supply of credit to the nontradable sector of the financial market. The flow of saving does not respond, and 70 Carios Massd domestic interest rates become 'outlyers," accelerating the process of bankruptcies. Negative effects on employment prevail over employment expansion generated by new investment opporunities. Unemployment grows, and the stability of the financial system is endangered. Policy Options Only three categories of major policy options are briefly discussed here, though the policy menu within each category deserves a more comprehensive treatment and is covered by other papers in this volume. Tfhe Speed of Reforms The first and most obvious policy option is to slow down the rate of change of relative prices by introducing reforms gradually. However, this option may reduce the chances of the policies being fully implemented, as it gives the opportunity for opposing interests to react. The speed at which reforms are introduced is to be determined not only by economic effects but also by considerations of political sustainability of the reform package. The proper balancing of these two considerations is a ouestion of political judgment on the basis of the relative strngth of forces pro and against reform. The stronger the pro-reform forces, the lower the speed at which reform measures can be introduced and the lower the economic and social cost to be paid. The Strengthening of the Domestic Fiacial Market In any case, in countries where both the trade and capital accounts have been relatively closed, I agree with those that believe that a first step to reform should be the s ening of the domestic financial system. The main elements for strengthening are capital and solvency reiments, and the quality of prudental regulation by the authorities. Interest rate regulations will need revision to allow for increase in rates that talke them to positive levels in real terms, but it seems unnecessary and inconvenient to allow rates to reach absurd levels. If the interest rate necessary to clear the credit market grossly exceeds sometbing like 10 to 15 percent in real terms, it will destabilize other markets, leading eventually to capital transfers that are difficult to Sutin. At very high levels, interest rates become simply a mechanism for eventual capital transfers between firms and sectors rather than an indication of rates of return in the economy. hitially, highar interest payments are simply financed, exacerbating Finanal Reforms, Opening Up, and Srabil4y: Policy Issues and Sequencing Consideraions 71 demand for credit. As a firm reaches the point of illiquidity, the financial system stops lending to it and capital transfers are expected to materialize, leading to traumatic legal and economic processes. Meanwhile, domestic interest rates do not play an efficient role in allocating resources or in increasing the flow of domestic saving. They do contribute to attract foreign capital, but they do so at the cost of an overkill. Legislation about bankruptcy procedures also need modernization, mostly in the direction of finding ways to allow the firm in distress to continue operations as a unit and of speeding up the process of liquidation. There are many cases where the recognition of a lower capital value of a firm leading to a reduction in liabilities is enough to facilitate its continued profitable operations, avoiding unnecessary additions to unemployment. In cases such as these, there is little creditors can do to protect their claims. Monetary and Fiscal Policies Monetary and fiscal policies should not press the financial markets firther. Expansive monetary policy would risk inflation, as the whole adjustment process is dominated by changes in relative prices that monetary policy by itself cannot change. An expansive policy stance of the monetary authority could lead to inflation without contributing much to unemployment reduction. On the other hand, a restrictive policy stance would only aggravate pressures on the financial markets. FLscal policy could be of great significance. By eliminating negative public saving it could positively contribute to the flow of saving in the economy, direcdy aReviating the pressure on financial markets. But it will probably need to do so by increasing government income rather than by reducing expenditres. Expenditure reduction may come at a later stage, when the economy is well on its way to recovery and positive stimulus from the expanding sectors begin to take hold. A second option is available to countries that have kept open capital markets for a long time. In such cases, it is likely that the nontradable sector of the financial market has been shrunk to be negligible from a macroeconomic point of view. If this is the case, trade reform can proceed at a faster pace as its effects on domestic interest rates will be minimized by capital inflows. Resources will be pulled out from negatively affected firms and/or sectors rather than pushed out of them by falling relative prices and unbearable interest payments. Unemployment effects will thus be minimized. The case of Argentina illustrates this option. Argentina has maintained open capital markets, either by design or de facto, for a long period of time. However, trade opening occurred only recently. Under these conditions, Argentina's open trade policy did not cause the disruptions in employment and growth of the type experienced, for example in Chile in the mid 1970s, or in Peru in the early 1990s. Similarly, real 72 Carlos Massad interest rates in Argentina during the trade opening period did not reach levels even half as high as those in Chile or Peru. Again, these results are not a consequence of opening up the capital market before opening up trade; it is rather a question of the relative iInportance of nontradable securities in the economy and of the speed at which they become tradable (table 3.5). The Exchange Rate Regime The exchange rate system of the country is quite relevant to this discussion. A credible fixed nominal exchange rate in relation to the intervention currency win help stabilize the price level, by providing a significant anchor for prices of tradable goods and services. If opening up occurs in an inflationary environment, a fixed nominal exchange rate will produce a separation between real interest rates in domestic currency and real rates measured in foreign currency, and the difference between the two may become substantial. For example, at a rate of inflation of, say, 15 percent per year, an interest rate of 20 percent in the domestic market will imply an internal real rate of roughly 5 percent, while the corresponding rate measured in dollars would be 20 percent, as the price of the dollar is not expected to change (table 3.5). The difference between these two real rates- is a strong incentive for capital inflows. Fixed nominal exchange rates in an inflationary environment imply a "subsidy' to capital inflows. The magnitude of the subsidy is equal to the difference between the domestic and the foreign rates of inflation. However, this peculiar subsidy is paid only when the inflow of capital is reversed. It particularly stimulates short-term inflows, ready to leave when the domestic currency is in danger of a devaluation. If devaluation expectations flare up, the authorities could do little to prevent a devaluation of the domestic currency. More flexible exchange rates provide a cushion to short-term capital movements, but at the cost of affecting domestic price levels and introducing additional uncertainty costs to tradables. Hence, flexibility is usually limited, either discretionally as in dirty floats or by some rule such as a band, a crawling-peg, or a tablita. These are intermediate cases that help distribute the impact of shocks among ddierent markets. Fiscal policy is the single most important domestic tool to affect the probability of a devaluation, as monetary policy is powerless in an open economy with fixed exchange rates, and of limited effect in intermediate cases. Once again, fiscal policy appears in the forefront. Funncial Reformu, Opening Up, and Stabiliy: PoIi:y Issues and Sequenwing Considerations 73 Conclusion: Need for a New Fiscal Policy? Fiscal policy directly affects expenditures, public or private, either by changing government income or spending, although changes in either are rather slow to implement. Given the limited short-run predictive capacity regarding the timing and magnitude of shocks, such slowness in implementation prevents the use of fiscal policy in many cases. In other cases, it generates undesired effects of aggravating rather than reducing the impact of shocks in the economy. This type of consideration is relevant when the government budget is not the source itself of major disequilibria in the economy. The general recommendation. to governments that the budget ought to be near balance at all times is perhaps one of the most important rules learned during the 1980s. But this is not enough. The more open the economy of a relatively small country, and the narrower the limits within which the nominal exchange rate is allowed to move, the lower the efficiency of monetary policy. Or, put in other words, the more open the economy, the lower the capacity of the monetary authorities to control the supply of money and domestic interest rates, and their powers to control private spending in case of need. As the economy is more open, the policy burden falls increasingly upon fiscal policy. Yet, tax changes in most countries can only be introduced by law, and the legislative process usually takes months. Tax changes, therefore, are ruled out as short-term policy tools. On the other hand, government expenditures do not respond easily to decisions to change them. Furthermore, a theoretical argument has been put forward to point out that under some conditions, changes in taxes or in government spending may not influence private or even total spending. According to the Ricardian equivalence principle, the private sector takes into account future tax changes that are expected to occur as a consequence of present changes and its own spending stream is not affected. However, this theoretical argument assumes no liquidity constraints, and households are assumed to take into account taxes on future generations when deciding about their own spending. Empirical evidence does not seem to be in accord with the theoretical argument (Bernheim 1987). The practical considerations mentioned above must be taken seriously in the design of policy. On the expenditure side, while governments may manage to increase their flexibility to alter spending, there are practical limits to it. Flexibility in spending requires not only an efficient decisionmaking process, which is a relatively scarce comnodity the world over, but also demands efficient unplementation and a generalized system of short-term contracts or provisions for change on short notice within prevailing contracts. Such provisions would certaily entail additional costs both in economic and political terms. Hence, they are generally ruled out. On balance, it seems that short-term changes in government expenditures can not be used efficiently ac a policy tool. Such changes, due to delays in implementation 74 Carlos Masse and predictive errors, may destabilize the economy. After the desired relative size of the state has been achieved, the best one can hope for on the expenditure side, is to aim to a roughly constant rate of growth of government spending. This rate should be similar to or slightly smaller than the long-term growth rate of the economy. In this way, the relative size of the state achieved would be preserved and the risk of errors on the side of an excessive increase of its size would be minimized. This argument is logically similar to that of Milton Friedman in his recommendation of a constant rate of change in the money supply through time implying rules and not discretion in the conduct of monetary policy (Friedman 1961). Obviously, it is impossible to apply Friedman's recommendation in a small, open economy that regulates its exchange rates. In that case, the supply of money would be determined by demand. But it is not impossible to apply the same logic to government spending. The story on the income side of the public budget can be quite different. Government income moves with the economy, as tax collections are sensitive to the level of economic activity. That sensitivity could be increased through appropriate changes in the tax structure that could be introduced once and for all. Furthermore, the authorities could be empowered by law to make transitory changes in some tax rates to support the built-in stabilizing characteristics of the government budget. The tax rates, which are first line candidates for transitory adjustnent, are pay- as-you-go type taxes of general application. Perhaps the most widely applied tax of that type is the value added tax, and since VAT is collected on each sale, the effects of changes in the tax rate will be visible as soon as the change is introduced. Variable tax rates could also apply to short-term capital inflows. This is already done in some countries through the introduction of obligatory noninterest bearing deposits of variable maturities and in different proportions of the inflow. With government expenditures growing at a constant rate while government income moves with the cycle, the fiscal budget could become a very important policy tool, to take up more and more of the burden that monetary policy is increasingly unable to bear. This is the road that Chile has just begun to follow. Congress has approved legislation allowing the executive power to introduce limited, transitory changes in the rate of the VAT, while government spending is expected to grow in consonance with the potential rate of growth of the economy. Financial Reforms, Opening Up, and SMabiltly: Policy ix.. zand Seqencing Cons Werations 75 Table 3.1. Chile and Mexico: Annual Average Percentage Change in Consumer and Wholesale Price Indices and Net Ccapital Inflow as a percentage of GDP, 1975-91 Chile Meico Year CPI VWP! Na CPI WPI Na 1975 374.70 4B1.90 (1.03) 15.20 10.70 2.85 1976 211.80 221.10 (1.27) 15.80 22.30 0.80 1977 91.90 86.10 2.78 29.00 41.20 0.06 1978 40.10 43.00 8.86 17.50 15.80 0.79 1979 33.40 49.40 7.65 18.20 18.30 1.42 1980 35.10 39.60 8.75 26.40 24.50 2.68 1981 19.70 9.10 10.66 27.90 24.50 3.01 1982 9.90 7.20 (3.65) 58.90 56.10 (5.77) 1983 27.30 45.S0 (5.65) 101.80 107.40 (8.84) 1984 19.90 24.30 1.25 65.50 70.30 (7.07) 1985 30.70 43-40 (4.46) 57-70 53.60 (6.896) 1986 19.50 19.80 (5.39) 86.20 88.40 (4.73) 1987 19.90 19.20 (3.67) 131.80 135.60 (3.80) L988 14.70 5.90 (3.67) 114.20 107.80 (6.60) 1989 17.00 15.10 (1.70) 20.00 16.10 (1.80) 1990 26.00 21.80 4.55 26.70 23.30 1.15 1991 21.80 21.70 (1.29) 22.70 20.50 5.02 CPI Consumer price index WPI Wholesale price index. NCI Net Capital inflow. Source: ECLAC. 76 Carlos Masd Table 3.2. Argenrina, Chile, and Mexico: Nominal Value of Shares Sold in Stock Exchanges as a Percentage of GDP, 1988 and 1922 Year Argiendna Cil Meico 1988 0.70 5.10 3.30 1992 8.30 4.80 13.30 Sore: Santiago Stock Exchange. Table 33. Arzcndina, Chile, andMexico: DomesdclInterestRates (U.S.dollars) 1979-93 Year Argeia (a) C7ik la) Mexico (b) 1979 - 37.78 15.02 1980 - 40.52 21.40 191 - 52.02 2276 1982 - 2554 (36.69) 1983 - (7.78) (25.30) 1984 518.39 10.55 6.87 1985 454.11 (13.44) 6.60 1986 (49.04) 5.38 - 1987 6,99Z.23 16.76 (9.85) 1988 40.56 16.95 47.43 1939 1,087.53 12.83 25.78 1990 16.436.85 31.56 20.43 1991 6.58 15.80 14.28 1992 18.58 22.23 14.03 1993 (4.30) 12.40 6.80 a. Lending rate. b. Based on lreasry bills. Source: Intemaional Monetary Fund, IFS 1993. Firancial Reforms, Opening Up, and Sability: Policy Issues and Sequencing Considerations 77 Table 3.4. Argentina, Chile, and Mexico: Domestic Real Interest Rates (DRI), and Saving as a Percentage of GDP, 1975-92 Argentina Chile Mexico Year DR (a) S ) DRI (a) S (b) DRI (a) S (0) 1975 - - 20.8 11.1 - - 1976 - 34.9 17.1 - - 1977 - - 37.1 12.4 - - 1978 - - 32.9 14.5 (5.9) 23.0 1979 - - 21.5 15.0 (2.7 24.7 1980 - - 8.9 16.9 (3.1) 24.9 1981 - - 27.0 12.4 2.2 24.9 1982 - - 49.1 9.4 (8.3) 27.9 1983 - - 12.2 12.5 (21.2) 30.3 1984 33.3 45.1 ij.- 12.5 (9.8) 27.7 1985 63.2 39.8 8.1 16.5 3.5 26.3 1986 9.4 36.6 5.7 18.4 - 22.4 1987 48-3 34.5 10.8 21.0 (12.4) 25.4 1988 8.8 41.7 7.6 24.2 11.1 22.0 1989 2,784.1 36.8 11.9 24.0 21.1 21.2 1990 9,432.4 19.7 17.1 23.1 3.8 20.7 1991 13.9 16.2 8.4 23.7 0.6 19.3 1992 8.0 15.2 10.1 23.4 3.5 17.7 DRI Domestic real interest rates. S Savings as a percentage of GDP. a. For Argentina and Chile, the DRY is based on lending rates. For Mexico, DR[ is based on hrsury bills. b. Calculated as the difference between GDP and total consumption, divided by GDP (nominal). Source: Intenational Monetaiy Fund, IFS 1993. 78 Carlos Massad Table 3.5. Argentina, Chile, Mexico, and Peru: Domestic Real Interest Rates, Selected Years Chile (a) Year Pesos USS 1976 34.90 58.24 1977 37.13 59.56 1978 32.86 26.59 1979 21.52 37.78 1980 8.91 40.52 1981 27.00 52.02 Airgentina (a) AMfexco () Pent (a) Year Pesos UsS Pesos uss Peso USS 1988 8.80 4056 ll OS 47.43 - - 1989 2.784.10 1,087.53 21.10 25.78 (125.00) 131.55 1990 9,432.38 16,436.85 3.80 20.43 (13.05) 29250 1991 13.90 6.58 0.63 14.28 248.20 504.90 1992 7.98 18-58 3.50 14.03 55.83 67.93 1993 (c) (4.30) 4170 6.80 16.25 40.30 47.93 a. Lelding rates in 1993. b. Data based on the average of dhe fist and seeond quarters of 1993. c. The interest rate is based on treasury bils. Financiol Reforms, Opening Up, and Slability: Poliy fssmes wnd Sequcig Consideratrons 79 References Bernheim, B. Douglas. 1987. "Ricardian Equivalence, An Evaluation of Theory and Evidence." NBER Macroeconomics Annual, Vol.2, New York: National Bureau of Economic Research. Bouzas,o R. 1993. "Mas Alla de la Estabilizaci6n y la Reforma? Un Ensayo Sobre la Economia Argentina a Comienzos de los 90." Desarrollo Econ6mico, 33 (129), (Abril, Mayo, Junio):3-28. Bustelo, E. 1987. Politica de Ajuste y Grupos mds Vulnerables en Aminca Latina: Hacia un Erfoque Alternativo." UNICEF, Bogoti, Fondo de Cultra Econ6mica. Corbo, V. 1988. 'Inflaci6n de una Econornia Abierta: El Caso de Chile." In F. Morande and K. Schimdt-Hebel, eds., Del Auge a la Crisis en 1982. Santiago: ILADES Dornbusch R., and S. Fischer 1993. -Moderate Inflation." The World Bank Economic Review 7(Janu.. y): 1-44 Edwards, S. 1988. "La Crisis de la Deuda Externa y las Politicas de Ajuste Estructual en Arn&ica Latina.' Colecci6n Esnrdios, CEEPLAN, No. 23, Santiago. Fanelli, J. M., and J. L. Machinea. 1993. "Capital Movemens in Argenina", ECLAC-IDRC Workshop II, New Pnvate Flows ito Latin Amenca, Saniago, (December 1993). Friedman, M. 1961. "The Lag in the Effect of Monetary Policy." Journal of Political Economy (October). Giovanmini, A. 1983."Savings and Interest Rates in LDC's." World Development (July, 1993). Massad, C., and R. Zahler. 1986. "El Progreso de Ajuste de los Afios Ochenta: La Necesidad de un Enfoque Global en America Latina." Sistema Monetario Internacional y Financiamiento Extemno. CEPAL. Santiago. 80 Carivs MAssad Meler, P. 1993. "Ajustes y Reforma Econ6micas en America Latina: Problemas y Experiencias Recientes." Pensamiento iberoamericano, (Julio 1992 - Junio 1993):15-58. Modigliani, F., and M. H. Miller. 1958. "The Cost of Capital, Corporation Finance and the Theory of Investment." American Economic Review (June):261-297. . 1963.-Corporate Income Taxes and the Cost of Capital-, American Economic Reiew (June 1963):433-443. Rosales, 0. 1990. 'El Debate Sobre Ajuste Estructural en America Laina". ILPES- CEPAL, Documento EIN-62, Enero. Solimano, A. 1987. 'Inflacion y los Costos de Estabilizar: Aspectos Conceptuales, Casos Hist6ricos y Experiencias Recientes."EI Trinestre Economico 56 (4), No. 224 Octubre-Diciembre 1989:765-797. Velasco, A. 1987. "Politcas de Estabilizaci6n y Teoria de Juego." Colecci6n Eoudis, CIEPLAN, No. 21. 4 FINANCLIL DEREGULATION AND BANK SUPERVISION: TlHE CASE OF INDONESIA Binhadi, Manging Director, Bank Indonesia Indonesia initiated financial deregulation beginning in the late 1960s. At that time, deregulation covered the foreign exchange market, investment and banking institutions. The implementation of the deregulaton policy, however, was affected by unfavorable economic conditions in the early 1960s, and was delayed by te oil booms in 1970 and the early 1980s. To accommodate such conditions, deregulation in monetary policies was not implemented until 1983, while deregulation of banking institutions was terminated in the early 1970s, only to be reinplemened again begnning in 1988. Meanwhile, deregulation in foreign excbange and investment contnued. In 1982 the Indonesian economy was affected by the world recession of the early 1980s. The situation worsened when declining oil prices led to sicant decreases in net oil exports. In addition, the growth rate of deposit mobilizaton by the banking sector continued to decline since liquidity credits were backed up by fthe oil boom in the 1970s. To cope with these problems, the government responded with a broad range of actions including deregulation of the finanial sector. Since then, the Indonesian goverment changed its economic policy to implement a sweeping deregulation, including deregulation of the real sector. Indonesia introduced bank supervision in the 1950s implemented by Bank Indonesia. In ilementing bank supervision, Bank Indonesia's techniques and approaches changed from time to time based on economic policies and developments within the financial and banking systems. Finanal System Deregulation Deregulation of the financial system in Indonesia was done in a sequence and was followed by other deregulaton and reductions in bureaucracy that covered a wide range of activities that affected the development of the economy. In the late 1960s the government conducted economic stabilization and rabilition programs designed to reduce inflation, provide adequate supply of basic needs, rehabilitate the economic intucture, and increase exports. In this context the government deegued 81 82 Binhadi foreign exchange control, encouraged domestic as well as foreign investments, and eased the establishment of new private and foreign banks. Because of very high inflation rates and undeveloped banking and money markets in the late 1960s, the monetary policy was implemented through a direct instrument: interest rate determination by the central bank. In the 1970s and early 1980s the oil boom greatly affected the Indonesian economy. This condition discouraged the implementation of the deregulation policy. On the otber hand, the evidence of a banldng crisis in the early 197as urged the government to issue a merger policy and to regulate the banking sector. Meanwhile, deregulation of foreign exchange and capital account continued. To overcome the economic deterioration caused by the world recession and the decline in oil prices in the early 1980s, the government furither deregulated the foreign exchange market. This was followed by deregulation of the financial, monetary, and bankng ssystem and deregulation of the capital market As part of the deregulation launched in 1990, Bank Indonesia further reduced its liquidity credit and improved the national credit system. Following the monetary and banldng deregulations the government deregulated the real sector in several stages. In 1991 Bank Indonesia improved its bank supervision system by introduicing bankdng prudential principles. To improve the legal framework, the new Banking Act, the Insurance Act, and the Pension Fund Act were enacted in March 1992. Deregulation of the Foreign Exchange AMarket The process of deregulating the foreign exchange market changed three key aspects, namely, the foreign exchange trading mechanism, foreign exchange market participants, and foreign exchange rates. The adjustment measures were selectively applied, covering either all three aspects simultaneously or only one aspect at a time. During 1967-70, several changes occurred concernig the foreign exchange trading mechansm and foreign exchange market participants, and some movements away from a multiple exchange rate system toward a single foreign exchange rate. A major change in the foreign exchange market mechanism occurred in 1967 when the government established the Foreign Exchange Bourse, and abolisbed the regulation that required official permission to use foreign exchange. For the first time since the existence of exchange controls, it was now possible to legally trade foreign currencies against the rupiah. Starting in 1970 the holding, selling, and purchasing of foreign exchange were no longer subject to restrictions. The requirement for exporters to surrender the export proceeds to a foreign exchange bank, however, was retained, although nothing prevented them from buying an even larger amount of foreign exchange at the same time. Foreign exchange banks, in turn, had to sell to Bank Indonesia the amount of foreign exchange they acqpuired. Importers and those who Financial Deregukation and Bank Supervsionit The Case of Indonesia 83 needed foreign exchange for whatever purpose had to buy foreign exchange from a foreign exchange bank. Thus, a foreign exchange bank could buy and seU foreign exchange to their customers. Foreign exchange transactions between banls and Bank Indonesia were conducted through the Foreign Exchange Bourse. During 1978-82, major changes occurred in the exchange rate system. The rupiah was devalued by 33.6 percent primarily to improve non-oil export and was pegged to a basket of the currencies of Indonesia's major trading partner in place of the U.S. dollar. The new system was expected to improve the competitiveness of Indonesian exports in the international markets, increase foreign exchange earnings, and stitmulate foreign capital investments. Beginning in 1982 exporters no longer bad to surrender foreign exchange proceeds to Bank Indonesia. They were allowed to hold export proceeds for their needs, and they could sell part or all foreign exchange proceeds to Bank Indonesia through a foreign exchange bank. The same applied to importers who wanted to buy part or all foreign exchange for their imports from Bank Indonesia. To encourage the development of the foreign exchange market, the government allowed the esablishment of brokerage firms. In 1989 the government abolished the Foreign Exchange Bourse and decided that transactions between Bank Indonesia and a foreign exchange bank must be done through dealing rooms. The government also removed the ceilings on a bank's foreign borrowings and applied the Net Foreign Exchange Open Positon regulation as a way to minimize a bank's risks. Furthermore, the procedures of liquidity and investment swaps between foreign exchange banks and Bank Indonesia were revised. These revisions of tl-- foreign exchange mechanism allowed the government to develop the cental bank mtervention pattern through the buying and seiling of foreign exchange in the foreign exchange market With ffie monetary and banking deregulation in 1988, the government reopened the opportumity to set up a foreign bank in the form of a joint-venture that could also be a foreign exchange bankl For private national banks the government had given some flexibility to operate as a foreign exchange bank provided the banks were sound. From 1989 onward, the exchange rate determined by the central bank was not the daily compulsory rate to be applied but rather only an indicative rate, which meant that the rate in the foreign exchange market could fluctuate on the basis of developments in the domestic and international markets. The central bank could exercise intervention in order to adjust this rate. With the Net Foreign Exchange Open Position regulation, which provided a squaring system for banks in te afternoon, the central bank determined the rate for squaring by simply examining the development of rates in domestic and international markets. In determining the indicative rate the central bank, however, still maintains the managed floating exchange rate system based on a basket of curencies. 84 Biniwdi The Monetary, Financial, and Baning Deregulation In 1983 the government instituted a far-reaching deregulation in the financial system. It removed the interest rate controls on state banks and lifted the system of controlling bank credits through administrated ceilings in favor of a more market oriented approach that relied on the development of indirect monetary instruments, namely, reserve requirements, open market operations, and discount facilities. To reduce the injection of money supply through liquidity credits, a significant part to those credits outstanding as of March 1983, were not renewed, and new liquidity credits were provided only for high priority sectors. For the implementation of open market operations, Bank Indonesia developed money market instruments, namely, Bank Indonesia Certificates (SBIs), and money market securities (SBPUs). SBIs were introduced in 1984 for tbree main reasons: first, as a monetary policy instrument, especially for monetary contraction; second, as a money market istrument, and trd, as an alternative apportunity for the banking system to invest temporary excess reserves. Money market securities were introduced in 1985 primarily for monetary expansion purposes. In addition to the SBIs and SBPUs, Bank Indonesia introduced discount facilities and improved the operations of the interbank market. The 1983 reforms had positive effects and enabled the economy to generate alternative sources of funds and significant increases in non-oil exports. The reforms played a crucial role in mobilizing funds, for investments, promoting non-oil exports, determining the market interest rates, affecting movement of exchange rates, influencing capital flow, contributing to the control of inflation, and improving the distribution of income. By and large, the banking system contributed to the considerable development of the private sector, enhanced the monefization and financial deepening thbroughout the country. Expansions in banling also increased competition among the banks, encouragig them to provide professional banking services and increase their efficiency. To further increase the role of the banldng system, the government introduced a deregulation package in 1988, called PACTO 27. The package was designed to mobilize finds, increase efficiency in the operation of banks and other financial institutions, effectively implement monetary policy, and develop the capital market and boost non-oil exports. In effect, these measures served as an integral part of the government's efforts on deregulation and debureaucratization. Policies on fund mobilization involved (a) opening opporunities for expanding the offices of banks and nonbank financial intmediaries (NBFIs) and establishing new banks including rural banks; (b) giving freedom to banks to fiurther develop savings schemes and to issue certificates of deposits; and (c) giving opportmities to NBFIs to act as deposit takers through the issuance of certificate of deposits. The promotion of non-oil exports requires expanded and improved banking services. For this purpose, deregulations involved (a) opening opportuies for Fnancial Dereguan andl ank Sapervisor The Case of JwJ la 85 expanding foreign exchange banks, establishing of joint banks, openin foreig bank sub-branch offices, and expanding of money changes to promote tourism; (b) improving foreign exchange swap and reswap arrangements; (c) abolishing foreign borrowing limits for banks and NBEFIs and intoducing the foreign exchang net open position; and (d) requiring joint banks and foreign banks to provide 50 percent of total credits for export purposes. To increase the efficiency of banks and NBFIs, a climate to induce sound competition was to be created. The expansion of banks and branch offices and the freedom to create new products for fund mobilization and credit expansion were the most import aspects for the inducemes of sound competition. In this conext, the government allowed the state comPanes to deposit their finds in private banks, with a maximum of 50 percent of the funds invested. Such deposits in a single private bank must be a maximum of 20 percent of the deposit by state companies. To fiuther increase the effctiveness of the monetary policy, the dergulatons also incorporated unification and standc'diztion of the reserve gre nt for banks and NBFIs, and the continuation of the conduct of open market operations wit the use of SBIs and SBPUs. For banks, the reserve r ent was reduced to 2 percent of deposits, and NBFIs were required to n the same percentage of reserve rqieents. To accelerate developments of the capital marit, deregulations includod a plan to equalize taxes on income from deposits wit that of income from seciies, and permission for banks and NBFIs to raise capital by issuing new shares hrough the capital market in addition to increase in the equity partciion of sharholders. For eql tax treatme, the principal measure was to charge a 15 percent final tax rate on interest income earned from timw deposits, savings deposits, and certficate of deposits, with the possibility of tax restitution for snull savers. However, since January 1992, the 15 percent final tax is only for deposits owned by indiiuals, whereas inter income on corporate deposits is teated as regular income. The Deregudaon of Capil Market The capital market in ndonesi was reestblished in 1976 and grw very slowly until 1988. Since the deregulaion in December 1988, however, the capital market has grown rapidly. The existence of the capital market is complementary to the banking sector to provide alternative source of funds for investment. NevertheIess, in the past, participation in the capital market did not increase as expected. Among the conditions needed to develop a strong capital market are an adequate mnmber of companies, who~ are willing to sell their shares in dte market, and supportive institutions and aive trading in the secondary market To overcome the lack of participation, the government initially introduced a policy package in 1987 that significantly modified and relaxed the requirements for the issuance of securities at the stock exchange. A parallel stock exchange was also established for companies who wished to issue their shares but could not fulfill the requirement of the stock exchange. In addition, foreign investors were- also allowed to partiipate in the parallel market in accordance with foreign investment regulations. As a follow-up, in December 1988, the government introduced another policy package for the capital market resulting in significant improvement in the activity of the capitl markt The measures in this package included the establishment of a private stock exchange, the opening of the stock exchange outside Jakarta, the expansion of marketable securities trade at the Jakarta Stock Exchange, and the development of fiane companies. The package provided various opportunities for new fnancing activities to encourage the development of finance institutions that could support the capital market To conduct their activities, foreign investors were allowed to participate in the ownership of a joint venture securities company and finance company. The National Crdit Policy Satisfactory developments in the banldng sector with respect to funds mobilization, improvement in efficiency and services to the public, interest rate movements, a favorable inflation rate, and exchange rate movement signified that the fiondation of the financial sector had become stronger, efficient, and more supportive to development efforts. hI the meantime, even though efforts were made to reduce the wuidy credits provided by Bank Indonesia, its yearly development showed an icreasing amount This was not favorable for the development of an efficient and strong financial sector. In 1990, the government took measures to improve the national credit system. Its objective was to gradually reduce the liquidity credits for various programs and activities, and at the same time to increase the role of the banking sector and general credit expansion. Furthermore, only a limited amount of liquidity credits from Bank Indonesia could be provided to support the efforts for food self-sufficiency, the development of cooperatives, and the enhancement of investments. Credits extended by banks and oth financial insfitutions, except for the above mentioned credit progas, would be based on their deposit mobilization. National banks were to allocate a minimum of 20 percent of their total credits for a guaranteed availability of funds for smal-scale industry and productive activities of cooperatives. In addition, foreig banks and joint banks were to provide at least 50 percent of their total credit for export purposes. Finacial Deregukilon and BRak Supervision: Thw Case of Indonesa 87 T7he New Banking Act, the Insurance Act, and the Pension Fund Act of 1992 Three new laws were enacted in 1992 to provide legal foundation to deregulation policies and to enhance the role of the banks. These were: the new Banking Act, the Insurance Act, and the Pension Fund Act. The new Banlding Act concerns the ownership of banks and stipulates that foreigners may purchase commercial bank shares through the stock exchange but are not allowed to become majority shareholders, while the state-owned commercial banks may issue shares through the stock exchange provided the majority of shares is still owned by the government. However, the scope of a bank's operations is principally universal banking. Banks in Indonesia are permitted to conduct all financial activities, although some activities must be done through subsidiaries. The ownership and establishment of insurance companies is treated in a similar manner as the banks. In the case of pension funds, two types of funds are available, namely employer pension funds and general pension funds. Banks are allowed to manage both types of pension funds. These new laws provide insuranc and pension funds a legal foundation for conducting operations in a deregulated manner. The development of these institutioPs will enable mobilizaton of long-term funds to support investments and capital market development. The Impact of F-mancial Deregulation Financial deregulation affects both macroeconomic variables and finacial institution development. In this section, the discussion first focuses on the impact on the banking sector, and then reviews the ways deregulation has affected macroeconomic variables. The Banking Sector There have been major changes in operational banking patterns and banking product developincnl since the deregulation in 1988. The Indonesian banldng industry experienced an impressive growth in the number of offices, the diversity of services, and the volume of business. In 1989 the number of commercial banks increased by thirty-five banks, ir. 1990 by twenty-five banks, in 1991 by twenty-one banks, and in 1992 by sixteen banks. At the end of 1992, the total number of new banks was 221 of which 7 are state-owned banks, 27 are regional development banks, 114 are private national banks, 10 are foreign bank branches, 20 are joint banks, and 13 are NBFIs. The total number of branches and sb-branches also increased substantially. During 1989-92, 3,681 new branches were established bringing the total number of 88 Biladi branches to 5,626 by the end of 1992. In addition, there are 3,195 BRI village units and 8,520 rural banks, of which 814 are newly established. This means that with a population of around 180 million, one office unit serves about 10,000 people. The mobilization of fumds also expanded rapidly during 1989 and 1990. Deposits increased by 44 percent in 1989 and by 51 percent in 1990. Credits increased by 37 percent in 1989 and by 51 percent in 1990. Because of the tight monetary policy that began in the middle of 1990, the growth of deposits in 1991 and 1992 slowed down to 17 percent, and credit expansion also slowed down to 17 percent in 1991 and 10 percent in 1992. The deregulation policy also encouraged the development of banking products and services. To mobilize funds, each bank tried to develop its own specialties to attract as many customers as possible. As a consequence, the number of savings schemes, for example, increased to about 140. The money market and the foreign exchange market have also developed in terms of both volume and product diversification. The interbank money market in rupiah increased from a daily average of Rp 180 billion in 1988 to Rp 435 billion in 1992. Short-term securities trading increased from a daily average of Rp 102 billion in 1988 to Rp 294 billion in 1992. The foreign exchange market grew substantially, from a daily average of US$2 billion in 1992. The products of the foreign exchange markets also expanded; in addition to spot transactions, they now include products, such as swap, forward, option, futures, and margin trding. The icrease of banks and bank offices and the increased vohlme of assets, ihabilities, and banking products and services affected the number of employees needed. Even though the banks' human resource development began much before the bankng institLtional deregulation, the sudden increase caused a shortage of skilled bank staff. This situation created keen competition and hijacking of personnel. The Impact on Macroeconomic Variabtes Since the early 1980s, the deregulation policies, as well as macroeconomic policies, such as the fiscal policy and the rephasing of huge government projects, have helped to improve the Indonesian economy. GDP growth, which had declined to 23 percen in 1982, increased to 4.2 percent in 1983. In the following five years the average growth of GDP reached 5.2 percent per year. On the other hand, the rate of inflation, which was 11.5 percent in 1983, declined to an average of 7.2 percent per year in the following five years. In 1988, as a result of further deregulation of banicng and relation of monetary controls, the GDP growth was around 7.4 percent during 1989 and 1990. The ensuing enormous increase in dermnd both in consumption and in investment caused an overheated economy. This led to a sharp increase in imports, a deterioration in the growth of non-oil exports, and an increase in inflation. This situation evennally Financial Deregutadon and Ran/ Supervision: The Case of Indonesia 89 forced the government to implement a tight monetary policy. To avoid an increase in foreign commercial borrowings, the government issued a prudential foreign conmmercial borrowing policy. Rapid credit expansion after financial deregulation in 1988 caused a rapid increase in money supply (Ml and M2 ) in 1989 and 1990. In 1989 both MI and M2 increased by 40 percent. In 1990, growth in MI declined to 18 percent, but M2 coninued to show a rapid increase of 44 percent. This development, on the one hand, supported the GDP growth, but on the other hand, increased the inflation rate from 5.9 percent in 1989 to about 9.5 percent during 1990 and 1991. The growth of MI and M2 then began to decline due to the implementation of a tight monetary policy in the middle of 1990. In 1991, both Ml and M2 increased by 10 percent, but in 1992, Ml growth was 9.3 percent and M2 growth 20 percent. Because of the new monetary policy, the interest rate that had increased to 25 percent in early 1991 started to decline at the end of 1991 and was 15 percent in 1992. With regard to exports, since the removal of the obligation for exporters to surrender their export proceeds and the deregulation policies in other sectors, non-oil exports began a sustained increase in 1983 and surpassed the value of oil exports by 1987. In 1989, non-oil exports were higher than imports, and given a slight increase in oil exports as well, current account deficit declined. There were some slack in non-oil export growth in 1990, but during 1991 and 1992 the growth resumed to about 21 percent. With the above development, Indonesian's share in international trade continued to increase. The ratio of exports to GDP was 18.2 percent in 1986 and increased to 22.7 percent in 1987, 24.5 percent in 1989, and 25A4 percent in 1991. The imports to GDP ratio was 15 percent in 1986 and increased to 16.5 percent in 1987, 17.3 percent in 1989, and 21.3 percent in 1991. Furthermore, the deregulation of foreign exchange market since 1982 and other macroeconomic policies stimulated domestic and foreign investment. Domestic and foreign capital investment approved by the government amounted to an average of Rp 2 trillion and US$933.8 million during 1982-87, increased to an average of Rp 5 trillion and US$1,311.0 million during 1982-87, and further increased to an annual average of Rp 31 trillion and US$7.1 million during 1987-92. Bank Supervision in the Deregulation Era Bank Indonesia as the cenral bank has been entrusted the reponsibility of bank superviion since the 1950s. Bank supervision techniques and approaches were adjusted from time to time based on economic policies and the development of the financial and banking systems. Several issues of bank supervision in the deregulation era can be further explored. go90 Bihdi The Philosophy and Objectives of Bank Supervision The philosophy and objectives of bank supervision have remained basically unchanged since the regulation era. Traditionally, the philosophy has been to protect the interest of depositors and maintain the public confidence in banking. As an institution of intermediary and deposit money banking, bank activities significantly affect, either positively or negatively, economic activity and monetary stability. Hence, further development in bank supervision must support tlids function. There are two types of depositor protection: indirect and direct. Indirect protection maintains the soundness of banks and banldng systems. Direct protection can be done through deposit insurance or deposit fund. Indonesia's bank supervision policy is related to indirect protection. Given these, the objectives of bank supervision are to achieve a sound and efficient banking system by maintaining public interest, allowing the banking industry to grow properly, and supporting the implementation of economic development and monetary policy. Scope and Strategy The deregulation era has produced basic changes in the type and complexity of operational banldng patterns and banldng product development The nmber of banks and office networks to be -supervised has also expanded. In addition, the banking industry- is closely related to other fmancial instituions and is affected by macroeconomic conditions. Therefore, the scope of banking supervision has widened, creating a need to develop a new strategy and technique in bank supervision. In February 1991; Bank Indonesia undertook a new strategy in supervising banks and NBFIs based on the following principles. * Banking supervision is to be carried out to achieve a sound and efficient banking system. * Bank soundness should become the concern of all related parties, namely, bank owners, bank management, bank customers, and Bank Indonesia as the banks' supervisor. * Two major aspects-should be taken into account: changes in the way of thinking of all related parties concerning the soundness of the bank, and adjustments of the bank supervision system in accordance with globalization and deregulation of the fiinancial system- Knancial Dereguiwion and Bank Siieisiw The Case of indonsia 91 * The supervision system involves six areas: prudential regulation, creation of a supervisory monitoring system, bank minion, effve discussion, cease and desist orders, and the development of a supporting mechanism. * In the case of a problem bank the central bank will no longer rescue by providing financial assistance but will indirectly support the bank by providing technical assistance, including possible merger with other banks or inviting new shareholders. The implementation of bank supervision strategy is furher elaborated in the following paragraphs. Prudential Regulaion as a Guideline for Bank Operation. Operational rules and prudential regulations are needed to guiide banks in conducting their activities and maintaining soundness. The soundness of banks should be maintained as soon as the bank is established. Therefore, prudential regulations should cover licensing provisions as well as operational guidelines and bank rating criteria. Inprovement of licensing provisions include, among others things, requirements and gaidelines for bank ownership and management, ui for opening new branches and sub-branches domestically as well as abroad, and guidelines for mergers and consolidation. Improvement of operational guidelines includes: capital adequacy ruirts; asset quality evaluation; provision for bad debt; some aspects of credit operations, ncluding legal lending limits and collateral; and foreign exchange operation that includes net open position, swap and reswap, margin trading, and bank guarante in foreign exchange. The rating system is based on the prnciple that in addition to quantitaive evaluation of capital, asset quality, management, earnings, and liquidity (CAMEL), other aspects that may influence the soundess of a bak are also to be evaluated. For this purpose, the bank reporting system has to improve, including its comuterization. Creation of an Earlt Warning System An early warnig system is an important aspect of bank supervision through off-site surveillance. The information needed for an early warning system includes reburns from the bank co , peer group analysis, economic developments, and other external aspects such as the condition of the bank's group. The early warning system is also used to reassess the fairness of the bank rating evaluation. To make the early warning system work, computerization and telecommunication systems are essential. Bank Examination Methods. A proper examiation is very iwportat for bank supervision to objectively evaluate a bank's financial condition and management 92 Blha capability. In this regard, the following aspects should be covered: (a) the asset evaluation system based on a quantitative method and judgment about soundness using qualitative analysis; (b) an exaiination method that is flexible enough to anticipate new banking products; and (c) the quality of the examination and timely reporting of results for which the number and quality of the examiners is very important. For this purpose, the use of public accountants is possible. Effective Discussion to Improve the Soundness of a Bank. To achieve good results, discussion with a bank should be based on open communications and effective problem solving. The discussion should be held periodically by using the early waning and bank examination results. Periodic discussions with a bank can also be used as a means of emphasizing to the bank owner and bank ngement the importance of protect the interest of depositors and other aspects of sound bang. Enforcement of Sanctions Including Cease and Desist Orders. Sanctions imposed on banks can be formal in the form of a fine or a cease and desist order. For the cease and desist order, the bank may be required to restrict the establishment of new offices limit the expansion of business activities and loans, and to replace its agement. If the cease and desist order is considered necessary, a bank supervisor may ask a bank to merge with another bank or to transfer all or part of the bank shares to new investors. Supporting the Eficiency of the Banking Business. Besides supervision, supporting ficilities are needed to increase the efficiency and soundness of the banking system in general. Supportng facilities can include a clearing house, a credit information system, the money market, foreign exchange market, and a code of conduct among bankers. The establisbment of the Indonesian Bankers Institute is needed for these purposes. Legal Framework A strong legal framework is necessary for long-term development of a sound banking system and effective banking supervision. Therefore, a new strategy and framework of the bank supervision system has been developed in the deregulation era, but needs to have a strong legal foundation in the articles of the new Banlking Act of 1992. This is the only act governig banking in Indonesia. Currently, te types of banks are simplified into commercial and rural banks. The scope of operation of rural banks is universal in ntre, focusing on the development of rual areas without actively participating m the giro payment schemes. Both commercal and rural banks are allowed to conduct banking business based on profit sharing principles under a special license. Ilnandal Deregutation and Bank SaqnrvLuion: The Case of Indonesia 93 In the new act, the licensing aspect is stipulated by a provision that covers the organization, structure, a minimum capital requirement, banking expertise, and a feasibility study. The Minister of Finance grants licenses to new banks after aking into consideration the recommendation of Bank Indonesia. Since Bank Indonesia is responsible for bank supervision, it stipulates provisions covering the soundness of a bank, capital adequacy, quality of management, profitability, liquidity, solvency, and other aspects related to banking operations in accordance with the prudential principles. As part of prudential principles, a legal lending limit is imposed. Besides stipulations on the type of business activities, some restrictions are also included in the Banking Act. With regard to bank surveillauce, a bank must submit to Bank Indonesia all information and clarification concerning its operations. A bank's annual balance sheet and profit and loss statements should be audited by a public accountant. Bank Indonesia conducts examinations of banks, both periodically and at any time deemed necessary. Upon request by Bank Indonesia, a bank must fuly cooperate in the audit of its books and files in its possession, and provide any necessary assistance in verifying any information, documents, and clarification submitted by the bank. The information is not to be made public and is confidential. Another aspect of bank supervision in the new Banldng Act is thie authority of Bank Indonesia to take action in the case of a problem bank. According to the act, Bank Indonesia is authorized to require shareholders to increase the capital or replace the board of management, to ask the bank to write-off bad debts and offset the gain of capitals, and to merge with other banks or sell the bank to other parties that are willing to take over all liabilities. Other actions can also' be taken by Bank Indonesia in accordance with the prevailing laws and regulations. Bank Indonesia is allowed to impose administrative sanctions or fines on the bank or related parties. When the condition of a bank endangers the banking system and all actions taken prove to be insufficient to overcome the problems, Bank Indonesia may propose to the Minister of Finance to revoke the license of the bank. Based on the rommendation of Bank Indonesia, the Minister of Finance shall revoke the license and order the board of directors to liquidate the bank. Improvement of Supervision Qualiy Because of the wide network of banks and banking offices throughout the country, supervision needs to be decentralizd. At the same time, there has to be an increase in the quality and the number of bank examiners and bank supervisors in line with the number of bank offices to be covered and the expansion of the banking business. In this respect, a good recruitment system for bank supervisors and bank examiners should be set up. The services of outside experts, consultan, and other 94 BsiJadL professionals, such as accountants, are also needed for improvement of the supervision system. In this context, it was agreed with the World Bank to provide fiLve types of expertise, namely, for credit, foreign exchange, computerization, bank examination, and early warming system. Another aspect of increasing the quality of bank supervision is intffational cooperation among bank supervisors, either through exchange of information, training in supervisory techniques, or the problem solving of similar matters. Such cooperation icludes membership in South East Asian Central Baukis (SEACEN), South East Asia, New Zealand, Australia (SEANZA), and the International Conference for Bankin Supervision. Bank Supervision and Macroeconomic Policy To achieve a sound banking system and to increase the role of the banking nstitution in economic development, bank supervision policy and approach should closely relate to the macroeconomic policy. The implementation of macroeconomic policy, especially the monetary policy, diety affects the condition of banks and should be the concern of bank supervisors. On the other hand, a prudential approach to bank superiion could affect the development of banking, which, in urn, is related to the performance of the economy. The Indonesian experience from 1989 to 1992 proved tbat the deregulation policy supported rapid growth of banking institutions, whicb has facilitated the high growth of the economy but also caused an overheated economy. Too rapid a growth of the bankig business has affected the quality of bank operations, which in turn affected the quality of banlkng assets. Efforts of the monetary authority to cool down the eacnomy has affected also the banking business and has forced some banks to consolidate. Prudential banking regulations, to some extent, limit the ability of banik to a grow. Whereas during period of slow economic growth, when an increase in banling credit is needed, banks may then exceed prudential regulations to extend credit. In such situetions, bank supervisors face the dilemma of balancing between strict and relaxed rgulation. Cloug Remrks Financial deregulation in Indonesia was implemented in sequence. It started in 1967, but general deregulation only began in 1983, when the government changed the economic policy approach, including deregulation of the real sectors. As deregulation proceeded during the 1980s, the Indonesian baiting industry experincedS an impressive growth in terms of number of banks and bank offices, and Financial Deregulation and Bank Supervision: he Care of Indonesia 9S diversity of financial services and volume of business. This development had a strong impact on macroeconomic variables. An enormous increase in demand, both in consumption and investment, caused an overheated economy. This was reflected in a sharp increase in imports, a deterioration in the percentage of non-oil export growth, and an increase in inflation. To cool down the overheated economy, the government imposed a tight monetary policy. This policy induced lowering the inflation rate, accompanied reduced growth of non-oil exports, increased interest rates, and lowered GDP lowered growth. On the other hand, the quality of credit granted in the overheated period worsened and forced banks to consolidate their operations. This enhanced the burden of bank supervision. The philosophy and objectives of bark supervision in both the regulation and the deregulation eras was basically unchanged. Traditionally, the philosophy was to protect the interest of depositors. In line with this philosophy, the implementation of bank supervision also strived to maintain the public's confidence in banldng and to strengthen the role of banldng in economic activities and monetary stability. Bank Indonesia developed a new strategy and approach to banking supervision in the deregulation era. The new strategy included a basic approach and changes to the way of thinking that the bank's soundness is a concern to all related parties. This approach should be followed by adjustment of the bank supervision system in six areas: operational guidelines, creation of an early warning system, examination method, effective discussion with a bank, cease and desist orders, and the development of support to banks. Decentralization of supervision is also part of the new strategy. Changes in strategies and techniques in the implementation of bank supervision requires high quality bank supervisors in adequate number. A strong legal framework is necessary for long-term development of a sound banldng system and effective banking supervision. Therefore, the new strategy and framework of the bank supervision system as developed in the deregulation era needed a stog legal foundation as stipulated in the 1992 Banking Act. In a deregulated environment, macroeconomic policy and bank supervision policy are closely related. The adjustment in the macroeconomic policy directly, and perhaps adversely, affects the development of the banking sector and thus becomes a concern of the bank supervisors. On the other hand, a bank supervision policy should consider its possible effects on macroeconomic policy objectives. To avoid negative impact, deregulation in the financial sector should be direcdy foliowed by prudential regulation and improvement of banldng supervision; To maintain stability and to support the development of the banldng system, financial deregulaton should also be supported by strong monetary instruments. 96 Binhadf Table 4.1. Number of Conmnercial Banks, 1982-92 Private Regiond Total nationa Foreign & development commercial Year Stae banks banks joint venture banks banks 1982 7 86 11 27 131 1985 7 80 11 27 129 1987 7 80 11 27 125 1988 7 77 11 27 122 1989 7 103 21 27 158 1990 7 122 28 27 184 1991 7 142 29 27 205 1992 7 157 29 27 221 Table 4.2. Number of Bank Offices, 1982-92 Privare Foreign & Regionad Total BRI- Stae national joint devlopment conmercal Rural vllge Year bank banks venture banks banks barks unfis 1982 854 339 86 196 1,475 5,808 3,569 1985 909 442 92 222 1,665 5,835 3,628 1987 992 552 91 235 1.870 5,783 2,358 1988 1,030 658 87 270 2,045 7,706 2,585 1989 1,137 1.512 92 335 3,076 7,770 2,843 1990 1.269 2,591 118 459 4,437 8,006 3.040 1991 1.333 3.282 123 580 5,318 8.296 3.210 1992 1,430 3,451 133 612 5,626 8,520 3.195 Financial Deregulation and Bank Supervisiown The Case of Indonesia 97 Table 4.3. Deposits, 1982-92 (billions of Rp) Foreign joint Regiond Private & venture dewelopmet Year* State banks national banks banks banks Rural banks Total 1982 6,165 1,186 908 411 41 8,711 1985 12,905 4,538 1,764 825 213 20.245 1987 22,165 10,924 2,534 1,234 - 36,857 1988 22.511 [1,132 2,515 1.300 401 37,859 1989 29,618 19.468 3,274 1,674 395 54,429 1990 40,535. 33,078 5,423 2,550 521 82,107 1991 41,812 43,142 6.935 3;228 606 95.723 1992 51,428 48,28 7,58 3,700 686 111,687 * End of the year. Table 4.4. Loans, 1982-92 Foreign & Rional Privae jon vetwr development Year* Ste banks nadonal banks banks banks Rirl bamns Total 1982 9,322 1,177 666 357 61 11.583 1985 19,125 4,106 1,073 640 234 25.178 1987 28,308 10.538 1,743 1.147 - 41,736 1988 35,984 10,809 1,913 1,196 501 50.403 1989 45,106 18,758 3,115 1,625 455 69.059 1990 59,811 35S2i7 6.177 2.302 656 104.163 1991 67,845 43,797 9,407 3,032 960 133,914 1992 76,718 43,797 9,407 3,032 960 133,914 * End of the year. 98 Binhadi Table 4.5. GDP, Export and Import, 1982-92 (millons of $) Export Import GCDP Oil & Non-Oil Percentage Oil & Non-Oil Percentage Year growth gas & gas change Toal gas & gas change Total 1982 2.2 15.869 3.878 -12.48 19.747 -4.433 -13.421 10.12 -17.854 1983 4.2 13.696 4.993 28.75 18.689 -3.832 -13.896 3.54 -17.728 1984 7.0 14.979 5.775 15.66 20.754 -2.937 -12.110 -12.85 -15.047 1985 25 12.549 5.978 3.52 18.527 -2.553 -10.152 -16.17 -M705 1986 5.9 7.740 6.656 11.34 14.396 -2.181 -9.757 -3.89 -11.938 1987 4.9 8.571 8.635 29.73 17.206 -2.227 -10.305 5.62 -12.532 1988 5.8 7.832 11.677 35.23 19509 -2.103 -11.728 13.81 -13.831 1989 7.5 8.914 14.060 20.41 22.974 -2.406 -13.904 18.55 -16.310 1990 7.1 11.931 14.876 5.80 26.807 -3. 22 -18.233 31.13 -21.455 1991 6.6 11.455 18.180 22.21 29.635 -3370 -21.464 17.72 -24.834 1992 - 10.492 22.010 21.07 32.502 -2.900 -23.581 9.86 -26.481 Table 4.6. Money Supply and Inflaion Rate, 1982-92 (jilio ofAp-) Percentage Percentfage Rate of -Yar ml growth II growth ltion 1982 7.2 10.9 11.9 22.2 9.69 1983 7.6 5.6 14.7 23.5 11.46 1984 8.6 13.4 17.9 22.3 8.76 1985 10.1 17.7 23.2 29.1 4.32 1986 11.7 15.6 27.7 195 8.83 1987 12.7 6.6 33.9 22.5 8.90 1988 14.4 13.5 42.0 23.9 5.47 1989 20.1 39.8 58.7 39.8 5.97 1990 23.1 18.4 84.6 44.1 9.53 1991 27.3 10.6 99.1 17.1 9.52 1992 28.8 9.3 119.1 20.2 4.94 Financial Deregulation and Bank Superui;onr: 7he Case of Idonesia 99 Figure 4.1. GDP, Non-oil Export and Non-oil Import Growth, 1982-92 40 30 20 10 -20E i E'I 1982 1983 1984 1965 1986 1987 1998 199 190 1991 1992 +----GDP -Export -*-lmport Money Supply Growth and Irltion Rae, 1982-92 45, 35 I 30 - " 25 : 20 15 - S - > _ / o er,'S wU-- 5U 1982 1983 1984 1995 1986 1967 1988 1989 1990 1991 1992 -|_ -InfWatin Ratet-+- -Ml - -- - Ml 100 Bindi Figure 4.2. Deposit and Loans, 1982-92 Deposits 120 - 100. , soo. , 80. 60 40 0 | 20 r ---X.-- ~ )K-- - -.6- -. 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 |-e-- - Smat -I -X- -PNB +J &FR - -x Toldl I Regional Development Banks. Loans 140 . 120 + 100 60 20 20 1982 1983 1984 1985 1986 1987 1988 1989 1990 L991 1992 |- - - -State -A-I -X--PNB ---OJN&FR -+-ToOW L Rcgional D cvlenBanks. Finacia Deregulationandapd B Supenrulsow The Case of hndoneIa 101 References Act of The Republic of Indonesia. 1968. Number 13 of 1968 Concering The Central Bank, December 7, 1968. 1 1992. Number 7 of 1992 Concerning Banking, March 25, 1992. Bank Indonesia. 1991a. 'Further Provision of PAKTO 27, 1988 Concerning Improvement of Bank Supervision," Fcbruary 28, 1991. Bank Indonesia. 1991b. "Supervision of Off-Balance Sheet Banking Operation in Indonesia." The SEACEN Meeting of Directors of Supervision, November. Jakarta. Binbadi. 1990. 'Financial Deregulation and the Role of Cental Bank in Indonesia." The Japanese Regional Bank Representative Mission, Orgnized by the Nippon Credit Bank, Singapore Branch, June. Jakarta. . 1990. "Monetary Policy in Indonesia." The 8th SEANZA Central Banldng Course, November. Bombay, India. . 1991. -How the Bankdng and Financial Services Sector Support Trade and Investment in Indonesia." The Inraonal Conference on Opportunities and Strategies For Doing Business with Pacific Rim Countries, June. Kuala Lumpur. . 1991. "Philosophy and Objectives of Supervising Banks and Fmancial Instiuions." The 19th SEACEN Course on Examination and Supervision of Financial Institutions, October - November. Jakarta. Binhadi, and Paul Meek. 1992. Implemenng Monetary Policy." SEANZA Course at the Reserve- Bank in Sydney in 1988. In Anne Booth, ed, South-East Asian Social Saence Monographs: The Oil Boom and After - Indonesian Economic Policy and Performance m The Soeharto Era Oxford University Press, London. 5 GLOBALIZATION, SPECULATIVE BUBBLES, AND CENTRAL BANKING Michael Dooley Private capital inflows have recently been a widespread and suprising problem for developing countries. Although capital inflows are something that governments would be willing to learn to live with, there is the concern that what flows in for reasons we do not fully understand could flow out for those same unknown reasons. If capital inflows are speculative bubbles, they could burst with dire consequences. It could be argued, moreover, that return in flight capital could prove to be very sensitive to economic developments, both in developing countries and in industrial countries. For example, if inflows have been generated by depressed yields available in developing countries, they might evaporate as yields return to more normal levels in the course of an economic recovery. On a more optimistic note, private capital inflows might fundamentally reflect improved prospects for recipient countries. Inflation declined in many developing countries in 1992, and lasting progress in fiscal adjustnent is evident in many cases. This would suggest that capital inflows are consistent with a fundamental improvement in the expected return on investment in these countries. Since unusual inflows seem to cut across regional lines, a discussion of policy options for dealing with inflows provides a focus for a more general discussion of economic policy in an increasingly integrated world capital market. Recent Private Capital Inflows into Developing Countries In the darkest days of the debt crisis, many believed that developing counties would not reenter international capital markets for a generation. The useful life of this prediction was even shorter than the average for economic prediction, because very large inflows to Latin America and Asia, along with spectacular booms in equity markets, were widespread after 1989. Table 5.1 shows the dollar value of private capital flows into developing counties in Asia and the Western Hemisphere. In some respects the experience of the two regions has been remarkably similar. 103 104 Miclhael Dooley Table 5.1. Privote Capital Flows and the Balance of Payments for Developing Countries in Asia and the Western Hiemisphere (biUions of U.S. dollars) Region 1984-88 1989 1990 1991 1992 Asia Private capital inflows 18 10 28 43 42 Increase in reserves 19 8 23 36 28 Goods and services deficit -1 2 5 7 14 Western Hemisphere Private capital inflows 10 10 24 39 43 Increase in reserves 1 2 15 17 15 Goods and services deficit 9 8 9 22 28 Source: IMF. World Economic Outlook, semi-annual issues, 1984-92. For the Western Hemisphere private capital inflows reduced dramatically following the 1982 debt crisis; inflows averaged less than US$10 billion from 1984 to 1989. However, inflows doubled in 1990 and doubled again in 1991 before leveling off last year. For Asiin developing countries private inflows were less affected by the debt crisis and averaged about US$18 billion after 1984. But here too, inflows increased sharply in 1991 and 1992, reaching levels roughly double the historical average. The counterpart of these private inflows can be divided into official capital outflows, conventionally measured by increases in international reserves and current account deficits. This is an interesting division because the offsetting c-apital flows are associated with sterilized intervention while trade deficits are measures of the contribution of foreign savings to real domestic consumption and capital formation. For the Western Hemisphere the small private inflows before 1989 financed current account deficits. In contrast, in the past three years the much larger private inflow:; have been about equally divided between reserve accumulation and deficits. For Asia almost all the private inflow was matched by reserve accumulations through 1989. The much larger private inflows since then have increasingly supported larger current account deficits, although for three years the reserve buildup stili accounts for two-thirds of the private inflow. These data suggest that policy decisions to offset, or sterilize, the domestic effects of private capital inflows have been an important force in shaping the balance of payments flows in developing countries in recent years. Globalizailon, Spectlative Bubbles, and Cntral Banking 1OS Considerable caution should be exercised in commenting on structural relationships among variables that are jointly determined and that are tied together afterward in an accounting identity. But some interesting empirical regularities seem to emerge from these data. First, a widespread response to capital inflows has been official intervention in foreign exchange markets and increases in international reserve assets. This response suggests that differences in formal exchange rate arraangements might 1?e less important than governments' attitudes toward the effects of inflows on exchange rates, interest rates, and the monetary base. Second, renewed access to international capital markets has allowed many countries to increase investment or consumption relative to GDP. Finally, a sudden reversal of private capital inflows would constitute a significant shock to a majority of the countries in the sample. A rough idea of the economic size of capital inflows is presented in table 5.2, where capital flows are expressed as a percentage of GDP of the recipient country. Not surprisingly, the regional data conceal important differences across countries. For the Western Hemisphere, capital inflows to Mexico in 1991 were 8 percent of GDP, a larger annual inflow than recorded before the debt crisis. With the exception of Brazil and Ecuador, inflows to other countries are impressive, although generally much smaller than levels seen before 1982. For the Asian sample, inflows to Indonesia, Nepal, and Thailand were large by historical standards, while Singapore showed the opposite trend. Implications for Economic Management One of the challenges in adapting the practical work of central banking to a world in which resident's access to international credit markets is growing is to modify the balance sheets that capture the economic interactions between the official sector and the private sector. A failure to keep pace with the new fmancial positions available to residents can result in very unpleasant surprises for monetary authorities. A recurring problem is that which speculative behavior and accompanying market movements at first may appear to be irrational this assessment is often the result of looking at only one part of a new pattem of international transactions. The recent capital inflows experienced by developing countries may be an example of this phenomenon. While we cannot rule out the possibility that such inflows are motivated by follow-the-leader behavior or speculative bubbles, I argue below that arbitrage opportunities created by the monetary authorities explain the size and widespread nature of such capital inflows. The key to the argument is the careful evaluation of contingent liabilities taken on by governments. These implicit liabilities create powerful incentives for private financial transactions, some of which show up in the balance of payments. International capital markets magnify the effects of these incentives. 106 Midwal Dooley Table 5.2. Capital Account Balance as a Percentage of GDP, Western Hemisphere. 1970-91 Yea Argemina Bolivia Brzil Phineps Colobi Ecuador Meaico hdonaia Unigjny Venezela 1970 1.0 -0.4' 2.4 1.7 4.2 7.0 2.8 2.8 0.3 0.8 1971 -0.2 0.1 3.3 1.3 5.5 H.8 2.2 3.2 1.0 2.2 1972 0A 0.3 5.5 2.0 4.0 7.2 2.2 6.0 -2.8 1.3 1973 02 -1.2 4.8 2.2 2.1 3.2 2.6 4.7 -2.3 -.1 1974 -0.0 -0.6 5.6 5.2 2.0 1.8 3.7 0.3 1A -4.9 1975 0.5 3.1 4.7 5.6 1.4 5.0 4.4 -2.9 2.0 1.7 1976 0.5 1.3 5.4 5.8 2.7 5.5 2.7 3.7 4.2 -1.1 1977 1.1 2.0 3.3 3.7 1.1 11.4 2.9 2.1 6.8 6.7 1978 -0.1 1.0 5.6 8.2 1.2 9.6 3.3 2.6 5.8 11.3 1979 4.3 4.7 3.6 6.3 4.3 6.9 4.1 OA 10.1 8.4 1980 1.6 -9.8 4.0 3.7 4.4 7.3 4.5 -1.1 18.0 1.5 1981 1A 10 4.7 2.2 6.1 5.0 6.1 0.8 8.6 -2.1 1982 -3.8 -3.3 3.3 3.5 7.4 3.7 -0.3 3.5 -6.6 -6.8 1983 -3.5 4.1 -0.9 1.4 3.8 -15.0 0.1 7.5 -S. -7.8 1984 0.2 0.3 1.2 -1.1 2.2 -7.8 -0.8 3.5 -0.5 -8.6 1985 -0.7 0.0 -4.2 -6.1 5.9 -7.2 -2.5 1.7 -2.8 -4.3 1986 0A 12 -3.2 -4.7 2.4 *7.9 0.1 1.4 -1.9 -5.2 1987 -4.8 40.7 -3.7 -5.0 -OA 4.2 1.7 3.0 -2.6 1.6 1988 0.3 3.8 -2.8 -2.2 1.6 -76 -2.5 1.6 4.7 2.2 1989 -5.0 -1.1 -1.3 0.6 0.6 -3.0 1.7 2.1 -2.6 -4.1 199C -11 3.5 -12 2.1 0.5 -SA 2.7 5.8 -02 -9.3 1991 1.7 3.8 -1.0 3A 1.4 -4.7 8.2 6.0 0.8 2.4 Of Tahm% Year Pens Korea Malaysia Mymar Nepal Chie Sinapore iLaka Oiiinah ailand 1970 -2.1 5.7 0.1 -0.2 1.3 0.9 42.0 1.9 0.6 2.6 197 -02 6.2 3.8 -0.3 1.5 .0.5 50.5 1.2 0.7 2.4 1972 0.3 3.7 4.9 0.5 1.4 -0.1 31.9 1.5 0.6 3.3 1973 3.3 4.2 1.5 1.7 1.3 -0.1 23.5 1.3 -t.O 2.8 1974 8.4 S.9 72 IA 1.1 -3.1 25.3 2.7 to 4.5 1975 5.8 9.7 5.1 0.6 12 0.1 17.3 1.8 3.9 4.0 1976 2.0 5.3 1.8 0.3 0.7 12 14.8 1.6 0.6 3.2 1977 3.8 3.6 -0.9 3.8 12 5.1 9.2 1.7 -5.3 5.8 1978 -0.0 1.9 1.1 4.2 0.0 12.0 14.2 2.6 -5.9 4.8 1979 -0.8 6.9 -0.6 4.3 1.6 10.8 13.3 8.1 -0.6 7.4 1980 1.1 7.5 3.1 2.7 0.8 I1.8 19.0 9.0 4.5 5.8 1981 3.7 5.4 8.1 3.1 2.7 14.9 17.1 9.1 9.3 7.5 1982 5.8 3.6 12.5 4.0 3.4 3.9 16.2 11.5 -1.7 2.2 1983 -1.4 1.2 11.6 2.7 3.1 -3.1 9.6 9.1 -2.1 6.4 1984 -6.9 1.6 6A 2.0 2.4 5.8 102 5.9 -5.1 6.4 1985 -7.1 1.3 5.8 2.1 2.0 -7.5 7.6 5.1 4.3 4.4 1986 -4.7 -3.8 4.6 2.2 62 4.4 1.3 5.S 9.3 0.8 1987 -4.7 -6.4 -3.7 2.5 5.1 -3.7 4.4 3.9 10.0 2.8 1988 -5.4 -2.8 -8.3 1.5 10.9 -6.0 1.7 4.4 -9.5 6.5 1989 -3.3 -1.0 3.8 2.3 8.2 -3.0 01 4.6 -8.3 9.1 1990 -2.2 0.7 8.8 1.4 12.6 5.9 9.3 4.7 -9.3 13A 1991 0.8 2.7 S.8 1.7 8.4 2.3 1.9 5.1 -IA 12.7 Source: Internauonal Financial S dcs (IFS). Globalization. Speculaf e Bubbes, and CctmIl Baning 107 In a tightly controlled financial system such incentives are difficult to exploit, but as capital markets become more integrated and direct controls on financial ransactions break down investors will find new and better ways to take advantage of explicit or implicit government guarantees. The volmne of such capital flows may have little to do with the economic fundamentals we usually associate with private capital flows. Instead, the scale of transactions is determined by the resources expected to be available'to support the government guarantee. The important difference between this and a bubble is that when a bubble bursts some private speculators lose to others. When an implicit guarantee is called the government loses. Experience in Industrial Countries This problem is by no means unique to developing countries. In fact, there are valuable analogies in the experience of industrial countries. Over the past twenty years the monetary policy regimes in indusi countries have evolved in response to increasingly open financial systems. One result of this has been growing reliance on open market operations rather than direct controls over financial markets. This process has generally occurred slowly as the monetary authorities have adapted their policy tools to a situation in which direct control of finarcial intermediaries has become less effective because of increased domestic and international competition for in ary services. A strong market view of this process, popular in the United States in recent years, is that residual moral suasion or regulatory policies available to the monetary authority are of historical interest but have little consequence for the behavior of the economy. A by-product of this process of deregulation has been the ability of private investors to exploit arbitrage opportunities in ways not possible in a more rigid regulatory environment. At the opposite end of the spectrum, there are very few countries where the central bank directly controls the balance sheets of all financial intermediaries so that market incentives are of little consequence. For example, if an increase in the monetary base is offset by an increase in required reserves and the potential market implications are blocked by controlled interest rates and effective quantitative controls over credit creation, it makes sense to focus on the balance sheet of the finanial system rather than the relatively small and uninteresting balance sheet of the cental bank. Historically, this is the reason internatonal organizations have focused on domestic credit as a summary measure of monetary policy. Given a wide array of mar-ket and control structures, domestic credit or a similar broad measure of financial intermediation is thought to be the best source of timely information about the success the central bank was having in translating intentions into reality. In this setfing, 108 Mchawl Dooley incentives to exploit government guarntees are blunted by quantitative controls on financial positions. Expeience in Developing Countries A useful way to think about the globalization of credit markets is that direct control of prvate sector balance sheets becomes more difficult. Thus, the focus of the central bank's analysis has to shift from direct monitoring of the credit or liability side of a set of private fiancial intermediaries to the central bank's own balance sheet and the incentives that drive the system via market forces. This is not an easy transition to deal with and it seems clear tat authorities in emergig markets today face a pardcularly difficult problem. In contrast to the expience nm industrial countries, international banidng centers and other forms of international financial intermediation are in place, well understood, and alert for new sources of business. Thus, the authorities do not have the luxury of learning along with the private sector as new forms of financial intermediation are developed. As residents gain access to foreign markets, controlled domestic intermediaries are usualy at a competitive disadvantage. When the formerly protected domestic instittions lose their market share, the information content of their balane sheets is also eroded. The typical experience of industrial countries is that domestic intermediaries have lobbied for removal of the restictions on their ability to compete. This generally has meant a rapid dismantling of interest rate ceilings, reserve requ>irement, and otber quantitative controls over their activities. The rapid erosion of the effectiveness of direct controls is likely to expose arbitrage opportunities based on govrernment contngent liabflities. This freedom to exploit these opportunities can create capital flows that are limited only by the ability of the government to live up to its contingent liabilities. lBubbles or Arbitrage? A popular explanation of short-term capital flows that seem to be unrelated to the fumentals is simply that free markets are inherently unstable. A plausible story is that market participants in transition economies have little experience or track records to go on in forming expectations in the newly liberalized environment. It is a fact of life that financial markets are dominated by expectations of future events and that the possibility that bubbles dominate pnce movements cannot be ruled out. From the point of view of economic theory, the issue is that the level of price of an asset depends in part on the expected change in the price. In modern exchange rate models, for example, there is usually an equilibrium path for exchange rates that is characterized by an increasing rate of appreciation. From a more practical perspectve Globalizatiorn, Speculative Bubbles, and Central Banking 109 it is somewhat unsetding that the closer one gets to such mark-ets the stronger the belief among market participants that price movements are often self-reinforcing. Governments have often cited the ill effects of speculative bubbles as the reason for exchange market intervention. In industriai countries, officials were once very straightforward about this, although since the mid 1970s it has been more fashionable to talk about maintaining orderly markets. Thus, bubbles and the central bank intervention to explode them has long been an important motive behind intervention in foreign exchange markets. It has been argued, for example, that market participants get caught up in the excitement of a speculative move in prices or capital flows and carry the value of an asset well beyond that consistent with the fimdamenals. An appealing story is that market participants recognize that an asset is becoming overvalued, but because they made money by holding it yesterday they go along for the ride. The relevant condition aftr all is not that the speculator is correct in valuing the asset but only that someone is willing to pay a higher price tomorrow. I should admit that my own research dating back to 1975 suggests that it is very difficult to reject such a model of behavior in foreign exchange markets. A common feature of such bubbles is that market participants believe that the underlying value of the asset is not important, but expect that the game will last long enough for them to get out before the bubble bursts. It is in this area where the central bank can play a strong role in promoting bubble behavior. This is clearly the case when the government is the someone else expected to provide a backstop, so private investor's losses are limited even if private expectations should change. One can find dramatic examples of this process in both developed and developing countries. The combination of deposit insurance and a relaxation of controls over portfolio selection in the United States led to explosive growth in inflows into savings and loans, and an eventual collapse of the bubble. The problem, clear in retrospect, was that the contingent liability of the U.S. government provided the private investor with a virtual guarantee that high yields offered by savings and loan deposits would not be matched by depositors' losses. Depositors did not question the ability of some savings and loans to offer deposit rates 200 basis points over the market. As long as deposits were probably guaranteed there was little downside risk. The important lesson is that the size of the capital inflow to the savings and loans did not depend on the fundamental quality of the assets they were acquiring. Depositors were protected by the very considerable ability of the govermnent to bail them out Moreover, the interest rate offered by the institutions did not accurately reflect the expected rate of return on the assets. For these reasons it was a mistake to interpret the capital inflow as reflecting an irrational evaluation of the prospects for these financial intermediaries. In developing countries a similar process helped generate the debt crisis of 1982. In this case the government of debtor countries offered guarantees of the dollar 110 Micdael Dooky value of residents' liabilities. The rapid buildup of private external debt in the 1970s was matched by a buildup of contingent liabilities of the government. This contingent liability became an important, but little noticed part of the government's overall financial position. To some extent the offer of the guarantee caused the private capital inflow. Perhaps more important the usual market checks to private inflows were short- circuited by the government guarantee. We would normally expect large capital inflows to depress the rate of return as the most profitable investment opportunities are exhausted. We now know that in many countries this natural limitation was avoided as residents recycled the inflows into private capital outflows, what afterward was labeled capital flight. Thus, calculations from Dooley (1982) suggest that about one-third of the buildup in external debt in the 1980s was matched by unrecorded private capital outflows from the debtor countries. This is a clear example of an unintended and little noticed pattern of financial transactions made possible by the globalization of intemational capital markets. Capital flows of this type can appear to be bubbles because they lead to very large flows that seem to defy the usual market rules that flows are self-limiting. In most cases I believe that what appear to be speculative bubbles in international capital flows are better understood as a part of an arbitrage game being played against a government. This was certinly the way large inflows to U.S. savings and loans avoided self-limiting forces, and explains a major part of the capital inflow into developing countries in the 1970s. Recent Capital lows Is it possible that such a process is again at work in internatiml capital markets? It is interesting to note that the bubble or overlending hypothesis for the buildup of external debt of developing countries was often explained by the story that a bubble can occur once in every generation of bankers. Thus, as the story goes, memory of the widespread defaults by developing countries in the 1930s died off with the bankers of that time and set the stage for irrational, follow-the-leader lending to the same countries in the 1970s. A repeat of the same argument to explain recent widespread inflows to developing counties implies a dramatic reduction in the half-life of bankers, or at least their memories. In this business one learns not to rule out an explanation simply because it assumes a high level of incompetence. But in the current circumstances I believe there is a much better and more important lesson to be learned. The alternative explanation is that a combination of fundamentals and arbitrage trnsactions are again at work. First, the fundamentls clearly favor a repatriation of funds to many debtor countries. Fiscal reform has been impressive in many countries, debt restrucrg has been put in place, and privatization, as well as intentions, has been remarkably successful. At the same time, yields available in industrial countries are at their lowest levels in many years. All these factors make capital inflow to Globalization. Speculative Bubbles, and Cenaral Banking III developing countries unsurprising. But are these factors sufficient to explain the large private capital inflows in the past eighteen months? I have serious doubts. It seems likely that once again private capital inflows are being sustained not only by the more favorable investment climate but also by arbitrage opportunites generated by the governments of developing countries. The form of the incentive is different when compared to the external debt-capital flight pattern that led to the 1982 debt crisis. But in one important respect the recent private capital inflows are similar because they are sustained by a contingent claim on the govermment. In a recent case private capital inflows took the form of foreign purchases of domestic currency instruments. This is certainly different from the dollar- denominated, government-guaranteed, syndicated credits that comprised the buildup in debt before 1982. Thus, it is clear that if a mistake is being made it is not the bankers that are making the same mistake. In the current pattern of capital flows it is not obvious therefore that the government of the borrowing country has provided a guarantee. However, an implicit guarantee might be implied by the increasingly popular use of the nominal exchange rate as an anchor for inflationary expectations. In basing its credibility on the maintenance of a fixed or managed exchange rate, the government, in effect, provides an exchange rate guarantee for the investor in domestic currency denominated instruments. Moreover, the credit risk is also implicitly or explicitly guaranteed by the government. When inflow is to govermnent securities, the guaantee is explcit. Bank liabilities are implicitly guaranteed especially if the deposit is denominated in domestic currency. The current pattern of private capital inflows and official outflows also blunts limiting price adjustments. Recall above that in the 1970s capital flight in the form of private capital outflows recycled fimds to the international market and prevented a limiting fall in yields on debtor country liabilities that would normally limit the inflow. In ffie recent round of inflows, funds are recycled to international markets in a form that is easy to measure because it is typically an increase in the international reserves of the central bank. For Latin America about half of the recent private capital inflow has been matcbed by increases in official reserves and, for Asia, about two-thirds of inflows have been recycled. The engine that keeps all this going is the willingness of the central bank to intervene in the foreign exchange market to prevent appreciation of the domestic currency and to intervene in the domestic money market to prevent declines in interest rates. In effect, the private sector relies on the political difficulty that the government will have in allowing a depreciation of the exchange rate should things trn around. What limits this process? As long as the developing country central bank offers a higher expected yield on its domestic currency liabilities than it earns on its reserve assets there is in principle no limit to the round-trip capital flows generated. Of course, in reality the central bank's resources are limited, but the scale of private 112 Mickael Dooley capital inflows and official capital outflows necessary to exhaust the central bank's expected net worth can be very large indeed. One way to quantify the constraint is to calculate the quasi-fiscal deficit that the government can support through borrownmg and taxation. This is not a bubble in the usual sense, because the private capital inflow is consistent with the expectation that the process will crash at some point. The crash does not lead to private losses because the government will bail out the private investor. The baUilout in this case involves the rapid exhaustion of the cental bank's international reserves and assistance to the banking system to meet deposit outflows. This is a dangerous policy stance for the government of developing counties. Solutions? The key to the problem in my view is the commitment to the exchange rate as an anchor for and-inflation expectations. Although the exchange rate is a powerful way for the government to signal its intentions to the private sector, exchange rates can encourage international arbitrage transactions that can eventually break the bank. The nominal exchange rate target does not contnrbute to credibility in an environment where private expectations for the real equbrium exchange rates are changig rapidly. Surely, economies undergoing transitions are those for which we have little previous information about the equgilbrium real exchange rate. It seems important, therefore, to break this pattern by introducing greatei wiertanty for the i onal investor. This can be accomplished either by allowing the capital inflow to depress domestic interest rates, that is, by not sterilzing the capital inflow, or by allowing the exchange rate to adjust to market forces. Neither policy will greatly affect the net prrvate capital inflow that is fillly justified by the economic reform packages that have been put in place in many developing counties. It will stop round tnp capital flows that have domiated all fixed exchange rate systems since the globalization of international capital markets. If I can once again cite the experience of the industial countries in dealing with increased capital market integration, most have opted for a flexible exchange rate regime with limhited intervention- The anchor for inflation expectations has been a monetary aggregate, or more simply, a target for inflatiom Many very open developing countries will find it better to fix the exchange rate and relinquish control over domestic interest rates. Like all good lessons, this is a very old one simply dressed up in a new stmctwe of international capital flows. 6 THE FINANCIAL SYSTEM AND MACROECONOMIC STABILITY: THE CHILEAN EXPERIENCE Guiliermo Le-Fort, Central Bank of Chile Over the last fifteen years, the history of the Chilean economy and its financial system has been characterized by deep structural changes and accelerated development, and also by instability and crsis. Overall the experience has been successful, and the Chilean economy has not only recovered from the crisLs but is now much more developed and solid than when the reform process began in the 1970s. However, some consequences of the crisis remain, the main one being the large domestic debt of the centra bank, on the order of one-third of GDP, that generates the central bank's annual qu2si-flscal losses, 15 percent of GDP. The valid policy lessons that can be drawn from this history of failure and success are discussed below. The paper has two major sections. The first section provides an overview of the developments in the Chilean fianmcial system over the last fifteen years, begnming with the start of the reform process in 1974, after which the first symptoms of financial inability became apparent (see macroeconomic indicators presented in table 6.1). This section contiues with a description of the financial cnsis of the early 1980s, the strategy developed to confront it, and the post-crisis financial reforms directed at increasing fnancial stability. The section ends with a discussion on recent financial system developments and current issues. The second part of the paper discusses the relationship between macroeconomic equilibrium and financial stabilty. As is clear from the Chilean experience, macroeconomic and financial stability are closely tied, but their causal relationship is complex and bidirectional. To some extent, macroeconomic instability seems to have played a very important role in generating the financial system crisis, however, it is also clear that microecononic problems within the financial system were in itself a source of overall macroeconomic instability. The paper summarizes some policy lessons drawn from the Chilean experience of the last fifteen years, looldng pardcularly at ways the fnncial system can develop consistently with macroeconomic stability. The most important premise is that a freely operating financial system with strong international links is conducive to economic development. The financial system, however, needs to be regulated to avoid problems that weaken financial stability, such as moral hazard originating from subsidized 113 114 Gdilrnno Le-Fort deposit insurance and the limited liability and small equity levels characteristic of fmancial institutions. In any case, a stable and efficient financial system is a necessary but not a sufficient condition for macroeconomic stability, which also requires well- designed and coherent macroeconomic policies. Overview of the Chilean Financial System By 1973, the Chilean financial system had become highly repressed. Most banks were under goverment control, real interest rates were negative, and quantitaive credit was restricted. The liberalization of the financial system began in 1974 with reprivatization of commercial banks, elimination of restrictions interest rates and banking activities, and authorization for the operation of new financial institutions. Iternational capital flows, however, condnued to be strictly limited until mid-1979 when some restrictions to medium- and long-term international capital movements were lifted. The final steps in this first deregulation process were taken in April 1980 when quanitative limits on monthly capital inflows were lifted. The financial reform was characterized by an important development of the inancial system that was interrupted by the financial crsis of the early 1980s. Total bank credit increased from the equivalent of 5.6 percent of GDP in 1975 to more than 77 percent in 1982, and M2 grew from 6 percent to 26 percent of GDP over the same period. The process of financial deepening was interrupted by the emergence of a financial crisis in 1982, and total bank credit fell to 60 percent and to 17 percent of GDP by 1985. The financial deepening process has resumed since, and by 1992 MN has the equivalent of 25 percent of GDP. In addition, total financial savings increased from 26 percent of GDP in 1980 to the equivalent of almost two-thirds of GDP by 1992 (table 6.1). Liberalization and Financial Distress The liberalization of the financial system was part of an overall govremment strategy to establish a market-oriented economic system open to foreign trade and international capital flows. The liberalization of the financial system was based on the idea that market-determined interest rates would stimulate economic growth, positive real interest rates would increase savings, and market-based allocation would improve the quality of investnent. While the economic system was being reformed and the economy rapidly recovered from the 1975 recession, the financial system accumulated a number of imbalances that increased its vulnerability to external and domestic shocks. In 1982, external shocks and the nonsusminable extenal current-account position led to a large expenditure adjustment that triggered Chile's worst financial and economic crisis since the Great Depression. The Financial System ad Macromecnonmuc Stability: The Cilean Erperience I15 Table 6.1. Financial System Aggregates As percent of total As percentage of GDP loans Financal system Year loans M2 M3 Non-performing las 1975 5.6 5.6 - - 1976 7.1 5.9 - - 1977 11.0 8.3 - - 1978 16.1 11.2 - - 1979 23.1 13.5 - 1.6 1980 36.8 13.4 26.4 1:2 1981 53.7 212 28.3 2.3 1982 76.9 26.2 36.6 4.1 1983 68.0 18.9 34.3 8.4 1984 66.1 17.8 34.8 8.9 1985 59.6 17.1 35.2 3.5 1986 59.9 18.1 41.4 3.5 1987 55.4 19.3 44.6 2.7 1988 52.6 20.0 46A 2.0 1989 54.2 21.2 50.5 1.8 1990 54.6 22.5 57A 2.1 1991 50.6 22.9 61.5 1.8 The financial liberalization process was followed by two periods of financial distess. Te first outburst of distress (1975-78) began with the failure of the whole savings and loan systems, followed by the failure of several unregulatd financial companies and Banco Osorno, a medium-sized commercial bank. Next, several large inancial cooperatives failed in 1977 and 1978. The second period of financial dtrss, and the most severe in Chilean history, began in November 1981 when ffinncial institutions representing 8 percent of private bankcs deposits had to be taken over by the public sector. This period reached its peak in January 1983, when almost all the remaning pnvate commercial banks became insolvent and were taken over by the public sector. Non-performing bank assets plus loans at risk increased to several times the capital of the financial system as a whole. Non-performing loans increased from 1.2 percent of total loans in 1980 to almost 9 percent in 1984 (table 6.1). With a financial system repressed by quantitative credit control and interest-rate ceilings over many years, the Superintendency of Banks had lost the ability to exercise prudential regulation in an environment almost free of restrictions. In addition, in what proved to be a gross misunderstnding of the true meaning of financial liberalization, some officials and analysts argued against the enforcement of all forms of financial 116 G;uilermo La-Fort regulation. The lack of rules and the loose enforcement allowed the buyers of privatized banks to operate with little or no capital and to concentrate their lending on firms owned by shareholders. In addition, financial companies were unregulated, and some of them lent substantial sums to their owners, who used them to buy shares of the eleven commercial banks that were privatized by late 1975. In January 1977, however, the difficulties of the recently privatized Banco Osomo forced government intervention to avoid the bank's failure. As a result of that action, depositors had every reason to believe that they were insured by the govermnent. All the responsibility for policing bank solvency was passed on to the regulators, with no help from private depositors. This event reinforced the conditions for the classical moral hazard problem of subsidized insurance, and is seen as a key to the behavior that led to the financial crisis of the early 1980s. The first outburst of financial distress resulted in the accumulation of non- performing assets in the banking system, but the situation, while bad, was still manageable until the crisis erupted. On November 2, 1981, the Superintendency of Banks decided to take over eight minor banks and financial companies where regular bank examination had exposed insolvency or fraud. The deep recession initiated in late 1981 undermined the solvency of the remaining private banks and a full-blown fnancial crisis ensued. The crisis originated from domestic macroeconomic policies and an extremely vulnerable financial system. The vulnerability resulted from the backlog of bad loans brought about by the risky behavior of domestic banks that loaned funds to clients overexposed to currency and interest rate risks. A deterioration in the terms of trade and the reversal in foreign credit flows ignited the crisis. The current account deficit of 14 percent of GDP in 1981, however, had reached nonsusble levels and the adjustment was unavoidable. The adjustment of such a large disequilibrium. was not a controlled process and resulted in a severe contracton in economic activity that contrbuted to the financial systeem crisis, which in turn amplified the recession. The Crisis, 1982-85 During 1982 the banking sector solvency deteriorated rapidly. It was evident that, for a great portion of debtors, the level of liabilities exceeded the value of assets.1 How did the banks as well as their depositors and debtors allow this to happen? For many debtors the change in relative prices brought about by the reforms of the 1970s had reduced the value of assets to the point of almost eliminating their net worth. From then on the debtors' optimal behavior was to continue borrowing regardless of the interest rate, hoping for a reversal of the economic situation (even though the 1.- For a description and different interpretations of the crisis see Barandiaran (1983), Ramir and Rosende (1989), Reinstein and Vergara (1993). he Financal System andMacroeconouc Stabilrty: The Chileon Expenence 117 probability of this event was very low). The banks that had become agents of economic conglomerates willingly financed risky loans provided they were in the interest of the congiomerate. Banks had extremely low equity levels that created incentives to finance risky projects because they could reap all possible profits while losses, above a certain level (the value of the bank equity), were not a bank problem. Depositors accepted this bank behavior because they were concerned with geting the highest possible interest rate and felt protected from bank failure by the implicit deposit insurance. Finally, regulators were very much behind events, and bank supervision was extremely limited. Durir- 1982 banldng sector solvency deteriorated rapidly. In January 1982 loans in arrears net of loan loss provisions accounted for 10.8 percent of the banking system capital and reserves. This percentage jumped to 19.9 percent in March, to 27.4 percent in June, and to 57.5 percent in September 1982. In March 1983 it rose to 89.5 percent, in June to 104 percent, and in September to 181.2 percent. The rapid increa m non-performmg loans resulted from the extremely high real lending interest, which ranged from 40 percent to 60 percent annually in 1981 and 1982. The extremely high interest rates were the result of several factors, includig the very limited access to foreign credit and the higher external iterest rates, the restrictive monetary policy in operation, the exchange risk and the ewpectation of currency depreciation, the moral hazard that induced banks to increase the interest rates paid on deposits, and the expectation of a discriminatory bailout of bank debtors. After two discrete steps toward currency depreciation in June and August of 1982, the exchange risk was reduced, and real intercst rates began to fill. Chile's two largest econonic conglomerates and many smaller independent investors, which operated on high leverage, went bankrupt as a result of their large foreign currency exposure in early 1983. On January 13, 1983, the government took control of five banks, including the two largest private ones owned by the most important economic conglomerates. Three were closed down, and two were subject to direct supervision. These entities represented a combined 64 percent of capital and reserves of the private financial sector. Several problems of the financial system created the conditions for the financial crisis, including the following: * The implicit deposit insurnce that dissociated yield and risk of bank Liabilities. On the one hand, and due to the existence of an implicit government insurance on bank liabilities, bank creditors were not concerned about portfolio risk.2 On 2. The view that bank liabiities were insured by the government was confirmed in 1976 in the Banco Osorno epsode. Foreign creditors may have also perceived that their assets were insured by the government, as was the case in dte Banco Osorno episode, and later during the 1982 financial crisis when prvate foreign debt received an expost public guarantee. 118 Gillermo Le-Fort the other hand, the government did not regulate and supervise banking activity because the economic authorities assumed that the financial market, like any other market, should be free of regulations and thus left risk evaluation to market participants. 0 The excessive optimism in the economy's growth possibilities overstated wealth and deflated consumers' perceptions of the real level of their indebtedness. The emerging financial difficulties were deemed temporary, and refinancing was easily extended. After awhile, banks' capital fell drastically because of increases in their past due loans. The banks continued to refimance their debtors, attracting resources at high rates and expecting unlikely favorable conditions. In any case, banks were no longer risking their own capital but only that of their insurers and creditors in general. * The most important banks were part of large economic conglomerates. The bank's own capital was marginal within the conglomerate, so preference was given to rescue the firms by granting them loans under conditions ihat were utvorable for the bank. Cleanup and Reform By 1981 the Chilean economy was confronting serious macroeconomic imbalances. Excessive aggregate expenditure had increased the current account deficit to 14.3 percent of GDP in 1981, which led to a high level of foreign indebtedness that peaked in 1985 as the foreign debt to GDP ratio reached 143.3 percent. The domestic side of the foreign debt problem was that local debtors were unable to pay their obligations to commercial banks, which in turn could not discharge their obligations. Facing such a critical situation the authonrties had few options. The first one, which could be called the liberal option, was to allow banks to fail. Thus creditors would have to take over bank assets, sell them, and face the losses generated by the difference between the market value of bank assets and the bank's liabilities. A second alternative was 'to melt the debt" by reducing the real value of obligations via inflation. Nevertheless, in the Chilean case the generalized indication of bank assets and liabilities would have had to be eliminated for this solution to work, or, alternatively, very high rates of inflon would have been required. A third alternative was a monetary reform to reduce the nominal value of bank liabilities, that is, depositors would have to face a substantial portion of the financial system losses. Finally, the last alternative was to socialize losses and defer them over time through a debt restructuring process financed by the central bank. The Financial System and Macroeconomic Stabily: The Chilean Erperlence 119 The Rescue Strategy The Chilean authorities chose the last option. This brought about public guarantees to private sector foreign debt, government intervention in the main local banlk, and the implementation of a massive program to support banks and their debtors. n additional component of this solution was the special guarantee given by the state to private foreign debt. The expectation of receiving this guarantee may help to explain the behavior of foreign banks about private overindebtedness. This solution, however, meant that the central bank had to extend subsidized credits to local debtors while backing domestic and foreign bank liabilities, generating the conditions for its present quasi-fiscal deficit.3 .The central bank bought the non-performing bank loans at par value in exchange for central bank promissory notes thus notably improving the quality of the asset portfolio of banks. The transaction also inciuded a provision to eventually buy back the bad loans from the central bank with a fraction of future bank profits. In that sense the deal meant the defennent of banks losses. The relative size of the assets to be bought back and the bank profits, however, imply that part of these loans will never be bought back. To avoid the monetary consequences of the purchases of Md loans, the central bank carried out most of the operations with debt instruments, %allich in most cases could be sold in the secondary market. This is precisely the origin of the current domestic debt of the central bank. Loans sold to the central bank by the two largest private banks are presented in table 6.2. Table 6.2. Loans Sold to the Central Bank (as percew of total loans) Years Banco Chile Banco Santiago Total 1985 7.0 5.2 16.3 1986 1.9 2.0 5.7 1987 0.4 0.2 0.6 The solution adopted some advantages: it prevented the undershooting of asset prices caused by their simultaneous forced sale; it prevented the closing of activities damaged by very high interest rates, but with reasonable yields and positive capital at the interest rates of central bank refinancing; and it avoided a run on deposits in the financial system. This alternative also involved several difficulties. First, the 3. Quasi-fiscal operations include financial ransactdons of central banks that entail some foim of explicit subsidy or cash transfer. 120 Guiltermo L-Foan refinancing terms allowed debtors to postpone payment of their obligations and await better negotiating conditions. Second, given the difficulty of establishing a correct criterion regarding an accurate evaluation of the degree of insolvency, the conditions set in the refinancing were relatively mild, thus creating expectations of debt forgiveness. Third, these conditions implied an important transfer of resources from the taxpayers to the debtors, and resources that belonged to everyone were used to help a few debtors and banks indifferently. Finally, the refinancing conditions allowed for the continued existence of firms with negative capital, and implied the diversion of resources from the generation of new assets. The rescue strategy also incorporated other important medium-term elements: a policy of macroeconomic adjustment consistent with external conditions, sustined real currency depreciation that more than doubled the relative price of tadable goods between 1982 and 1988, centralized negotiation with foreign creditors, and a new banking legislation (1986) that overcame shortcomings in the area of financial regulation. The major programs designed to help domestic debtors and banks are analyzed in the next section and their effects on the quasi-fiscal deficit. Quasi-Fiscal InWlications The support programs for the financial system began in December 1981 when the central bank and the state bakc granted emergency loans to four banks and four financial companies. These credits were completely written off by the central bank and wiUl never be repaid. Between 1982 and 1985 the central bank extended emergency loans to commercial banks that were not under govermnent control and purchased their non-performing portfolio. The credit assistance enabled commercial banks to continue operating, because the value of their non-performing loans significantly exceeded that of their capital and reserves. The central bank acquired "? percent of the non- performing loan portfolio with promissory notes that yielCf eal annual interest of 7 percent, redeemable in four years. In addition, the sale generated an obligation to repurchase from the central bank at a real interest rate of only 5 percent per annum, redeemable with bank profits over an indefinite term. The orginal arrangement implied a predetermined and finite dme to repurcbase the debt portfolio sold to the central bank. However, in 1989 the authorities decided to change the conditions so that banks use every year up to a given percentage of their earnings to repurchase the portfolio sold to the central bank independently of the time they need to redeem all the debt. It is esfimated that some banks will never be able to do so. AU the loans sold melted together, and the total package is what is now called the subordinated debt, because its payment is subordinated to the earnings of the debtor banks (table 6.3). T7he Ftnancial System and Macroeconomic Stability: The Chilean Experiene 121 Table 6.3. Subordinaed Debt Year US$ millions Percentage of GDP 1984 1,705 8.8 1985 2,738 16.1 1986 3,319 19.7 1987 3,713 19.6 1988 3,513 15.9 1989 3,586 14.1 1990 3.925 14.1 1991 3,920 12.5 Thus, the portfolio purchase program had two implicit subsidies. The first one arose from the fact that the promissory notes yielded a higher rate of retrn thm the cost of the central bank funds at the moment they were issued. The second was the discount rate of more than 5 percent, applicable to bank profits with which the portfolio was to be repurchased. The cental bank had to absorb the losses of banks under government control through allowances to the accounts for risky credits and, in addition, had to purchase the overdue debt portfolio and subsidize new stckholders in order to privatize the banks. The portfolio purchase for these banks was similar to that of the other banks, the sole difference being that these banks must use only up to 70 percent of their profits to repurchase the loans. The process of real currency depreciation began in 1982, which substmntiay inreased the real value of the foreign debt. The government decided to assist foreign curency debtors through an exchange subsidy that ended up being one of the main sources of quasi-fiscal loss. The subsidy consisted in the sale of foreign exchange at a fraction of the official rate. The loss generated by the differential between the official and subsidized rates was met by the central bank. At the early stages the central bank sold the subsidized U.S. dollars directly. Later, the subsidy operated through promissory notes that were given to each debtor according to the value of th6. %xchange subsidy entitlements. The notes were denominated in the indexation unit and were tradable in the secondary market. in addition to the assistance to foreign currency debtors, the central bank refinanced domestic currency debts. The program tcok place mainly between 1984 and 1985, and its beneficiaries were debtor firm, mortgage debtors, and consumption loan debtors. This essentially enabled debtors to transfer short-term liabilities at market rates to long-term liabilities at subsidized rates. This prog,ram also benefited banks because it raised the quality of their loan portfolios. 122 Gzuilermo Le-Fort Other transactions, even though they did not represent financial assistance to debtors or banks, had important quasi-fiscal implications. The exchange rate insurance mechanism (swaps) allowed economic agents to sell U.S. dollars to the central bank at the prevailing exchange rate, with the option of repurchasing them at the same rate corrected by the domestic to foreign inflation rate differential. Given the process of real currency depreciation, the aforementioned mechanism involved large losses for the central bank and profits for the private agents that sold foreign exchange through this mechanism. The cumlative cost of these programs is estimated to be between US$7,000 and US$7.0 to US$9.0 billion, the equivalent of up to one-third of annual GDP.4 One way to estimate this cost is to add up the increase in the cental bak net indebtedness US$5.7 billion in the period 1981-90 and the inflation tax collection, estimated at US$1,800 over the same period. Inflation tax collections are estimated at an annual average of 0.5 percent of GDP, or around US$180 million in 1992 (Eyzaguirre 1992). One of the consequences of the financial crisis is the central bank quasi-fiscal deficit, equivalent to approximately 1.5 percent of GDP in annual cash flow losses.5 This deficit is financed mainly with the issuance of additional domestic debt and with inflation tax collections. The continuation of the sustained reduction in the inflation rate, that begau in 1990 is projected to reduce the inflation tax so that the financing of the deficit will rest mostly on domestic debt. Indeed, given that the domestic debt represent nearly 40 percent of GDP, a 4 percent anmnal GDP growth allows the complete financing of the quasi-fiscal deficit while keeping the debt-output ratio constant Since the long-term growth rate is projected at 5 to 6 percent (in the last nine years it has averaged around 6 percent), the financing of a quasi-fiscal deficit equivalent to 1.5 percent of GDP is consistent with a gradual reduction in the domestic debt to GDP ratio. 4. Detailed estimations of the cost of the crisis can be found in Eyzagurre and Larranaga (1990). They also analyze the macro consequences of the quasi-fiscal deficit. 5. This figure includes promissory notes for US$7,000 million that -the nonfinancial public sector transferred to the central bank in order to balance part of the losses. These publc-debt promissory notes cannot be sold in the secondary market, yield Libor plus 0.5 percent, but annual payments through 1995 are limited to 2 percent of the outstanding debtL The balance is capitalized and will be amortzed in tweny annual installmet beginning in December 1995. 7eFinncial System adMacroeonomic Sabilfiy: The Ckilean Eperince 123 Reform: Thte Banking Law of 1986 The lack of appropriate regulations is among the leading factors that created the conditions for the financial crisis.6 As already mentioned, the pre-crisis baning legislation did not contemplate explicit deposit insurance. Nevertheless, in practice, different signals given by the authorities had created a generalized perception of an implicit deposit insurance, which allowed banks to finance risky loan portfolios with deposits. In addition, regulation and supervision of the system were inadequate and inconsisten with this implicit insurance. There were no regulations to ensure continued supervision of portfolio nsk, loan concentration, and the level of bank equity. Generalized problems of high loan concentration, massive loan mliovers, and increases in deposit interest rates to confront liquidity shortages show the absence of prudential regulations. Moreover, there was no supervision that allowed the early detection of financial problems, and financial legislation was not specific on the actions to follow when financial institutions were in trouble. In practice, the governent had to take control of financial institutions only when failure was imminent, and the law provided no mechanism to order the recapitalization of a failed insttution. The 1986 banking legislation attempted to establish a framework to confront the problems associated with the generation of the financial crisis. This legislation gives great importance to bank superimsion. Under this law, the Superintendency of Banks and Fmiacial Institutions (SBIF) must make available detailed information on the fiancial conditions of each bank. Moreover, the SBIF must publish in the press, at least tbree times a year, its opinion on the status of each banking institution. Private sector agencies are also allowed to participate in bank supervision. To evaluate and classify the qaality and risk of bank assets currently each bank must have at least two private appraisers. The new legislation also places great emphasis on the regulation of activities. Bank activities mainly involve intermediate deposits into loans or central bank documents, confingent operations (letters of credit, fuues, and others), excbang operations, and other services (collections, discounts, and so on). These activities are subject to a number of regulations, including limits on loans by debtors, limits on position in different foreign currencies, limits to mortgage financing, and otbers. It was also ruled that loans to individuals or corporations that were direcdy or inrectdy related to the owners or managers of the bank could not be granted more favorable terms than those approved to third parties in similar transactions. In addition, the 1986 law allows banks to perform new activities in the areas of securities intermediation and financial services through bank subsidiaries. The first area includes securities agencies, stock trading agencies, mutual funds, and investm 6. See Ramnrez and Rosende (1989), and Reinstein and Vergara (1993) for a more detailed description of the 1986 Banking Law. 124 Guiemo Lf-Fort fimds. The second area includes leasing, credit card management, and financial consulig. The authorization to expand banking activities, however, strictly separates traditional banking and subsidiary business. The law establishes different kinds of rules. First, the creation of a subsidiary must be done with segregated bank capital. That is, if aba kwishes to invest part of its own capital in a subsidiary, the contribution is deducted from the bank capital for the calculation of different ratios (debt to capital, and so forth.) Second, the law establishes the separation between the bank and its subsidiaies in terms of personnel and equipment. Third, bank loans to its subsidiaries are subject to limits. One of the main innovations in the 1986 legislaton is related to the provisions the banls have to make for their loans. The provisioning rules requir banks to write off non-performing loans that are estimated to have higher than normal nsks and would force an effective increase in bank capital to compensate for the loss. The rule focuses capit requ ie enitson the economic rather tban the book value of bank equity, which is clearly the relevant one in order to implement effective prudential regulations. The new legislation establishes a maximum debt to equity ratio of twenty. Should a bank exceed this limit, it is assumed that the bank has been involved in activities that might endanger its financial situation, and procedures provided by law are then initiated to normalize the sitation To protct the payments system. an explicit and full insurance is applied to demand deposits. In the case of a run on a bank, the central bank must provides at the earliest possible time, th resources to fuUy guarantee demand deposits. If a bank fails the central bank would have prefeience over other creditors up to the value of the emergency loans. In addition, and to limit the moral hazard problem of deposit insurnce bthe new legislation includes an explicit partal isurance on time and savings deposits. The deposit guarantee covers up to 90 percent of each individuial time or savings deposit, and up to 120 index units, equivalent to about US$2,500 per depositor, per year (articles 141 and 142 of the Generl Banldng Law). The latter limit applies to deposits in the overall banking system and cannot be bypassed by distibuting deposits in different bank. Futher Developments and Current Problems of the Fimancial System The emergence of new forms of financial itermediation during the post-reform period has meant a reduction in the bank's market share, a process of disintrmediation. These new forms of intermediation have a competitive edge for banls, because banks are subject to lower capital standards and looser regulations, but at the same time tbey cannot benefit from explicit deposit insurance. With the emeagence of new financial intermediaries, the banks have been progressively authorized to establish subsidiaries to perform financial activities, such as leasing, ne Fmawcil Syrfem and Macroeconoaic Szbilry: The Chlkan EWerence 125 brokerage, investment bankdng, mutual funds management, financial consulting, and other activities. Technological and financial development, as well as the increasing role of institutional investors, have contributed to financial disintermediation. Besides market trends, regulatory and supervisory institutions may also have some responsibility when they do not keep up with these developments and they maintain restrictive legislation that hinders bank activitiec. The Post-Reform Banking System A particular source, of concern is the unorganized emergence of different financial intermediaries. The problem is not that there are many and different intermediaries. On the contrary, it is probably positive, because, it allows more competition and increasing alternatives to savers and investors. The point is that these events happen with so much dynamism bypassing the regulatory mechanism Thus, for example, it is common to find that regulations and supervision for the same operation, but performed by two different intermediaries, are not the same. The challenge then is to achieve consistency regarding the norms that rule different financial intermediaries. This is especially relevant for countries like Chile, where financial reforms and baning legislation are quite advanced, and where new bank financial irmediaries, most importantly pension funds, have acquired an extraordinary dynamism and play a key role in the capital market. The processes of disintermediation raise new challenges that have not taken place in other counties. It requires to gradually adapt financial legislation and supervision according to changes as they occur so as to deepen the mdernization, efficiency, and contnbution of the finmcial system to the ongoing process of saving and investment In concrete terms, the challenge is updatng the legislation designed to prevent financial instability without hindering the development of financial markets. A second post-reform trend has been the globalization of financial Markes. In spite of the pressures, the authorities have taken a gradual approach to globalization. The sudden and unplanned opening of the economy to international capital flows has been knownL to lead to an undesired appreciation of the real exchange rate, a rapid increase in foreign indebtedness, and financial instability. The opening of the capital accot should be approached gradually to avoid generating macroeconomic imbalances. In Chile during the last three years very important actions have been taken in the direction of opening the capital account eliminating restrictions to capital outflows. Persons and firms are allowed to invest abroad using the informal exchange market and informing the central bank of the operation. An explicit authorization of 126 Grm;eowoLI-Feor the Centrl Bank is needed to use the fomal exchange market for these operations.7 The situation is more restrictive for banks and other institutional investors. Banks can invest only up to 25 percent of their U.S. dollar deposits in highly rated foreign securities. A relaxation of these standards, however, is currently being stidied. Domestic banks can only finance nonresidents in international trade operations involving one Chilean party or in operations within the Latin Amedcan Integration Association (ALADI). Of the other institutional investors only pension fiuds managng entities can invest up to 3 percent of their portfolios abroad in low risk, highly rated securites. 7The Capital Market and the Pension Funds One of the major developments in Chile's financial market in the last decade has been the emergence of private pension fund agencies (AFPs). Chile switched from a public pay as you go system to a privately funded capitlization system in the early 1980s.l Since then social security contnbutions are deposited in AFPs, which in tam imvest these fimds in various istruments, including bank deposits, privately issued secuties, central baik bonds, and stocks. The AFPs have acc latd significant resources that currently represent about 30 percent of GDP and that have been very importan in the deepenng of the Chilean finanial markets. The AFP and the life nsurance companies (whose significant development is also a direct consequence of the new pension system) are the main holders of long-term financial instuments. In Chile, unlike other Latin American counties, there is a deep long-term capital market. Currendy, bonds and deposits issued by the financial system with an orginal matrity of at least one year represent around 20 percent of GDP (table 6.4). 7. The retativey small exchange rate ddifrentil benveen fte formal and informal markes of the last two yeaws is an indicator of the degre of openness of the capital accounto 8. The social security reform consits of the privatizaton of the pension system and fte replacement of the pay as yow go system with one of individual capitalization. The govement sets regulations but does not administer funds. A transitional fiscal problem arose because of the payments of redrement pensions under te old system dLat generate public expenditure, while social security contnbutions are no longer part of public reveue. The Fwancial Systent ad Maeroeconomic Stability: The Culean Eerce 127 Table 6.4. Pension Fund Assets per Financial Instrunents (as of December of each year) Trn deposits Credit notes finacial finacal FmsharArw ad Years SecrdWes insrwnem instrmms debenwes 1981 28.1 61.9 9.4 0.6 1982 26.0 26.6 46.8 0.6 1983 44.5 2.7 50.7 2.1 1984 42.3 12.8 43.1 1.8 1985 42.6 20.9 35.4 1.1 1986 46.7 269 2555 0.8 1987 43.1 32.0 22.2 2.7 1988 35.5 37.5 20.6 6.4 1989 41.6 21.5 17.7 19.2 1990 44.1 17.3 16.1 22.4 1991 38.3 13.3 13.4 34.9 Finazcia and Macroeconomic Stability: Links and Causaty The lfrts between financial and macroeconomic stability are multiple, complex, and difficult to disentangle. In this part of the paper, we attewmpt to identify some of the links and analyze the extent to which the disruptions to macroeconomic stability can be originated in microeconomic problems of the fnancial system, or, alteratively, if macroeconomic and financial stability appear to be closely related. Despite these limitations it appears that the microeconomic problems of the financial system played a significant role in creating the conditions for the macroeconomic crisis of the 1980s. Macroeconomic Oiguis of the Financial Crisis Macroeconomic stability in the Chilean experience appears to be directly related to the value of two key prices that are indicators of the level of aggregate spending and of the sustinability of macrmeconomic policies. The real exchange rate, the relative price of tradable and nontradable goods, and the real interest rate were grossly out of line in the period before the crisis. (Corbo 1985; Corbo, de Melo, and Tybout 1986). From 1978 to 1981 the real exchange rate appreciated by 22 percent, while the real borrowing interest rate averaged 27 percent in annual terms. The two key prices refleed a serious macroeconomic imbalance. Over the same period aggregate 128 Gliitermo Le-Fort expenditure increased at an annual average rate of 10.3 percent while GDP increased only 7.4 percent. Consequently, the external current-account deficit increased from 7.1 percent of GDP in 1978 to 14.5 percent of GDP in 1981. The Importance of Key Prices The macroeconomic imbalances were a source of financial fragility. The adjustment process that followed resulted in a severe deterioration of the loan portfolio. Since the level of aggregate expenditure and the relative prices of tradable and nontradable goods were grossly out of line, the income of borrowers and the profitability of investment projects financed by bank loans were severely reduced by the adjustment. Many bank borrowers were unable to discharge their contractual obligations. The inadequacy of the Chilean macroeconomic policy of the early 1980s has been linked to the fixed exchange rate policy in effect since June 1979. A possible interpretation of this inadequacy was that demand management policies were not appropriate, resulting in an expansion of aggregate expenditure far beyond what could be reasonably considered consistent with the fixed exchange rate. Despite the small fiscal surplus averaging 1.4 percent of GDP in 1978-81, fiscal policy afterward may have been too expansive, partly explaining the rapid expansion of aggregate expenditure that took place in that period. It could be argued that the accumulation of contingent fiscal liabilities, although excluded from the definition of public debt, were considered by private agents to estimate their expected wealth and to decide their spending plans. Among the contingent fiscal liabilities could be included the implicit deposit insurance, the exchange rate insurance, and other contingent support given to balking system borrowers. If these contingent liabilities were taken into account, public sector liabilities would have been growing at a fast rate in the period before the financial crisis of the early 1980s. Monetary policy at the same time proved to be extremely ineffective, especially considering the ambitious inflation objective reflected in the fixed exchange rate. The monetary policy along with the exchange rate policy that basically followed the monetary approach to the balance of payments was inconsistent with the overall macroeconomic situation. To the extent that monetary policy can be represented by the level of interest rate, monetary policy was contractionary because real interest rates were extraordinarily high. These high rates, however, were ineffective in moderating the expansion of real expenditure. The elimination of capital account restrictions under tbese conditions generated a massive inflow of foreign capital that accommodated an even larger increase in aggregate demand, while interest rates remained at high levels. The prevalence of high real interest rates and a rapid increase in expenditure can be reconciled considering that borrowers' expectations included an indefinite refinancing of interest payments or some other form of debt bailout. The Financial System and Macroeconomic Srabilitry The Chikan Erpeflence 129 An important lesson from this experience is that aggregate demand policy should be directed toward obtaining stable macroeconomic conditions for which the sustainability of the key relative prices is a necessary condition. The level of the real exchange rate and that of the real interest rate should be of central importance for policymakers. To attain this goal, demand management, and particularly fiscal policy, should control aggregate spending over the medium tern and not solely concentrate on intermediate targets such as the fiscal deficit. As the Chilean experience indicates, excessive expenditure, whether private or public, ends up being a source of serious macroeconomic disruption, and the main tool to limit the expansion of aggregate expenditure is fiscal policy.9 Moreover, a possible interpretation of this experience is that to a large extent the macroeconomic imbalances had their origins in the mnicroeconomic problems of the financial system. These problems could have resulted in expansive macroeconomic policies due to the accumulation of contingent fiscal liabilities and the ineffectiveness of monetary policy. The debtors' expectations of a bailout and the implicit deposit and exchange rate insurance could not be excluded as factors behind the nonsustinable level of real expenditure. The failure of the macroeconomic policy at that time was that authorities failed to recognize that expenditure expansion was excessive and thus did not attempt to restrain it. The alternatives for policy action included reducing aggregate expenditure using a restrictive fiscal policy or, acting on the source of the problem, attempting a financial reform by limiting deposit and exchange insce and expectations of a debtor's bailout. If neither option was feasible, it was necessary to accept a less ambitious inflation target and adjust the rate of currency depreciation accordingly, to avoid the overappreciation of the real exchange rate and the severe deterioration of foreign accounts. This last alternative would have limited the expansion of domestic expenditure with an inflation tax and may have created some room for contractionary monetary policy. The Chilean experience of the 1980s reinforced the importance of concentrating efforts to keep key macroeconomic prices at sustainable levels and carefuly monitoring the level of aggregate expenditure. Perhaps one result of this macroeconomic policy stance could be a lower rate of GDP growth in one particular year, but under this stance the average growth rate would be higher, and the growth process would be much more stable. 9. The effectiveness of monetary policy is seriously limited by the targeting of the real exchange rate. 130 GuilertmLe-Fort Capital Account Opening The extremely high level of the domestic real interest rate did not eliminate excessive domestic expenditure, but it did attract massive external capital inflows. To a large extent capital inflows were intermediated by the financial system and by private analysts and bankers lobbying for a rapid elinination of the remaining capital account restrictions. Quantitative restrictions on foreign borrowing by banks were seen as the cause for the high domestic interest rate. The rapid removal of quantitative restrictions created more problems than it solved. Interest rates fell but continued at high levels, and the massive capital inflow financed an even larger increase in domestic spending. Eliminating restrictions to international capital flows when domestic interest rates exceed international rates, after performing the appropriate exchange rate and risk corrections, amounted to an expenditure shock that created pressures on domestic prices and expenditures. Although today in Chile there are no quantitative restrictions on foreign borrowing, some restrictions to capital flows still apply. A 30 percent reserve requirement is imposed on all foreign capital inflows, with the sole exception of equity fiacing. This reserve requirement is not remunerated and must be kept in the central bank for a year, thus increasin; the cost of foreign financing, particularly for short-run operations. The removal of these restrictions should be step-by-step in order to avoid a sudden increase in expenditure and to allow for a careful monitoring of bank lending. The removal of restrictions affecting capital movements and exchange transactions is in itself desirable, but they should be part of a well-planned strategy. The excellent macroeconomic performance of the last three years should not be endangered by an impulsive opening of the capital account. Openness to Trade and Vulnerability to Enaernal Shock One of the most notable characteristics of the international economy during the 1980s was the rapid growth of international trade. Between 1983 and 1989 the volume of international trade increased at an average annual rate of 6 percent compared to an annual growth rate of the world's GDP of only 4 percent. Overall, during the 1980s world trade volume increased by 50 percent; and an important share of this growth has take place with goods of a high technological content, and with a greater number of countries taking part in technologically advanced trade. Technological progress in management, communications, and information processing has globalized the marketplace because numerous institutional obstacles and political frontiers have been removed. The dynamics of this process should be sustained because it has increased the welfare of citizens of many countries. The emergence of interational trading blocks may segment international markets, endangering the gains derived from trade and specialization in a global marketplace. The FInancal System and Macroeconwmic Sabilhy: The Chikan Experience 131 Chile has favored multilateral openness as the best way to promote specialization, efficiency, and growth. Chile began to unilaterally open to international trade almost two decades ago. Over this period the relative size of Chilean exports have doubled as compared with domestic activity with world trade. Exports of goods and services that represented 15 percent of GDP in 1970 reached 35 percent of GDP in 1992. In 1970, for each million U.S. dollars of world exports, Chile exported US$940 of non-copper goods, whereas in 1991 for each million U.S. dollar of world exports Chile exported US$1,540, almost doubling its share in world trade. However, if copper exports are also included in this calculation, the share of Chilean exports in world exports would have fallen from US$3,800 per million dollars of world exports in 1970 to US$2,600 per million of world exports in 1991. This reflects the importance of copper exports in Chilean trade and the significant reduction of the real price of copper over the last thirty years. Over the last two decades Chilean trade restrictions have been drasically reduced. In fact, in the early 1970s trade tariffs averaged more than 100 percent, had a wide variance, and were complemented by quantitative trade restrictions. Today, quantitative restrictions have been eliminated, and a single uniform 11 percent tariff rate is applied. Under this strategy of openness to internatonal trade the country is vulnerable to internatinal shocks, partcularly to the extent that exports are not widely diversified. Despite some diversification in recent years, copper and a few other export products still represent more than 70 percent of Chilean exports and more than 20 percent of Chilean GDP. This risk could be reduced by additional trade diversification, but some of the benefits from trade, which come precisely from specialization, would then be lost To allow for a reduction in risk faced by domestic residents and to still fully benefit from trade induced specialization, contingent claims should be traded in international capital markets. Openness to international capital flows, mcluding direct foreign investment, allows for an efficient diversification of risks, thus reducing national exposure to external shocks. Along the same lines, the regulation and supervision of the domestic financial system is oriented to reduce risk to the financial system. However, to the extent that the domestic production system becomes more specialized, possibilities for effecively diversifying loan risks become more restricted. Consequently, overall risk can be reduced if domestic financial institutions seek investment opporunities abroad. Of course, the relaxation of restrictions on bank investments abroad creates the need to improve regulations on international lending transactions so as to limit and monitor commercial and currency risks of intemational lending by domestic banks. 132 Gadlermo Le-Fort Financial Origins of Macroeconomic Disequilibrium While macroeconomic instability can negatively affect the financial system, the inverse is that financial instability can damage the economy. A financial system that is not functioning well can affect the economy because it introduces inefficiencies at the micro level in the intermediation of savings. In addition, such a financial system can be a 'contributing factor in the beginning of a macroeconomic crisis or, more commonly, a factor that helps deepen a crisis. The following paragraphs analyze the externalities imposed by a financial system that does not work properly, followed by a review of the intrinsic fragility of the financial system. In both cases the importance of good regulation and supervision of the fimancial system becomes clear. Finally, macroeconomic costs of the Chilean financial system in the early 1980s are briefly discussed. Financial System Externalities Financial problems can negatively affect the real economy, first, because the fnancial system, through performing the role of intermediatinn of savings, can affect the efficiency of investment and the growth possibilities of the economy. Given the limitations on financial intermediation, (for example, information and transaction costs, the need for project screening and for monitoring loans, among others), specialized institions are required to efficiently carry out intermediation. Moreover, a well- organized fiancial system allows for more efficient risk management, risk sharing, and reduction through transactions in contingent claims. In discharging this function, the financial system can make a significant contribution to development, but when it fails, severe macroeconomic disruptions follow. The financial system also has a very important role in providing the economy with an efficient payments system. Demand deposits are the most important means of conducting transactions. Consequently, the failure of a financial institution can create serious disruption in the payments system. Not surprisingly, one of the main concerns of economic policy when confronting the failure of a financial institution is to provide for the continuing operation of the payments system, and to avoid the problem of an individual bank escalating into a general crisis. The disintermediation of savings, the difficulties for carrying out risk-sharing operations, and the limitations for carrying out current transactions through a failed payments system can generate significant economic costs. It is clear that the financial industry is the only one where localized problems can develop into a bank run and an economywide crisis. The externality that affects the financial system implies the need for a protective net for depositors that limits the risk of a bank ran. Endogenous financial problems have been the object of study in economic literature (Charles Kindleberger (1978) describes a process of financial manias. panics, and crashes). T7he Financial System and Macrocconomkic Sabillry: DTe Chilean iperience 133 Strucural weaknesses of the financial system increase the likelihood of this self- destructive process. Structural Weaknesses of the Financiol System With deposit insurance, depositors have no incentive for discriminating between financial institutions according to the risk of their portfolios; theiy are only conerned with the return offered.'0 Financial institutions tend to increase the risk of their portfolios to increase returns and attract a larger share of deposits. This risky behavior tends to be aggravated by the low level of bank capitalization. In theory, the solution to this moral hazard problem requires bank supervision and regulation. Regulations should atempt to impose minimum capital requirements directly related to the risk of the loan portfolio. Higher levels of risk demand stricter loan provisioning and higher bank capitalization. Deposit insurance should be explicit and limited, and if possible, the cost should relate to the risk of failure faced by the banking institution. Finally, bank supervision is needed to enforce regulations and to publicize the information on the financial conditions of the banldng institLtion using well-known accounting criteria. Quasi-Fiscal Losses The origin of the Chilean financial crisis has been related to implicit deposit insurance and contingent liabilities and transfer given out by the cental bank to depositors and debtors. The effect of the implicit insurance and the contingent transfers on macroeconomic stability has already been discussed, bwu the effect of these quasi-fiscal operations continued both during and after the crisis. At the outburst of the crisis the central bank provided the contingent liabilities, and the equity of the cental bank was seriously impaired. The central bank debt issued to finance the tEaLisfers will represent a burden on fiscal resources for several years to come. The solution of the Chilean financial crisis was built upon financial assistane and transfers given by the central bank to depositors and debtois. The financial assistance and transfers resulted in important losses that continue to generate cash losses for the central bank. These losses, or quasi-fiscal deficits, are equivalent to ofter types of public expenditure that have a detrimental effect on national savings and should be taken into account when designing fiscal policy. ikf fiscal policy does not compensate for the lower savings ratio generated by the quasi-fiscal losses, the central 10. For a formal modeling of these issues see Le-Fort (1991). 134 Quiflenmo Le-Fon bank may be forced to increase the real interest rate in order to avoid pressures on prices and external accounts. Moreover, in an extreme case of fiscal irresponsibility, monetary policy may lose its effectiveness1 owing to a large international interest rate differential and the inconvenience of further real appreciation. The monetary authorities may then be forced to use the inflation tax to finance the quasi-fiscal deficit. Concluding Remarks Financial system reform in Chile has finally been successful. This success was obtained at an important transitional cost in tx form of a crisis that ended with almost all the banking system under government control and the worst recession since the Great Depression. The lifting of financial repression, inspired by the work of McKinnon and Shaw in the 1970s (McKirnnon 1973; Shaw 1973), was poorly implemented. The crisis generated a significant fiscal cost that, over and above the lower output of the recession, is estimated to have reached up to one-third of the annual GDP. Moreover, this cost will continue to contribute to the central bank's quasi-fiscal deficit, which is equivalent to around 1.5 percent of GDP for many years. The Chilean crisis of the early 1980s may have been triggered by inadequate macroeconomic policies and external shocks, but the structural weaknesses of the financial system created the conditions for a crisis that sooner or later was going to burst Problems of moral hazard in the financial system had resulted in particularly risky behavior by financial market participants that weakened ban' solvency and increased the likelihood of a crisis. To a large extent the macroeconomic conditions of excessive aggregate expenditure that existed before the outbreak of the fins icial crisis were the result of inadequate operations within the financial system. It can be argued that economic agents believed that the government was insuring the deposits and foreign currency exposure of both banks and bank debtors. The perceived contngent wealth tansfers that were implicit in the insurance mechanisms helped foster excessive spending and weaken the effect of high interest rates on aggregate expenditre. Given the importance of externalities generated by the proper operation of the banking system, it seems inadequate to eliminate all forms of deposit insurance. With deposit insurance there is the need for adequate bank supervision and regul? 'on to avoid the moral hazard problem associated with subsidized insurance. The development of the domestic financial system and the integraton of the international financial system had a positive effect on macroeconomic performance and stability- The experience of the 1980s indicates that the structural change involving the financial system must be done gradually in order to minimize the transitional costs that can be created. Finally, financial liberalization should consist of reforming regulations affecting financial transactions, but not in the complete elimination of these regulations. Regulations should be directed at avoiding the excessive risk taking fostered by inappropriate incentives present in an unregulated financial system. The Ftnancial System and Macroeconomic Stability: 7he Chtilean Experience 135 Finally, in a dynamic world it is important to have flexible legislation that can adapt to new changes. In this sense the processes of globalization and disintermediation impose a new challenge to the regulatory authorities. The challenge is not to discourage the normal development of a modern financial system but to avoid processes within financial development that may bring about a perverse structure of incentives that finally ends up in a financial and macroeconomic crisis. 136 Guillermo Le-Fort References Barandiaran, Edgardo. 1983. "Nuestra Crisis Financiera.' Estudios Pziblicos 12 (Spring). Benston, G. J., et al. 1986. Safe and Sound Banking. Cambridge, Mass. MIT Press. CEPAL. 1991. Iternalizaci6n y Regionalizaci6n de la Economna Mundial: Sus Consecuencias para Ameica Latina, (September). Corbo, Vittorio. 1985. 'Reforms and Macroeconomic Adjustments in Chile during 1974-84.' World Development 13 (August). Corbo, Vittorio, Jaime de Melo, and James Tybout. 1986. "What Went Wrong with the Recent Reforms in the Southern Cone.' Economic Development and Cultural Change 34 (April). Diamond, Douglas W. 1984. "Financial Intermediation and Delegated Monitoring." Review of Economic Studies 51 (July). Eyzaguirre, Nicolis, and Osvaldo Larranaga. l990. Macroeconomra de as Operaciones Cuasifiscales en Chile. Santiago, Programa Post-grado de Economia ILADES Georgetown University, (Serie de Investigaci6n, 31) Eyzaguirre, Nicolis. 1992. "Financial Crisis, Reform and Stabilization: The Chilean Experience Up to September.' Paper presented at the Senior Policy Seminar Financial Sector Reforms in Asian and Latin American Countries: Lessons of Comparative Experience, May 25-28. Santiago, Chile . Kelly, Edward D. 1991. "Supervision and Rescue of Financial Institutions: The U.S. Experience." Paper presented at the 2&h Meeting of Governors of Central Banks of the American Continent, October. Santiago, Chile. Kindleberger, Charles. 1978. Manias, Panics. and Crashes: A History of Financial Crises. New York: Basic Books. Larranaga, Osvaldo. 1991. "Autonomia y Deficit del Banco Central." Colecci6n de Estudios CIEPLAN 32 (June). 7he FinWACiOJ System awid Macroeconomic Stabiliry: TJhe Ghkta Ezperience 137 Le-Port, Guillermo. 1991. uFinancial Crisis in Developing counties and Structnral Weaknesses of the Financial System," AndIisis Econ6mico (November). McKinon, 1. 1973. Money and Capital in Economic Development. Washington, D.C.: The Broolings Institution. Marshall, Enrique. 1991. "El Banco Central como Regulador y Supervisor del Sistema Financiero," paper presented at the 28th Meeting of Goiemrors of Cental Banks of the American Continent, Santiago, Chile (October). Modigliani, Franco, and Merton Miller. 1958. -The Cost of Capital, Corporation Finance, and the Theory of Investment," American Economic Review 48 (June). Ondarts, G. 1992. 'La Nueva Integraci6n" en Integraci6n Lainoamericana No. 175 (January - February). Ramfe, Guillermo, and Francisco Rosende. 1989. "Anilisis de la Legislacion Bancaria Chilena," Senie de Eseudios Econ6mtcos 35, Cental Bank of Chile. Reinstein, Andr6s, and Rodrigo Vergara. 1993. 'Requerimiento Optimo de Capital para el Sistema Bancario Chileno," Estudios Pbl1cos (Summer). Rosende, Francisco, and Rodrigo Vergara. 1986. QOpciones de Politica para el Sector Fmanciero," Cuadernos de Economia (December). Shaw, R. 1973. Financial Deepening in Economic Development. New York Oxford University Press. Smith, Clifford W., and Jerold B. Wamer. 1979. 'On Financial Contracting, An Analysis of Bond Covenants," Journal of Financial Economics 7 (June). The Department of the Treasury. 1991. "Modernizing the Financial System Recommendations for Safer, more Competitive Banks," Washington, D.C. (February). Valdes, Salvador. and Alejandra Lomakin. 1987. "Percepci6n sobre la Garantia Estatal a los Depcsitos durante 1987 en Chile," Cuadernos de Economia 25 (August). 138 Guilermo L-Fora Wisecarver, Daniel. 1983. "Dogmatismo y Pragmatismo: una D6cada de Politica Econ6mica en Chile," Esrudios P,2blicos, 11 (Winter). Zahler, Roberto. 1992. "Financial Sector and Liberalization: Opening Remarks" Paper presented at the Senior Policy Seminar Financial Sector Reforms in Asian and Latin American Countries: Lessons of Comparative Experience, Santiago, Chile (May). 7 SECURITIES MARKET REGULATIONS AND REFORMS IN THAILAND Ekamot Kiriwat, Deputy Governor, Bank of Thailand As with many fast developing countries, Thailand has struggled to develop an infsmcture to keep pace with the economic development- The government raised massive fncial resources to pour into sectors that are regarded as essential for the rapid growth of the economy. However, not only were the end users of these fmancial resources trying to keep up with the changing pace, but so was the whole system in which money could effectively be raised to accommodate those needs and demands. The capital market in Thailand, though attracting huge interest among investors and fund-raisers in recent years, was too ill-equipped to support financial innovation so necessary to propel Thailand into the international securities market The Thai capital market was governed by outdated laws that restricted direct financing by companies. Moreover, jurisdiction over the capital market was split among various entities ranging from the Ministry of Finance, the Ministry of Commerce, the Bank of Thailand, and the Stock Exchange of Thailand. As a result, there was no coherent policy direction for market participants to follow. Thus, the market was left to develop through a maze of laws and regulations that were sometimes irreconcilable. This paper illustrates in more detail the problems facing the Thai capital market and how those problems were tackled by the reforms brought about through the promulgation of the Securities and Exchange Act. Essentially, the act created a new environment for the capital market and paved the way for financial innovations. Weaknesses of the Old Legal Framework and Financial Innovation A major weakness in the old legal system, which hindered financial innovation, was that there were many laws that securities market participants had to observe, namely, the Stock Exchange of Thailand Act, B.E. 2517, governing the activities of the Stock Exchange of Thailand (SET); the Act on the Undertaking of Finance Business, Securities Business, and Credit Foncier Business, B.E. 2522, governing the business of securities companies; the Public Company Act, B.E 2522, governing the public offering of shares and debentures; and the Civil and Commercial Code providing 139 140 Ebcamol rivu general provisions reprding the civil and commercial practices in Thailand, such as the setting up of limited companies. The greater the number of laws, te greater the chance of inconsistency among them which certainly created a burden on market participants. After the enactment of the Public Company Act and the amendments to the Civil and Commercial Codes in 1978, limited companies were not allowed to make any iritial public offering (IPO) of their shares, nor were they allowed to offer debentures to the public. New shares had to be initially offered only to existing shareholders. These provisions significantly obstructed the mobilization of funds in the primary market, thus having an adverse effect on the secondary market According to the provisions of the Public Company Act, only public companies could make an IPO of their shares and debentures. However, after over ten years since the Public Company Act had been enacted, only thirty-three companies had become public companies because of the strict provisions and some other weaknesses of the act The problems with the act included the severe criminal and civil libilities on companies' directors, and the requirement that any company having more than thy- tree shareholders had to convert itself to a public company regardless of its intenon whether to raise funds from the public or not In 1984 the Stock Exchange of Thailand Act was used to overrule some of the provisions in the Public Company Act and the Amendments to the Civil and Commercial Codes. The Stock Exchange of Thailand Act allowed listed and authorized companies in the Stock Exchange of Thailand (SET). Companies that were tentatively approved to be listed and authored in the SET and were waiting for final approval from the Finance nster could offer their sbares and debenunres to the public. The Stock Exchange of Thailand Act, however, was primarily a regulation on the trading of secuities in the secondary market, and was not intended to be a regulation on the primary market Tbus, most of the listed companies in the SET were limited companies according to the Civil and Commercial Codes, and hence were not regulated by the Public Company Act Consequently, there was no effective legal framework to supervise the primary market, that is, nobody cared about the development of the primary market. Weaknesses in Supervision and Enforcemnem Apart from the number of laws that needed to be observed, there were also variou supervisory agencies in charge of the securities businesses, namely, the Minister of Finance, the SET, the Bank of Thailand, and the Ministry of Commerce. Thus no single supervisory agency had an overview of the securities busiesses. For example, the Bank of Thailand supervised securities companies but did not supervise the actvities in the SET, md the SET supervised member brokers but did not have the authority to regulate nonmember brokers. These inconsistencies and the large number Securities Markct Regulations and Reforms in hrailand 141 of supervisory agencies involved created inefficiency in the enforcement of securities regulations. The Need to Develop Financial Instruments Apart from the obstacles and weaknesses in the legal framework that had to be eliminated, it was also necessary to enhance the development of financial instruments. The Thai financial market was becoming more deregulated; market forces were allowed to work more fully with less intervention. Therefore, the financial managers' or individuals' needs fnr financial instruments to assist them in managing cash flow and risk were expected to gain more importance than in the past. Several debt and equity inuments werc useful in seeding direct fiancing. More varieties of financial instruments would promote the mobilization of savings, especially to support the long- term savings as a result of the availability of secondary markets of several financial instruments. Consequently, the enactment of the Securities and Exchange Act was designed to achieve four main objectives: * To set a framework for the development of financial instruments to become an imnportant funding vehicle for Thai businesses. Limited companies that are granted approval by the Securities and Exchange Commission (SEC) can issue debt instruments to the public, increasing the supply of debt instruments in the market. * To provide better protection for investors. * To make the securities supervisory systems more transparent and more unified. * To facilitate the development of securities business and the Stock Exchange of Thailand. The Securities and Exchange Act introduced several important changes to expand and improve the securities market in Thailand. The following paragraphs describe these changes. Unification of Supervisory Agencies The Securities and Exchange Commission was established to supervise all aspects of securities businesses. Currendy, eleven commission members are appointed 142 Eaamol Iiriwat from the various agencies that previously supervised securities businesses from financial, accounting, and legal companies and from professionals in the securities business to represent every party concerned. The SEC is supported by the Office of the Securities and Exchange Commission, which is headed by the secretary-general. Issuance of Securities The Securities and Exchange Act stipulated that the mobilization of funds from the public by using stocks or equity instruments is limited to public companies, while debt can be issued by both public companies and limited companies. These companies, however, must obtain approval from the Office of the Securities and Exchange Commission to mobilize funds from the public. Public Offering of Securities As mentioned earlier, the public offering of securities by public companies was formerly regulated by the Public Company Act, while the public offering by listed companies was partially regulated by the Stock Exchange of Thaiand Act. No provision regulated public offering of outstanding securities by existing shareholders of limited companies that were not listed or authorized in the SET. Companies thus could circumvent the laws by selling new securities to existig shareholders who later sold the securities to the public. The Securities and Exchange Act closed this loophole by empowering the SEC to supervise all types of public offering of securities: short-term, long-term, new, and outstanding securities. One of the key concepts of the Securities and Exchange Act is to create transparency in the securities business. The issuers must disclose as much informaton and be as reliable as possible to the investors to enable them to make informed investment decisions. Debt Instrwnents, Investor Representatives and Trustees Under the Securities and Exchange Act, the definition of debentre includes all kinds of debt instruments, both short and long term, except "bill" as defined in the Civil and Commercial Code. This is intended to prevent companies from circumventing the law by caling basic debt instruments by other names. The act also introduced the concept of investor representatives and trustees under the section on secured bond and fumd management. This provision gives investors protection and prepares the groundwork for the supervision of sowritization of assets. Securities Market RegAdtions and Reforms in haiad 143 Securities Business The Securities and Exchange Act clearly defined rules, procedures, and supervision of each type of securities business, and added private fund management as a new type of securities business. Thus, securities companies can now manage the surplus fund of persons and foundations. For a mutual fimd company, the mutual fund approved by the Office of the SEC will be a Thai juristic person and a separate legal entity from the securites company. The provision on the supervision of securities business emphasizes the importance of capital adequacy and increases the required capital for securities companies. The SEC is empowered to set the capital requirement in relation to the volume of activities or the net position of securities companies. The Stock Eichange of Thaiand (SET) The Securities and Exchange Act restructured the authority and responsibility of the SET by transferring the enforcement work from the SET to the SEC. The SET is responsible for overseeing the operational work of the Securities and Exchange Commission. The act also permitted smaller, over-the-ounter markets to be set up with no cross-listing wifth the SET and laid the groundwork for the establishment of various securities related organizations, such as the Securities Depository Center, the Securities Registrar Office, and the Securities Cleaing House. These organizations will be privately run. Investors' Protection The Securities and Exchange Act specified a number of requirements to improve investors' protection, namely, such as the requirement of SEC approval before a company can issue securities to the public, and the requirement for disclosure of information before a company can make a public offering. The act also made the penalty for insider trading more severe. Impacts of the Securites and Exchange Act on Financial Reform The promulgadon of the Securities and Exchange Act is expected to change the financial system. Direct financing for businesses seeking financial sources without having to go through commercial banks or finance companies will help boost the source of fimds for businesses while offering the public additional investment tools. It has set 144 Ekamol Kiwt up a framework for supervision and development of both primary and secondary securities markets. In addition, the act encourages financial innovation, which should lead to new types of businesses. A new dimension has been added to the Thai financial markets, and the securities business will benefit from more transparent and efficient supervision and legal framework. Finally the economy will be better served by more diversified and sophisticated means of financing. 8 PAKISTAN'S LIBERALIZATION OF THE EXTERNAL SECTOR Mohammad Ashraf Janjua, Deputy Governor, State Bank of Pidstan Though the genesis of the exchange and payments reforms can be traced back to 1972 when the Pakistan's rupee was massively devalued and import payments were hiberalized, more conscious and consistent efforts were made since 1980 when a tbree- year Extended Fund Facility (EFF) angement was agreed upon with the Intemational Monetary Fund (IMF). A central feature of the EFF program was a reform of the import system and a phased implementation of import liberalization. During 1980-81 a number of restrictions affecting imports were eased, including removal of the licensing ceiling on virtually all nonconsumer goods and the liberalization of vitually all raw material imports. A basic change was made in the fonrat of the import system by replacing the old system of publishing positive lists of permissible import ims with negative and restricted lists systems. The free list consisted of 435 items in 1979480. It was gradually enlarged to 575 during 1983-84, while the tied list, which comprised 23 items in 1979-80, was reduced to 20 in 1983-84. As a further measure of support to these efforts a flexicble exchange rate policy was introduced in January 1982 under which the value of the rupee is determined with reference to a basket of currencies of Pakdstan's major trading partners. As a major element of the macroeconomic adjustment and structural reform effort, a comprehensive medium-term program of trade liberalization and tariff reduction and rationaliztion was initiated in 1988. As part of the replacement program of most of the nontarf barriers into tariffs and in order to improve the transparency in the lists of banned and restricted goods, the government reclassified all categories of importable items that were subject to restrictions (the restricted list) and all categories of banned imports (the negative list). With respect to imports subject to restrictions, a number of additional categories were removed from the restricted list, thereby reducing the list to sixty-two categories. The 1991-92 Import Policy Order stipulated removal of further categories from the restricted list, with a plan to eventally eliminate the resticted list completely. The government also introduced a comprehensive reform and rationalizaton of the tariff system in 1988. The maximum ad valorem custom duty rate on all categories of imports was reduced from 225 percent in 1987-88 to 125 percent during 1988-89, except for alcoholic beverages and spirits and luxury vehicles. 145 146 Mohammad AshrqfiJaiua Reforms since February 1991 As part of the recent efforts to liberalize, deregulate, and privatize the economy, the external sector was opened up and a number of controls and restrictions were dismantled. In addition, beginning February 1991, the government introduced important reforms conceming foreign investment, the exchange and payments systems and foreign trade. They are as follows: * Allowing foreigners to invest freely in all industries, except certain specified industries, without pror approval of the govermnent; * Permitting foreign ownership of up to 100 percent of the equity of a business in Pakistan; e Allowing remittances overseas of dividends on shares, held by nonresidents, without prior permission from the State Bank of Paldstan (SBP); e Removing the requirement of government approval before shares held by foreigners can be transferred or foreign holdings of domestic capital can be remitted; * Allowing foreign nationals to invest in shares in existing companies through the stock exchange. Establishing a special convertible rupee account with an authorized dealer in Pakistan that can be opened by nonresidents to purchase shares quoted on the stock exchange with the balance in this account transferable outside Pakistan without prior approval of the state bank; * Liberalizing the rule for foreign controlled manufacturing companies (FMC) for working capital from domestic credit institutions without any limit and without prior approval of the state bank- * Obtaining foreign loans through industrial undertakdngs for seting up new industies or for expansion without any restrictions on interest rate and with no involvement in the front-end fees provided by the government repayment guarantee; * Abolishing the ceiling on payment of royalty technical assistnce fees; Pakistan ' Liberalzadon of die tuensal Sector 147 * Issuing National Investment Trust (NIT) units to Pakistan nationals residing outside Pakistan and foreign nationals residing in or out of Pakistan, on the basis of repatriation of capital and profits. To supplement Foreign Exchange Bearer Certificate (FEBC) scheme, doLlar bearer certificates (D)BCs) with a one-year mamrity were introduced in April 1991. These can be purchased by residents and nonresidents using foreign exchange. In addition, permission was granted for the opening of foreign currency accounts (FCAs) by residents of Pakistan on the same basis as nonresidents. Thus, balances held in FCAs will be freely transferable abroad, and there will be no limits on amounts of withdrawal. Other changes specific to foreign exchange are as follows: * Limits for financing certain invisible transactions were enhanced; * The requirement of declaration of notes, coins, and foreign exchange to customers' authorities in respect of outgoing and incoming passengers was dispensed with; * Permission was granted to Pakistan nationals and resident companies and fmns to r ork as money changers; * To protect domestic borrowers against the appreciation of foreign currencies, exchange risk cover was provided by the State Bank of Pakista so that their repayment liability is fixed in advance, irrespective of exchange rate fluctuations; * The limit of foreign exchange that resident Pakistanis could hold abroad was raised from US$500 (or the equivalent hereof in other currencies) to US$1,000; * The raising of foreign currency funds from abroad, by investment banks through the issue of certificates of investment having maturity of at least three months has been allowed; * PFive-year foreign currency bearer certificates denominated in U.S. dollars, deutsche marks, pound sterling, and Japanese yen have been issued, and these certificates, which can be cashed after a minimm holding period of two years, will bear fixed rates of profit payable half- yearly; 148 Mohammad Ashraf Jaena * Licensing requirements for items on the free list and other specified items were abolished; * The maximum import duty was reduced from 125 percent to 90 percent in 1991-92 and further to 80 percent in 1992-93; * And about forty-four items have been removed from the negative list in the import policies since 1990. At the end of 1990 there were 118 items or groups of items that were on the negative import list. Finally, foreign companies have been allowed to undertake export trade, and the public sector monopoly for the export of rice and cotton has been abolished. The Impact of the Reforms Although it is premature to assess the impact of these extensive exchange and payments reforms, it can be safely said that these reforms have started paying dividends. The following paragraphs summarize the impact of these reforms on the capital account. Foreign Investment The reforms introduced to attract foreign investment have shown positive results. Foreign investment was US$216 million during 1989-90 and increased to US$246 million during 1990-91, reaching US$335 million during 1991-92. During the first six months of 1992-93 direct investment stood at US$77.7 milion compared to US$90.6 million during the same period of 1991-92. Portfolio Investment The permission to overseas Pakistanis and foreign nationals to open convertible rupee accounts and to make repatriable investments in shares of exising listed companies with stock exchanges has attracted US$219 million during 1991-92. During July to December 1992, inflows from portfolio investments amounted to US$75.8 million as compared to US$97.9 million in the comparable period of 1991-92. Pakbin' s Liberalztion of the Ertenza Secor 149 Foreign Private Loans The permission to industrial undertakings to raise foreign loans without any restrictions on interest rates, front-end fees, and repayment periods has resulted in a considerable increase in the flow of foreign private loans. The disbursements were US$150 million during 1989-90, increased to US$158 million in 1990-91, and fiurther to US$559 million in 1991-92. Thus, the 1991 to 1992 disbursements showed an increase of 253.8 percent. The rising trend is continuing. During the first six months of 1992-93 disbursements amounted to US$420.2 million as compared to US$373.3 million during the same period of 1991-92. Government Securies The govremment's scheme regarding the issuance of one-year dollar bearer certificates and five-year foreign currency bearer certificates, introduced in 1921-1992, has attracted sums of US$53.1 million and US$62.7 million, respectively, as of December 1992. Foreign Exchange Bearer Certificates (FEBC) FEBCs, denominated in Pak rupee with three-year maturity, were first issued in August 1985. Tne outstanding amount of FEBCs, which at the end of June 1989 stood at US$280 million, rose to US$381 million at the end of June 1991. After decliing to US$338.4 million on June 30, 1992, they rose to US$400 million by the end of 1992. Foreign Currency Account (FCAs) The foreign currency accounts scheme, which was intoduced in January 1973, was initially meant for Pakisuni nationals residing abroad. The scope of the scheme was gradually widened. Permission to Pakistani residents to open and maintain these accounts and general permission for credit to these accounts with the proceeds of FEBCs, dollar bearer certificates, travelers checks, and currency notes was granted as part of the overall package of foreign exchange reforms announced in Febnruy 1991. As a result, the deposits under the scheme have gone up sharply: deposits averaged US$1,871.4 million during 1987-90, and started rising rather sharply since March 1991. From US$2,346 million at the end of March 1991, these deposits rose to US$3,867.5 million by the end of December 1992. The breakdown of these deposits in terms of residents and nonresidents indicated that residents deposits showed a steady 150 MohammadAslaf J&ai steep rise from US$32.8 million at the end of March 1991 to US$1,618.7 million by the end of December 1992, and nonresident deposits peaked at US$2,446.1 million at the end of October 1991 but started declining and stood at US$2,248.8 million at the end of December 1992. These deposits can be fiurther classified as deposits to banks and financial institutions and deposits to companies (non-financial institutions). The deposits to companies steadily increased from US$1,332 million in February 1991 to US$3,058 million in December 1992, an increase of 127 percent, while deposits to financial institutions declined from US$961 million in February 1991 to US$804 million in Janury 1993. The decline in these deposits mainly reflected the impact of the abolition of credit ceilings. The net increase in foreign currency account (FCA) deposits during the past four years has been steady, but the rise has been particularly sharp in 1991-92 when these deposits recorded an increase of $1,104.1 million compared to US$476.2 million in 1990-91 and US$266.3 million in 1989-90. The sharp increases in 1990-91 and 1991-92 mainly reflected increases in residents' deposits after the goverment granted prmission to resident Pakistanis to open and maintain FCAs on the same basis as nonresidents. The breakdown of the overall increases in terms of residents and nonresidents indicated that while the deposits of the former went up sharply by US$157.6 million during April-June 1991 and US$1,164.7 milLion during 1991-92, those of the latter increased only by US$88.6 million in April-Tune 1991 but fell by US$60.6 million during 1991-92. While the deposits of the residents continued rising during July-December, 1992 and stood at US$1,618.7 million and $263.6 aillion at the end of December 1992, nonresident deposits recorded a furither fall of US$92A million over the same period. The sharp increase in resident deposits could be attributed to diversion of part of the home remitances, sales, and encashment of FEBCs; under or overinvoicing of exports and imports; and a division of the differential between the maximum permissible and acuala commission paid to importers by Paldstani exporters. Table 8.1 shows how each of the above factors is estimated to have accounted for the sharp increases in resident deposits. Table 8x1. Factors Affecting the Rise in Resident Deposits (USSnion) Faaor 990-91 1991-92 JuLyDec. 1992 Home remittances 94.1 474.9 166.0 FEBCs 151.0 485.1 - Commissions - 109.0 42.0 Total 245.1 1.069.0 208.0 Pakisanr's Liberalization of the Enerma Secar 151 For the purpose of balance of payments, the way part of the FCA deposits pertining to residents have been evaluated has been recently changed. Formerly, these deposits constituted liability to nonresidents and formed a component of the capital account. Now these deposits are treated as a part of unrequited transfers in the current account. This experience of capital account liberalization is a recent one for Pakistan. The deliberations of this seminar on the experiences of other countries in Asia and Latin America will be helpful in formulating policies and management of extemal balances in the forthcoming years. 9 FINANCIAL LIBERALIZATION IN AN AGRARLIN ECONOMY: THE CASE OF PARAGUAY Reinaldo Penner, Central Bank of Paraguay The banking sector in Paraguay developed with the establishment of internaonal bank branches mainly during the 1960s and 1970s (table 9.1). During 1980-92, the sector experienced high growth when several local banks were established, even though some branches closed. An open system of free entry and exit was the basis of the expansion of modern banking. Local banks strd with an acceptable performance in this competitive environment Their entrance into the market usually began through a process called "spin-off," when local managers of foreign branches initiated their own maagement of a local bank. In addition, these local managers would buy closed branches and, in some cases, develop a joint venture with the former owners of the closing branches. As a result, the total assets and labilities of the banking system were more or less equally disibuted among local banks and foreign branches. By this time the economy changed from a small-peasant economy into an ago- export economy and income per capita doubled. But external conditions, in the form of price swings and climatological change, increased. Along with the stimulation of a primary export-oriented economy based on cotton, soybeans, and meat, the government fcllowed an expansionary policy through external as well as internal indebtedns on behalf of an investment program on public enterprises. The program, however, was highly inefficient and generated serious monetary instability. This policy produced recurrent recessions during 1982-83 and again in 1986. The situadon led to the resignation of the govermnent in 1989, and the curremnt administration took power. The Limited Growth of the Financial Sector from 1980 to 1988 Except for temporary slowdowns in profitability and a rise in the level of nonperforming loans, the economic recessions of 1982-83 and 1986 apparently had no lasting effect on the financial system. For instance, in 198243 the nonperformin loans of the banks grew from 6.1 percent to 14.9 percent (table 9.2). 153 154 Reinaldo Penner The recession increased the risk of default and led to many reschedulings of agricultural credit. Interventions by the supervisory authorities of the Central Bank of Paraguay, however, were limited to two banks, each with limited losses during 1980- 88. One important feature was the -unfavorable anticipation of the local currency devaluation by the banks in 198243. After more than two decades of fixed exchange rates (figure 9.1), the real exchange rate dropped by 140 percent in the first half of the 1980s, causing high losses to the banks. Since 1986, however, the rate has been quite stable, and by the end of the decade banks again assumed external liabilities. Since 1973, banks had been allowed to accept dollar deposits from the public and the deposit rates were freely negotiated between banks and clients, though these rates were usually lower than the international rates, and the banks channeled the fuids abroad to earn higher interest rates. The branches were allowed to use resources from parent banks abroad to grant dollar denominated loans to well-established local exportng fwms. These policies on capital account liberalization lhelped Paraguay avoid a financial crisis similar to the Chilean crisis in 1983, when Chile's banks were unable to afford fteir exchange rates and credit risks on dollar liabiiities. The substantial expansion of the financial sector during the 1980s was not quite consistent with the level of fiancial intermediation and savings. For instance, from 1980 to 1988 eight banks were incorporated in to the system, an increase of almost 60 percent, while total banking Liabilities and assets decreased in real terms (table 9.3). Indeed, the lower growth rate of deposits and loans increased the overhead expenses of the system because of the larger numnber of banks. Conseauently, this pushed the spread to higher levels, although the official interest ctilings prevented this to a certain degree. The average number of bank employees in 1991 was 151, whereas the average stock of available domestic resources per bank was US$29 miion. This was hardly enough to generate a turnover to pay salaries and bills. The financial intermediation ratios, which already were extremely low by intenational standards, decreased sharply during the second half of the 1980s. This resulted in a drop of all monetary aggregates. As a ratio of GNP, Ml decreased from 8.2 percent in 1985 to 7.9 percent in 1988; M2 from 15.4 percent to 12.7 percent; and M3 from 18.9 percent to 14.5 percent. The quasi-montary saving dropped particularly sharply from 7.2 percent to 4.8 percent for local currency and from 3A percent to 1.8 percent for foreign currency (table 9.4). The disintemediation process clearly indicated the growing importance of the informal financial sector. Increasingly banks operated through related nonbankdng institutions where reserve requirements were lower and unregistered transactions were easier to perform because of relaxed supervision. The state controlled National Development Bank (BNF) decreased its participation during the 1980s as did commercial banks, while savings and loan associations (SAPV) and finance compames grew. BNF loans fell from 22.6 percent of total domestic resource mobilization in 1981 to 20.9 percent in 1988. BNF's domestic resource mobilization, however, increased from 7.8 percent to 10.2 percent in the same Famda Liberazadion in Agrana Economy: The Case of Paraguay 15 period (table 9.5). Commercial bank credits were rduced from 70.3 prcent to 60.4 percet of the total domestic resource mobiizaion, although domestic resource mobilization for domesdc banls grew from 68.7 to 71.8 percent. Both the BNF and the commercial banks experienced a reduction in credits and resources during 1980-88 compared to credit and resource level in 1975. Conversely, SAPV loans grew from 8.1 percent to 10.9 percent of the total domestic resource mobilization, although the level of resource mobilimtion decreased. The finance companies also expanded their credit from 0.5 percent to 5.9 percent during 1980-88. Many local banlks and some branches are related to the expanding segmerts of nonbank financial intermediaries (NBFIs), such as finance companies, SAPVs, credit cooperatives, warehouses, insuce companies, and investment banks. Between 1979 and 1988, nineteen fince companies came into existence, each operating with an average of ten employees (table 9.6). The main reason for the increase in intermediation by the NBFIs was the high legal reserere ents for banks, set at 42 percen for sight deposits, 30 percent for time deposits, and 15 percent for dollar denominated deposits. Meanwhile the SAPVs had a reserve requirement of only 5 percent and the finanial companies only 10 percent. In addition to the reserve r e , a stamp tax on financial trnsactions increased the disintermedion process. The tax pushed retail banking and personal loans out of the formal sector. The additional irem of a minimm asset of 50 pere invested in agriculture, industry, or export financing, and the ceiling imposed on interest rates, played at the same time an important role in this process. A thid element was the expansionary rediscount policy of the central bank at subsidized rates. The local banks are not only financially connected to the NBFIs, but they are also tied to large corporations in the commodity or service sectors constituting a broad economic group. Such an inerlocking economic unit is pardally self-funding but is aiso funded by banks and the NBFIs. During the economic recessions, banks had to take over many bankrupt firms that were then merged with the groups. A special case of finacially related firms exists in agriculture. The ago rting fims and trading firms buying up and transporting crops to silos and warehouses need large amounts of credit to pre-finance their farmers. In many cases the cororations operate as near- banks when they select their clients, although unie banks they do not assume the total risk. The nsk is partially passed on to the NBFIs, commercial banks, or the state controlled banks who, in tur, transfer part of the risks to the central bank through the rediscount facility. The state controlled BNF, the National Workers Banks (BNT), and some eligible commercial banks offered loans for agricultural actvities by rediscounting these at the central bank widL subsidized intrest rates (figure 9.3). Until the beginning of the 1980s, the BNF had access to abundant external loans from multlatral ogaizations. But after the ietional finacial crisis of 1982 the central bank provided fimds equivalent to almost 80 percent of the loans channeled by the BNF for 156 Rcine Pewter commercial crops (cotton, soybeans, wheat, and sugarcane). While the BNF financed planting, the commercial bank took a more active role in the financing of the harvest and pre-export loans. Overall, the commercial banks (including the state controUed BNT) financed approximately 75 percent of commercial agriculture and the BNF financed the remaining 25 percent. Commercial banks also increased their share of rediscount operations, from 11.5 percent in 1980 to 51.5 percent in 1988. Under these circumstances the banks did not need to augment their iner tion, but they increasingly relied on the rediscount facilites offered by the central bank and channeled them to agro-exportng firms. The mobilizaion of domestic resources were transferred to the NBFIs, which offer, in many instances, higher and positive real interest rates. Monetary policy was inadequate to manage the distoions created within this type of financial system. The control of the monetary expansion of all the financial institutions was difficult, not only because of the rediscount facility of the central bank, but also the central bank financed an ambitious public sector investments program that raised the public debt over US$400 million during the second half of the 1980s. To cope with the monetay expansion, the central bank increased the reserve qi to 42 percent in some cases. This, however, sdmulated disintermediation and growth of unrgulated and informal instituons, in some cases run by formal financial intemediaries. Based on a study performed in the 1980s the system as a whole seemed financialy sound on the surface because the bulk of the commercial banks appeared profLtable and adequately capitalized. Behind this apparent success, however, the limits of financial expansion appeared to be clear. Several factors caused such limits: diminiShing financial indiaton, a concentated market, inflexible regulations, inefficient supervision, high and deentiated reserve requirements, and indirect taxes on financial transactions. The result was overspecized institutions for the size of the market and fradgmtion in most categories, while a good number of unregulated informed intitutions were acually rn by financial instutions in the formal sector. The Policy of Fnancial Reform, 1989-92 The unstable macroeconoc situation caused by mismanagement and inapprriate poLicies during the 1980s ended in an intal coUapse at the end of 1989, leaving the central bank without international reses. In February 1989, the new government initated an general economic liberalization accompanied by financial liberalizaion, supported by the International Monetary Fund and the World Bank The reform process has been synthesized in four areas: (a) extrnal liberaliation involving foreign exchange, current and capital account; (b) internal libealizaion conceming interest rate and credit allocation; (c) monetay maragement; and (d) fianial supervision. Finncial Liberalization in an Agranai Eonom: The Case of Paraguy 157 External Liberalization The first measure of financial reform was to abolish the multiple exchange rates and establish a free-floating exchange rate. Since then all operations of the private sector have been channeled directly to the foreign currency market where banks and exchange houses can operate freely. But as long as the public sector's transactions in foreign currencies, including those of the nine public enterprises, were still performed by the foreign exchange desk of the central bank, although at the prevailing market rate, the liberalization did not fully release official control on capital account- transactions. The active intervention by the foreign exchange desk of the cental bank on both the demand and supply sides resulted in a very powerful instrument of monetary and exchange rate regulation. Oter related measures on current and capital account liberalization soon followed. In July 1989 the external tariffs were fixed at a weighted average of 8 percent.1 The reserve requirements on dollar deposits were lowered and unified at 15 percent, but it was increased in two steps until it reached 30 percent by June 1992. This was done to unify the requirements between local and foreign currency deposits and to reduce the huge capital inflows durmg the period. In October 1990 all banks, not just the branch banks, were allowed to mobilize external funds and provide foreign currency loans to exporting firms. By the end of 1992, Congress audtrized banks to grant such loans for import-substitution activities by investing their domestically funded foreign currency liabilities. Internal Liberalization The second part of the reform process consisted of a series of measures during July 1989 and October 1990 to improve financial intrmediation. First, all interest rates on deposits, except the rediscount operation of te central bank, were ihberalized. Second, all ceilings on credit interest rates and the minimum portfolio requirements, except for the newly created Certificates of Deposit (CDs), were also discontinued. In July 1989 the central bank authorized the banks to issue CDs at market rates. The CDs were tax exempt and subject to a reserve requirement of only 10 percent The funds, however, were chaneled to the agro-exporting sector. Since January 1990, the financial companies were also authorized to operate in the agro-exporting market. The CDs had to have a matrity of 180 and 360 days, like the rediscount opemons of the central bank, which they were supposed to replace. In October 1990, the central bank relaxed the limitation on investing CDs funds in agriclture, but the rediscount rate was raised to 30 percent, which was the 1 . In July 1992 ariffs were considerably simplified. 158 Reinatdo Penner prevailing average interest rate for CDs. This rate was lowered in 1991 when inflation rate decreased to the current level of about 18 percent. Monetary Management The third part of financial reform concerned the monetary policy on which the central bank from the very beginning received assistance from the International Monetary Fund. Basically, the assistance consisted of the development of financial programming and management of the money and foreign exchange departments. In addition, other important organizational and administrative areas of the central bank also received such assistance. Since October 1990 the cental bank followed a policy to achieve a weeldy target of the monetary base through open market interventions and reserve requirements, and to better -ontrol the rediscount operation. A policy of budget control and a tax reform in 1992 reduced public sector loans to a minimum resulting i a fiscal surlus in 1990, 1991, and 1992. In November 1990 the bank introduced issued Letas de Regulacion Monetaria (LRM) to perform open market interventions. The LRMs are exempt from all taxes, have a flexible maturity (average thrty days), and are sold at weeldy auctons at an average yearly yield of 18 percent. In January 1992, the money and the foreign exchange desks were unified to form the Open Market Interventions Department to better coordinate the activities of both. The objective was to equalize the reserve requirements on foreign currency deposits with requirements on domestic currency deposits, by, first, slowly reducing the reserve requirements on domestic deposits to acceptable levels and gradually replacing them with LRMs. In October 1990 the reserve requirements on domestic currncy deposits were reduced from 42 percent to 37 percent. Concomitantly, the reqirements on CDs were raised from 10 percent to 15 percent, and on the finance company deposits from 5 percent to 10 percent. In July 1991 the requirement of 37 percent was reduced to 30 percent, equaling the requirements on foreign currency deposits in June 1992. Since July 1992, the central bank began to remunerate 50 percent of the reserves at a 10 percent interest rate. In June 1992 the investment and mutial funds were also subject to reserve requirements of 30 percent, to be complied with gradually. The policy on rediscount operations was to reduce the operations to a minim level and replace them with bank mobilization of funds tbrough CDs. Further elements of the policy on rediscount operations were to equalize its interest rate to those-of CDs and to equalize it with the average yield of the LRMs. Financial Liberalwion in an Agrarian Econony: The Case of Paraguay 159 Financial Supervision The fourth part of the reforms concerned the improvement of the supervision system of the financial system. For this purpose the central bank first increased the minimum capital requirements for banks and finance companies to five times the nitial level during October 1989 and April 1991. In January 1991 the government signed an agreement with the World Bank for a technical assistance program with the following objectives: (a) modernizing and strengthening the supervisory capacity of the central bank; (b) replacing the Centml Banling Law and the Banking Law, dated 1941 and 1952, respecively (this was also a main objective of the assistance program approved by the IMP); and (c) developing the capital market, setting up a stock exchange, and increasing the banking capacity to be able to grant long-term loans to the real sector. The supervisory program consisted of two steps: (a) developing a centralized database for preventive control and risk evaluation; and (b) developing regulons on asset classification based on the international standards on risk evaluation and capital provision. This was designed to improve the solvency and transparency of banks and finance companies. For the first time banks had to evaluate their asset risks at the closing balance sheet of 1992 and afterward. This was subject to capital proviion by June 1993. But the Centrl Banking Law and the Banlkng Law were prepared and revised many times, and no version has reached Congress. The critical elements of these laws are the autonomy and the supervisory power of the central banks. But concern has repeatedly caused postponement of the approval of these laws. The Capital Market Law had less opposition and was approved in 1991, creating the National Stock Exchange Commission, which is engaged in developing stock exchange promotion. The Results of Liberalization The economic reform and liberlization policy undertaken after 1989 has stabilized inflation, which declined from 44.1 percent in 1990 to 17.8 percent in 1992. The fiscal surplus was 3.4 percent in 1990, 1.5 percent in 1991, and 0.2 percent in 1992, which supported a modest increase in domestic savings and helped in .the achievement of monetary targets. Notwithstanding these positive results, the economy entered into a recession in 1991 resulting in lower cconomic growth rate of 2.5 percent in 1991 and 1.7 percent in 1992 together with an increasing rate of unemployment 160 Reido PenCEr Issues of Stability The fragility of the monoculture economy took its toll on stability. This is similar to what happened in the 1980s. The policy of stabilization and adjustment that followed after 1989 did not overcome the negative economic effects of the agricultural crisis, caused by a fall of yield per acre because of climatic conditions. There was a 45 percent price drop in soybeans and a 50 percent price drop in cotton during the 1989- 92 period (figure 9.2). The unstable prices and the unfavorabie climatic conditions led to a credit rationing in the agricultural sector, which, in turn, hampered -its diversification. Consequently, financial reform seemed to become a paradoxical choice: on the one hand the financial system needed to be liberaized and modernized to increase domestic saving and to mobilize international finance, but on the other hand the experience so far showed that liberalization would not automatically ensure better levels of financial intermediation and financial deepening in an economic climate cbaracterized by stagnation. On the contrary, the central bank had to expand its old style rediscount during 1992 to prevent a total collapse in cotton and soybean production. Thus, an unstable agrarian economy is not the best climate to liberalize because it demands a policy of diversification to dismantle the stmcture of monocuure, which, in turn, demands financial assistance. Many analysts questioned whether the policy of financial liberalization should have gone faster to cope with some of the bottlenecks, or slower to ensure the necessary adjustments in other areas, such as economic diversification This remains a controversial matter. Domestic Savngs and Intermediation Gross domestic savings increased from 20 percent of GNP to 25 percent in the 1988-92 period. The change in the national accounting methods since 1991, however, makes these data somewhat unreliable. Notwistnding, a modest increase in saving may be plausible. Banldng intermediation, especially in foreign currency, increased as well, although some illusory effects exist in the ratio of money to GNP to the extent that a rise may reflect less of a rise in deposit and more of GNP stagnation. The rise in banking intermediation has occurred at the same time as a shift of deposits from NBFIs to banking institutions. This shift is explained by two elements: the reduction of the reserve requirements on bank deposits and the increase of those on NBFI deposits; and the virtual elimination of the transaction tax, which makes retail banking attractive again for banks. -Because of a lack of data on broad monetary aggregates, which should include both bank and nonbank deposits, the analysis of financial intermediation centers on three monetary aggregates, Ml, M2, and M3, as shown in table 9.4. Ml, which includes the current account, increased from 7.9 percent in 1988 to 8.7 in 1992; M2, FinancialLiberatization in anAgrarian Economy: The Case of Paraguay 161 which includes all banking savings deposits in local currency, increased from 12 percent in 1988 to 14.4 percent in 1992. In both cases the ratios matched the levels in 1985, which show slow progress in intermediation, that is already low compared to siniar economies, especially concerning savings in local currency. M3, which also includes banking savings in foreign currency, increased from 14.5 percent to 22 percent between 1988 and 1992 because of the increase in foreign currency deposits from 1.8 percent to 7.6 percent (table 9.4). The slight increase of banking intermediation during 1989-92 is the result of a shift in financial intermediation away from the BNF, the SAPV, and the finae companies to the banking system. The connnercial banks increased their share of total domestic resource mobilization from 71.8 percent in 1988 to 84.7 percent in 1992, whereas the BNF share decreased from 10.2 percent to 5.1 percent, the SAPV share from 14.4 percent to 8.0 percent, and the finance companies' share from 3.6 percent to 2.0 percent over the same period. The Credit Expansion Following the increase in the level of financial intermediation, the banks increased loans to the private sector. from 11 percent of GNP in 1988 to 17.2 percent in 1992. This is above the level reached in 1985 (table 9.4). This reflects not only the inreased banlkng intermediation but also the adjustment on public sector expenses, which reversed the crowding-out effect experienced by the Paraguayan economy during most of the 1980s. As a result the bankdng system expanded steadily, incorporating six new banks (one branch and five local banks) between 1989 and 1992 (table 9.1). The fiancial companies expanded as well with some twenty new mstimtions during 1989- 92. Many of these finance companies were operating illegally and requested official authorization to conduct their business. The stiking result of the growth due to liberalization is the change in credit flows away from development loans for primary and manufacturing activities to consumer loans, such as personal loans, retailing activities, and housing. The share of development loans decreased from 32 percent of total loans in 1988 to 22 percent in 1992, whereas the share of consumer loans increased from 49 percent of total flow of loans in 1988 to 63 percent in 1992 (table 9.7). The declining trend in the flow of funds to the real sector during 1989-92 was especially strong in the core activities of the economy, namely soybean and cotton cultivation. An important reason for the increase of loans channeled to retailing acdvities and housing could be the overvaluation of local currency due to an inflow of short-term capital to the economy. This capital inflow reduced relative prices of nontadables (construction, services, and imported goods) against those of tradables (exports and import substimtion goods). 162 Reinalto Penner The decrease of loans to the core activities during 1991-92 is similar to what happened during the economic recessions in 1982-83 and 1987. Thus, the situation will probably settle when the agriculture recovers. A review of the evolution of interest rates, however, shows that the reduction in real terms is somehow contradictory to the reduction of credits channeled to core activities. The increase of loans for exports was either because of the agrarian crisis, or because international banks refused to finance Paraguayan exports while the accumulation arrears in the external debt was so great. (The external debt was not serviced untl the end of 1992.) Another reason for the strong rise in export financing was primarily because of the policy of external liberalization. Nominal and Real Interest Rates Nominal interest rates sharply increased in 1990, the year all ceilings on rates were dismantled. This mainly reflected the rise of inflation rates from 28 percent in 1989 to 44 percent in 1990 (table 9.8). After the fall of the inflation rate to 12 percent in 1991, the interest rates decreased as well, but for the first time stabilized at positive real levels. The difference between commercial, housing, and development interest rates is still notable, reflecting the segmentation of the market. Real interest rates of the commercial banks were around 15 percent in 1991, and those of the finamce companies were about 28 percent. Development rates were slightly lower, because of the rediscount facilities of the central bank. Deposit rates in 1991 were also positive for the first time, especially for CDs, which yielded 5 percent in real terms. During 1992 the nominal commercial rates increased from 27 percent to 30 percent. Because the inflation rate increased from 12 percent in 199i to 18 percent in 1992, the real commercial rates decreased by about 3 points. The nominal CDs rates increased from 18 percent to 20.6 percent, and their real interest rates also remained at positive levels. Although most nominal rates were reduced in 1991, and real rates were reduced in 1992, the level of real rates is still above 10 percent. This is high when compared to international standards. Notwithstanding the measures regarding external liberalization and free market entry, the internal rates did not show a clear tendency toward equaliztion with international interest rates until now. The expectation is that the regularization of the arrears of external debt will reduce the country risk, and internal and external real interest rates will begin to equalize. The Limits of the Reform The reduction of the flow of funds to core activities has reduced the political support for continuig policies of financial reform, designed to allow banks to mobilize funds through CDs and to channel these to core activities, such as cultivating and Financial Liberalization in an Agrarian Economy: lhi Case of Paraguay 163 harvesting soybeans and cotton. This, in turn, should have allowed the central bank to withdraw gradually from its involvement in development financing by adjusting its rediscount rate to positive real levels. The po0cy of steadily augmenting the rediscount rate, however, prevented banks from maing use of the rediscount facility. Thus, total commercial banking loans subject to rediscount (soybean, cotton, wheat, sugarcane) fell from 40.5 percent in 1989 to only 4 percent in 1992 (table 9.9). As a result, rediscount loans from the central bank to agricultural activities have declined steadily as a percrentage of total rediscountable funds from 48 percent in 1989 to 21 percent in 1990. The banks, instead of mobilizing local currency finds through CDs, had more success in attracting foreign currency deposits and channeling them offshore or toward consumptive activities. The real flow of fimds from commercial banks to the plantations declined from 1989 onward, because the banks invested their assets in less risky activities, such as export financing, which in 1992 received twice the funds it received in 1989 (table 9.9) Under these circumstances the central bank has maintained its involvement in rediscount operations. The real flow of funds from the state controlled BNF to the core activities remained stable during these years and were financed mostly with rediscount loans from the central bank as in the 1980s. Rediscount operations resumed in 1992 when the government decided to expand these operations because of the bank's unwiingness to take over the credit market for commercial farming activities. Apparently the central bank was not prepared to assume the risks. Or another explanation could be credit aversion ky the banks when clients reach a certain upper level of risk. External liberalization allowed banks to increase their foreign currency loans to local firms, which increased their participation in total commercial bank loans to the private sector from 6.0 percent in 1989 to 18.6 percent in 1992 (table 9.10). Foreign currency flows augmented the total flow of funds from commercial banks to the core activities from 30 to 35 percent for cotton and 40 to 50 percent for soybeans in 1988- 90, and to 50 percent for cotton and 60 percent for soybeans in 1991-92. These funds, however, are limited to the segments that earn foreign exchanges; thus, they do not really reach the farmers. This means that foreign currency loans are used mainly for the harvest season and less for the cultivation season. - To conclude, the rediscount operations are still of fimdamental importance but do not support objectives of financial reform to develop a competitive financial market that would be able to intermediate between domestic savings and investment, especially in core activities. The major resources generated by liberalization through increased domestic saving, however, did not flow to core activities of the economy, especially to farming activities. Rather, the flow was directed toward consumer financing. Therefore, the financial system still relies on central bank rediscount operations that are channeled basically to the BNF at negative real interest rates. The HNF in trn 164 Reinuddo Penner channels them to farmers or agro-exporting firms at slightly positive real rates of 4 percent. The agro-exporting firms pass the funds fiurher down to the farmers or to commodity intermedaries (firms buying up and transporting cr I., t real interest rates of about 17 percent. Finally, many farmers get the loans from the intermediaries at real rates that reach as high as 57 percent (figure 9.3). This system has large overhead costs for the financial intermediaries, is quite expensive for the farmers, and lacks the necessary transparency to be competitive. The need to reform this system is not only important for the intermediation process but also has significant importance for the central bank, which is engaged in monetry targeting. Instead of being the fist lender of the economy, it should assume the role of the lender of the last resort, but as long as domestic savings do not flow to the needy sectors, the current system must reman operational. The limits of the reform financial system have not yet resulted in a slowdown of credit expansion, nor are there open financial tensions at present that could lead to crisis. The reform process, however, has stagnated, and a decision on central issues is needed to redesign a system that is appropriate for an unstable agrarian economy at an immature stage of economic diversification. Decisions have not yet been made to begin sulch a process. To a certain extent it may be concluded that financing economic diversification is a precondition for stabilizing the economy and developing a system based on a liberalized financial system. Fimacal Liberalizaon in an Agraian Econoaxy: Te Case of Pagway 165 Figure 9.1. GDP, Inflaion, 1981-92 GDP (ercentage variaion) < ' \ ~~~~~~~I ' 20 -~ 19B1 1982 1983 1984 I985 1986 1987 1988 1989 1990 1991 1992 -X- Total GDP -0-AgricueM GDP| Inflation (December) so 40 A '£ -\ 30 20 /Z 01 X "-- 1981 1982 L983 1984 1985 1986 1987 1988 1989 1990 1991 1992 -X- Idiadon -U- Real EzW Ra )-4- - / 1. / ('4 I / - 0 >4' N N s4 0 'N a 00 -, >( I 'S. / N I NI 0ft N S .4 / a 4. 'N ..  I 4 U - __ __ .4 __ - C' S C C' I Finanial Liberalization in an Agranan Economy: Thie Case of Paraguay 167 Fwre 9.3. The Process of Financial Intermediation in Agriculture Real Interes rames Domestic Savring Bank 19S9-90 /1\9 / External redismunt irms /b/u/Saving p loam ~~~m md tanspional g Coaercial Bank tco rting p mDevelopent Bank /e / 4 / 8hans / & Pera Bants Meraants - i / / ~~~~~F'nms buying up boo S p Fnmns buying up X /~~~~~~m /tnpof crp and tr ansprigcrops / / / ~~~~~Famies {E) Farmers I~~~~~~~~Pat IP Peasant 168 Reinaldo Penner Table 9.1. The Paraguayan Banking System, 1992 Countq Year Brncds Parentm ba of onig fomndd Branchs Eimloyees Foreign banks Citibak NA. Citicorp EE.EU.U. 1958 7 110 Ascin S.A. Bco. Central S.A. Spain 1964 13 164 lloyds Bunk PLC LLoyds Bank U. Kngdom 1920 8 128 Intbawco SA. Bco. Nacional S.A. Brazil 1978 3 87 Holmnis Unido Algemiee B. Neder L Netierlands 1965 3 98 Do Brasil SA. Beo. do Brasil Brazil 1941 1 92 Real del Par. SA. Beo. Real Brazil 1974 2 72 Sudamnri SCA Banque Sudameris Framcc 1961 5 113 Naci6n Argentina Bco. Nac Argenti Argentina 1942 3 93 Enorior SA. Bco. Exter. de Espafia Spain 1961 2 66 iParmai SA Bca. do EsL de Paxani Brazil 19S0 5 158 BANESPA SA. Bco. do ESL de S. Paulo Brazi 1977 1 52 LN.G. Bank I.N.G. Bank Nerbedands 1992 1 22 Subtotl 54 1,255 Lcal prive banks Uni6n SA. Paraguay 1978 21 270 General SBA Paraguay 1987 11 157 Continental BAJ Paraguay 1980 10 119 Corporaion SA. Paraguay 1987 3 101 Invorsiones SA. Paraguay 1984 11 242 DesalNo SCA. Pargay 1971 2 75 Fimmirica S.A. Paaguay 1988 7 98 Alein Paraguayo SA. Paraguay 1989 2 67 Bapr S.A. Paragy 1981 12 157 Paaguayo Orienal SA. Paraguay 1988 3 104 Unifa SBA de Iv. y Fom. Paraguay 1990 5 160 Regional SA. de lov. y Foa Parguay 1991 4 45 Bawosur SA. de bar. y Fom. Paraguay 1992 1 39 Subtal 92 1.634 Pbic bank National Dvedopment Bank Paraguay 1961 47 935 National Workers Bank Parauay 1975 12 180 Subot 59 1.115 Source: Central Bank of Paraguay, Supeintendency of Banks. Financial ibemraition in an Agrarin Economy: The Case of Parguay .169 Table 9.2. Nonperforming Portfolio of Commurcil Banks, 1981-92 Czegory 1981 1983 1985 1987 1988 1989 1990 1991 1992 BMions ofgnaranes NonperFomiing loans 6.6 19.2 18.2 12.9 11.4 17.3 13.6 24.0 41.3 Total portfolio 107.5 129.0 165A 254.7 325.5 608.8 5255 853.2 1,198.4 Loan loss provision 1.0 2.3 2.8 3.0 2.5 4.3 5.0 8.2 11.8 Percentages Nonperfomninghlotal 6.1 14.9 11.0 5.1 3.5 2.8 2.6 2.8 3.5 Loss provlnonpeiform. 15.2 122 15.6 23.2 21.6 25.1 36.5 34.1 28.7 Souce: Central Bank of Paraguay and World Bank estumates. 170 Reitaido Penner Table 9.3. Consolidated Balance Sheet of the Barking System, 1981-89 (billions of guanes. 1980 constant pnces) Asseu and iabiies 1981 1985 1987 1988 1989 1990 1991 ASSE7S Liquid asse 13.3 10.2 8.3 11.4 40.1 35.2 30.6 Reserv.W/ccnral bank 45.4 3S.1 50.4 45.4 44.7 37.7 38.1 Net oans 96.6 79.0 82.1 83.8 94.2 94.1 94.5 a) Loam 98.9 81.7 84.4 85.3 96.0 95.7 95.4 (i) In operation 89.2 69.5 78.3 81.3 90.9 90.4 92.7 (ii) Nonperfonning 9.6 12.2 6.0 4.0 5.0 5.2 2.7 b) Minus: loan loss provision (2.2) (2.7) (2.2) (1.4) (1.7) (1.5) (0.9) Building and cquipmnmt 2.6 1.7 936 2.9 2.7 3.6 5.2 Other assets 11.7 15.2 10.0 7.3 9.1 11.6 9.2 TOTAL ASE 169.7 144.3 151.9 151.0 191.0 182.5 177.7 Deposit 101.8 84.2 94.0 89.1 121.5 118.7 135.4 Borrowed funds 25.3 29.8 27.8 31.3 32.5 27-4 12.6 Odher liabIlities 13.2 11.2 10.4 9.4 11.6 9.8 6.2 Total liabilities 140.4 125.3 132.3 129.9 165.6 I56.1 154.3 Capital and resermes 24.3 20.1 19.0 18.5 23.1 24.2 20.2 Profit and loss 2.6 77.9 1.5 2.1 1.3 2.5 2.6 Net worth 29.3 19.0 19.5 21.0 25.3 26.4 23.4 TOTAL LBUMES 169.7 144.3 151.9 151.0 191.0 182.5 177.7 Source: Central Bank of Paraguay, Superintendency of Banks. Fnanaal Liberaizaon in an Agrarwin Economy: The Case of Paragnay Ifl Table 9.4. Indicators of Financial Intermediation Banking System Liabilities and Assets (percent of GDP) Monetary aggregates 1985 1987 1988 1989 1990 1991 1992* M 1 8.2 8.5 7.9 8.1 7.6 7.5 8.7 M 2 15.4 14.0 12.7 12.9 12.1 12.8 14.4 M 3 18.9 16.8 14.5 17.2 17.5 18.5 2Z.0 Financial saving 10.6 8.3 6.6 9.0 9.9 10.9 13.7 LOcal 7.2 5.5 4.8 4.7 4.5 5.2 6.1 Foreign 3.4 2.8 1.8 4.3 5.4 5.7 7.6 Bank credit to the private sector 13.2 12-0 11.0 10.8 12.1 15.3 17.2 * Prelimninay. source: Central Bank of Paraguay. 172 Reinaldo Penner Table 9.5. Size of Financial Intermediaries, Selected Years (Percenrage of total domestic resource mobiization) Financial itermedaries 1975 1981 1985 1988 1989 1990 1991 1992" National development bank Liquid asse 8.7 5.4 5.4 10.2 5.6 5.5 3.9 4.8 Credit 52.9 22.6 22.7 20.9 21.8 22.1 14A6 11.5 Domestic resource mobilization 14.8 7.8 7.6 10.2 6.3 7A 4.2 5.1 Other net 46.8 20.2 20.5 20.9 21.1 20.3 14.3 11.2 Commercial banks Liquid aSSetS 432 35.4 45.6 36.5 36.8 30.7 42.9 46.1 Credit 70.3 72.3 57.6 60A 68.0 79.2 60.5 62.3 Domestic resource mobilization 73.8 68.7 71.9 71.8 75.6 7617 85.3 84.7 Other net 39.6 39.0 31.2 25.1 29.2 33.3 18.1 23.7 Savings and loan association Liquid aSSetS 3.3 4.6 2.4 22 22 1.9 2.9 2.0 Credit 8.1 16.1 11.3 10.9 11-0 11.1 5.9 55 Domestic resource mobilization 10.8 21.1 16.9 14A 14.0 13.8 92 8.0 Ocher net 0.6 -0.4 -3.2 -12 0.8 -0.8 -0.6 -0.6 Private development banks Liquid assets 0.2 0.0 0.4 0.7 0.6 0.8 0.6 0.0 Credit 3.6 0.8 0.8 1.1 1.1 1.8 1.0 1.1 Domestic resource mobilization 0.3 0.0 0.0 0.1 0.3 0.6 0.0 0.0 Other net 3.6 0.8 12 1.7 1.4 2.1 1.3 1A Fmance companies liquidassets 0.0 0.4 0.4 05 0.6 1.1 1.0 0.0 Credit 0.5 7.6 6.8 5.9 5.6 6.2 4.7 5.4 Domestic resouroe mobilization 0.3 2.4 3.6 3.6 3.8 1.6 1.0 2.0 Other net 0.2 5.6 3.6 2.8 2.4 6.3 4.6 3.8 Toti Liquid asse 55.4 45.7 54.1 50.0 451 40.1 51. 53.8 Credit 133.4 119.5 99.1 99.3 107.4 121.1 86.6 85.8 Domestic resource mobilization 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Other net 90.8 65.2 53.2 49.3 53.2 61.2 37.7 39.6 *Prl Source: Central Bank of Paraguay. Fnancial Libalizton in an Agarian Economy: The Case of Pwraguay 173 Table 9.6. Swmmay Information on Finance Companies, 1992 Frnace Operion TOWl Tooal Exemal Lon porfbLio canir begnning brnac*es empoees audit examnaton La PaMguaya 1975 0 17 No No Corporai6a 1976 1 29 No No F. Paraguaya 1976 0 9 No No Rural 1965 0 12 No No Ununxey 1976 0 11 Yes No Cifa 1976 0 12 No No El Comercio 1976 0 28 Yes Yes Emole 1976 0 13 Yes No Finamirica 1977 0 5 No No Fianud I979 5 33 No No Finmncentro 1979 0 11 No No General 1979 0 12 No No Chaco 1979 0 7 No No bwrsifr 1979 3 31 No No Pantodo 1979 0 15 No No Uni6n 1979 2 17 No No Exterior 1979 0 4 Yes Yes Panati 1980 0 18 No No Corpus 190 0 8 No No C:ififlacicra 1980 0 4 No No Azr.i6n 1N0 0 13 No No FUMaaLzs & lVesiOn ls 1981 1 15 Ycs Yes S:ace 1981 0 6 No No Suda:;rica 1981 0 9 Yes Yes Fivancopar 1984 1 21 Yes Yes Veucedora 1997 0 12 No No St Ana 1997 0 8 No No Metroolimna 1999 1 19 No No Vangadia 1989 0 7 No No r'mexpr 1990 0 8 No No Roble 1990 0 6 No No Fitanban 1990 1 14 No No Tecnica 1990 0 8 No No Empresaril 1990 1 19 No No Mercantil 1990 0 11 No No Encarnacifn 1990 0 3 No No Cristal 1990 0 5 No No Agrofinaiera 1991 0 7 No No Estrela 1991 0 11 No No Fnantn:bol 1991 0 8 No No Alfa 1992 0 14 No No Efisa 1992 0 8 No No Agro Amnmbay 1992 0 4 No No Capital 1992 0 11 No No Familiar 1992 0 6 Yes Yes ndusa 1992 0 5 No No Itagna 1992 0 5 No No El Producor 1992 0 5 Yes Yes Total 16 564 Source: Celrtal Bank of Paraguay, Superintcaency of Banks. 174 Reiaodo Pewer Table 9.7. Bank Credit Extended to Private Sector by Composition, December 31, 1985-92 Casegoy 1985 1986 1987 1988 1989 19O 1991 1992 Billon Of .gardes Nadonal development bnlk 36.1 57.8 77.0 84.5 133.6 157.5 231.0 285.1 Agriculture 29.9 46.5 61.3 58.1 99.1 129.3 176.0 227.3 Live stock 0.6 0.9 1.6 1.9 3.1 3.5 4.3 6.3 Induwy 2.9 6.0 7.9 17.; 23.8 17.7 132 31.5 Commerce 2.2 4.0 5.5 7.1 7.3 5.3 18.1 132 Export 0.3 0.3 0.5 - - - 162 6A Other - - - 02 - I.S 3.0 0.5 Commercial banks 266.2 334.1 4622 669.1 1.054.2 1,6605 2,726.7 33904.4 Agrcute I5.0 81.6 86.0 134.4 227.7 318.5 338.7 317.6 Live stock 9.4 6.4 13.8 22.8 28.4 34.5 63.6 90.6 industry 50.6 48.9 73.6 83.3 127.5 155.7 267.3 2523 Cowmen 132.2 1315 208.3 324.1 419A4 722.6 1,024.3 2.044.8 -'XpOTt 42.0 46.4 57.2 63.0 149.6 329.6 469:2 597.4 Consructions and othes 16.7 19.1 23.0 41.3 101.3 99.5 383.4 601.5 TOTAL 302.4 391.9 S39.2 753.7 1,187.8 1,818.0 2,957.8 4,189.6 Percntg share Natonal developmmbank 12.0 14.8 14.3 112 112 8.7 7.8 6.8 Agriculre 9.9 11.9 11.4 7.7 8.3 7.1 6.0 5.4 Live smck 02 0-2 0.3 0.3 0.3 0.3 0.1 0.8 Industry 1.0 1.5 1.5 23 2.0 1.0 0.4 0.8 Commerce 0.7 1.0 1.0 0.9 0.6 0.3 0.6 0.3 Export 0.1 0.1 0.1 0.0 0.0 0.0 0.5 0.1 Other 0.0 0.0 0.0 0.0 0.0 0.1 0.1 0.0 Commerial banks 88.3 852 85.7 88.8 88.8 91.3 92.2 93.2 Agricutu 5.0 20.3 15.9 17.8 19.2 17.5 11.5 7.6 Lit stock 3.1 1.7 2.6 3.0 2.4 1.9 2.2 7.6 Industry 16.8 12.5 13.7 11.1 10.7 8.6 9.0 6.0 Commerce 43.7 33.6 38.6 43.0 35.3 39.7 40.7 48.8 Export 13.9 11.9 106 8.4 12.6 18.1 15.9 14.3 Cons_actions and othm 5.5 4.9 4.3 5.5 8.5 53 13.0 14.4 TOTAL 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 Source: Central Bank of Parguay. Fuandal Liberaizion inca Agrarian Econn.: The Case of Panac u 175 Table 9.8. Imerest Rate Structure, 1989-92 nmciL system rates in an i perceage) December Rates 1989 1990 1991 19 Deposit r Cmmrcial banks Saving deposit (at sight) :i.0 12.0 12.0 10.S Term deposits 60 to 90 days 14-18 16.0 13.0 16.6 91 to 180 days 14-18 17.0 12.9 14.0 More than 180 days 14-18 24.0 14.0 13.8 Saving deposit certificates - 27.0 1810 20.6 Housing, saving and loan associatons Saving depoit (at sight) 8-11 8-20 11.5 9.7 Term deposits 60 to 90 days 14-18 22.0 17.5 16.0 91 to IGS days 1413 26.0 24.0 22.6 More than 180 days 14-18 30.0 26.0 27.7 Fmancial insdtutions Pmisso notes isued against deposits 13-20 30.0 24.6 25.8 Saving deposit certificates - 29.0 25.8 28.0 Dev9opm banks Securides - - 16.3 16.3 Loan rmat Commercial rates Conunercial banks 28.0 40.0 27.1 30.1 National development bank 28.0 40.0 27.5 32.6 Fmancial insdtdons 35.0 42.0 40.2 44.2 Housing rates Housing, saving and loan associations 22-30 28-36 29.0 29.0 Development rates Commercial bank 28.0 40.0 25.8 28.8 National development bank 22-28 36.0 24.2 25.0 Development banks 28.0 40.0 25.8 28.8 Source: Central Bank of Paraguay. 176 Reinaldo Penner Table 9.9. Centrd Bank Rediscount of Banking System Loans to Export and Agriculture Sector, 1980-92 (millon of guaramues and percentages) Lamu by nadama developme bank 01N) Agdcuknr Epnt RcdEs- Radi- Radr- TOt cowu Yewr fn coned Lass coswed kn b a PerN2 1980 7,54 6.826 60 401 7,223 1.S25 21.1 1981 10.902 9.811 643 431 10.243 2.238 21.8 1982 10.486 9.43 8 5 9,443 7.653 81.0 19S3 10.11*Y 9.124 108 7.272 9.197 7.428 80.7 1984 14.574 13.116 559 374 13.491 8.142 60.3 1985 14,414 12.972 161 108 13.081 8,846 67.6 1986 '6.977 15.279 103 6.942 t5.349 10.58 68.9 1987 18.399 1659 153 103 16.66 14.012 84.1 1988 14.166 12.749 6 4 12.753 10,165 79.7 1989 19.167 17.250 1 1 17.251 13.121 76.0 1990 18.090 16.281 1 1 162 16.274 99.9 1S91 19.922 17,839 1.829 125 19,065 12.464 65.3 19921 19.386 17.447 570 382 17.830 14.354 80.5 Asgriaure Erpn Cau ed flE&- Red.- Radi- Reds- To- counted Tota cmned Yer Loai cowed Loan tuwed km by Q M" frn by CB Perma2 199D 4.573 4,573 35,045 23.40 28.053 3.14 11.4 35231 4.739 13.4 1991 5.043 5.043 31.055 20D.07 25. 3.S13 13.3 36.093 5.32t 16.1 1932 3.797 3.797 18.4 12451 16.249 2.8W 17.5 25.692 100 40.9 1913 4.281 4.231 15.511 10.44 14.745 3.977 269 23.942 11.40 47.6 194 4.975 497S 31.105 20.40 25.115 1.166 31.6 39.307 16.309 41A 1955 7.256 76 20.216 135 2t.301 17.010 31.7 33.82 25.36 7.3 1996 29.7Y9 29.799 16.953 11i362 41.161 21.604 S4 56.510 32.187 56.9 19B7 25.773 25.773 17.151 11.493 3771 15.644 41.9 53.933 29.657 54.9 19B 32.733 32.733 15.366 10,195 43.029 22.175 515 5S.793 32.341 57.9 1999 44.03S 44.035 21.930 19.33 63.41B 25.7 404 8.670 33.779 43.0 1990 44.564 44564 46.113 3D.195 75.46D 17.023 22.3 91.743 33.293 363 1991 38.139 33.139 52.130 35.396 73.5 623 5 92.601 18.753 20.2 19921 26511 26.511 52205 34.9n 618 2.476 4.0 79,319 16,830 212 1. June - Nocv. 30, 1992. 2. Percent 4f te total agriculture and export portfolio subject to rediscount (products subject to rediscoun: soybean, cotton, wheat, sugarcane). Financial LibeUzaldon i a- AgriaR Econaynu: The Case of Paragua 177 Table 9.10. Consoldaled Loans to the Private Sector by the Banking System, 1988-92 December Cgaregiy 198 1989 1990 1991 1992 Niliciw of panmec Loans in local currency 343.0 466.7 684.9 993.9 1,381.9 Rediscount facilities 93.2 111.1 117.1 112.2 170.2 Deposits 249.8 355.6 477.4 723.0 1,023.3 Mutual funds reso'c - - 902 158.5 188.2 Loans in fbreign currency (0$) 23.5 29.5 85.0 241.5 315A (From which mutual funds resources) - - - (3%) P%) Total loans 366.6 496.2 769.9 1,235.5 1,697.4 Percen Loans in local currency 93.58 94.04 88.96 80.45 81.42 Rediscoun facilities 25.44 2L38 15.22 9.09 10.03 Deposits 68.14 71.65 6Z.01 58.52 60.29 Muual funds resorces 11.72 12.83 11.09 Loans in freign curreny (US) 6.42 5.96 11.04 1955 18.58 (From which mutual funds macrca, Tota loans 100.00 100.00 100.00 100.00 100.00 Source: Centrl Bank of Paraguay. 10 JAPANESE BANKS, SPECULATIVE BUBBLE, AND BANK SUPERVISION Takashi Kazaki, Bank of Japan I will discuss two subjects based on the Japanese experience: the recent speculative bubble and the Japanese financial system; the framework of on-site examination by the Bank of Japan. Japanese Banks and the Speculadve Bubble First, let me begin by reviewing the recent fluctuation of asset prices in Japan and the stability of the Japanese financial system. It is true that recent business circumstances surrounding Japanese financial institutions have become increasingly severe. The major reasons for this are: the emergence and collapse of the speculative bubble in asset prices, the slowdown in the Japanese economy, and ongoing financial liberaliztion in Japan, :ncluding liberalization of interest rates, which bas raised fundig coss and intensified competition among Japanese financial institutons. Of these three reasons, the first one is especially significant. Emerging Risks Against the background of the shaip rise in asset prices in the late 1980s, Japanese banks competed excessively with each other, giving first priority to expansion of their bankdng activities. Consequently, credit, market, and management risks have emerged. Credit risk emerged as many Japanese banks, in expectation of a contiuous rise in land and stock prices, excessively extnded real estate loans and real estate secured loans in the late 1980s. In addition, Japanese banks used nonbank financial subsidiaries to expand these loans. Furthermore, in this process of credit expansion, Japanese bans began to pay inadequate attention to credit analysis, follow-up management of loans, and relied too much on coilateral. As a result, as asset prices declined, credit risk emerged. Japanese banks' asset quality, especially that of real estate related loans and loans to nonbank financial institutions, has deterirated since 1990, as can be seen in the increase i nonperforming loans. 179 180 Takashi Kazzaki Market risk emerged to cover the increase in funding costs due to interest rate liberalization. Many small and medium-size Japanese banks increased investments in high-risk, high-return securities such as stocks, stock investment trusts, and equity- linked structured bonds without adequate knowledge of these assets and a strong asset and liability management. Accordingly, these banks' asset portfolios deteriorated as stock prices declined. Management risk emerged as the expansion oriented policy of Japanese banks penetrated into the branches, exerted pressures on the staff, and weakened internal controls, thus inducing financial irregulaities. To cope with the emergence of these risks, the Bank of Japan, through on-site examinations and other supervisory tools, has guided Japanese banks to take comprehensive and effective corrective action. In response to the Bank of Japan's supervisory guidance and the lessons from the recent deterioration of asset quality and financial irregularities, Japanese banks are implementing various corrective measures. Corrective Measures First, the banks are shifting business priorities from quantitative expansion to enhancement of qualitative elements such as returns on assets. This shift is also due to tbe requirement to meet the Bank for International Settlements (HIS) capital standard. The plunge in Japan's stock market prices reduced Japanese banks' unrealied capital gains on securities holdings included in Tier-2 capital, thereby exerting negative effects on the BIS capital ratios of Japanese banks. Under such circumstances, Japanese banks are restraining asset growth. The second corrective measure is the reinforcement of credit analysis funcidon, or a separation of credit analysis sections from business promotion sections. In the process credit underwriting, creditworthiness, and financial conditions of a borrower are to be fillly examined by a credit analysis section. Japanese banks, based on the experience of the speculative bubble, have been strengthening credit analysis sections by giving them independent fnction and authority to reject inappropriate loan applications. The third measure is the rnforcement of internal controls, including the strengthening of the internal audit system and the dual control system. The fourth measure is the stengthe of controls over nonbank financial subsidiaries. This is especially important for Japanese banks, because in the late 1980s nonbank subsidiaries of Japanese banlk' extended real estate loans at a rapid pace without adequate credit analysis and proper control by the parent banks. In addition to the above mentioned corrective measures, in my view, Japanese banks should implement three additional measures. First, Japanese banks should efctively deal with the icrease in problem assets. Because of the recent inactive real -estate market in Japan, repayment dues have Japanesenks, Speadaive Bubble, and Bank Supervision 18 been increasing in real estate related loans made by Japanese banks and nonbank financial subsidiaries. Also, the liquidation of coUateralized real estate has been lagging. In this situation, Japanese banks, based on the principle of self-responsibility, should effectively handle problem loans by employing all possible measures, such as establishing a task force or a special division to collect problem loans and appropriately charge off bad assets. Second, Japanese banks should promote consolidation. By this I mean that Japanese banks should focus on strategically selected core banldng businesses while reducing unprofitable sectors to improve overall profitability. Broadly, consolidation may include bank mergers and acquisitions to enhance efficiency. Tbrough such consolidation, Japanese banks should enhance capital positions as a buffer to absorb potential losses. Third, Japanese banks should promote the disclosure of banks' financial conditions. Fmancial conditions of Japanese banls are published through financial statements, including balance sheets and income statements. However, since the quality of assets, in particular the condition of nonperforming loans, is so far undisclosed, it should be gradually disclosed, which will contnbute to the enhancement of the market discipline of banks. The Bank of Japan will continue to closely watch the effectiveness of the correctve measures of Japanese banks and wEIl provide supervisory guidance as needed. Bank Supervision Next, I will speak about lessons for bank supervisors. Based on our experience m dealig with the speculative bubble, I believe that bank supervisors should implement the following three methods of bank supervision. First, bank supervisors need to conduct on-site examinations on a regular and full-fledged basis. The merit of on-site examinations is that supervisors can directly review the overall operations of a bank, in parficular, they can thoroughly assess the asset quality. It is essential for bank supervisors to detect problems as early as possible and to take supervisory actions as promptly as possible by conducting regular and full- scope on-site examinations of banlks. Moreover, such examinations can help prevent the emergence of systemic risk. Second, on-site examinatons should be conducted at the discretion and judgment of supervisors and specific features of individual banks should be fully taken into consideration. To appropriately supervise banbs, the discretion and judgment of supervisory authorities is more important than the mechanical execution of rules and formulas as stipulated in laws and regulations. It is thus necessary for supervisors when conducting an on-site examinaton to make fair supervisory judgments flexibly, 182 Takashi Knz takng into full consideration the macro-financial sitLation and the specific features of the bank. Bank supervisors need to take effective and concrete supervisory action based on the actual condition of a bank, after due consideration has been given to corrective measures essential for each bank. Third, bank supervisors need to conduct examinations on a consolidated basis and closely monitor the financial conditions not only of parent banks but also of domestic and foreign subsidiaries. The bank supervisor should monitor the actual condition of the whole organization of a bank, including its subsidiaries, because the emergence of risk is unavoidable between a parent bank and its subsidiaries. In response to the globalization of Japanese banlks' business activities and the deterioration in the financial conditions of their domestic subsidiaries, the Bank of Japan has recently been promoting consolidated supervision. Putting this in more concree terms, the Bank of Japan has been strengthening the on-site examinations o- the overseas branches of Japanese banks. In the case of large banks, as part of regular on-s e tio the Bank of Japan sends examiners to major overseas branches to examine their asset quality and other financial conditions. In the case of medium-size banks, examination teams are sent abroad four times a ully to conduct on-site examinations of the branches of several selected banks in each major financial market. Furthermore, the Bank of Japan has been reinforcing the monitoring of domestic nonbank fiancial subsidiaries of Japanese banks. Although the Bank is not authorized to conduct on-site examination of bank subsidiaries directly, it obtains detailed information on the condition of such subsidiaries when it examines parent banks. Iu particular, since the domestic subsidiaries of Japanese banks have made substantial real estate loans and other loans through funding from their parent bank and also from other banks, the Bank of Japan analyzes the asset quality and other financial conditions of these subsidiaries by obtaining mformation from the parent banks. Bearing in mind the emergence of risks and the response of the Bank of Japan and Japanese commercial banks, I will next outine the impact of the recent fluctuation of asset prices on the management of Japanese banks and the Japanese financial system. The Fluwtuaion of Asset Prices on Bank Management Because of the recent decline in ierest rates, interest rate margins and, as a result, the operating profits of Japanese banks have been recovering. Thanks to this, most Japanese banks have been aole to absorb losses from past due interest and loan charge-offs caused by the recent deteroration of assets. In addition. retamined earnings of Japanese banks have greatly improved since the 1980s. Therefore, the financial risks can be managed by major Japanese banks, and the recent flucuation of asset prices has not strongly damaged the overall financial system in Japan. The Bank of Japan believes that the banking system wi;l recover from the urrnt situation if the banks effectively implement corrective measures. Japanese Banks. Speculaive Bubble, and Bank SWervision 183 I conclude my first subject by touching on the background of the recent speculative bubble from a macroeconomic perspective. The question is why the speculative bubble developed simultaneously in the latter half of the. 1980s in major countries, including the United States, the United Kingdom, France, and Japan. In my view, it was partly related to expansionary monetary policies implemented by most industrial countries in the 1980s. However, I think that the most important factor was inflationary expectations for asset prices such as real estate and stocks that provoked speculative transactions of these assets. Borrowers, lenders? and investors expected a continuous rise in asset prices and went to excesses. Such speculative and excessive risk-taking prevailed on a global basis as a result of the globalization of financial and real estate transactions. The most important lesson financal institutions can learn from the recent collapse of the bubble is to find ways to prevent the recurrence of excessive and speculative risk-talcing. On-Site Examination Next, I will move on to the second subject, namely the framework of on-site examination by the Bank of Japan. In supervising individual banks, the Bank of Japan conducts regular on-site examinations once every two to tbree years as well as off-site m:onitoring through day-to-day contact. The basic objective of an on-site eaamination is to thoroughly review the business condition of the bank being examined and determine whether it is financially sound and to take prompt corrective action if the examination identifies problems so as to protect depositors and ensure the stability of the fiancial system. This early detection of problems and prompt corrective measures based on on- sihe examinations have become increasingly important This is because the progress of deregulation, globalization, and increase in off-balance sheet transactions have created risks that are increasingly diversified, complicated, and globalized. This is also becmase credit risk, market nsk, and management risk are emerging because of the collapse of the speculative bubble. In terms of the scope and focus, the Bank of Japan's on-site e ion covers all areas of bank activity. Given the limited time for on-site examination, which ranges from two to four weeks, the most important components of on-site examination are: (a) to thoroughly examine asset quality while also considering capital adequacy; (b) to evaluate risk management capabilities; and (c) to evaluate profitability. I will discuss each of these in turn. 184 Takault Konzak Asset Quality A deterioration in asset quality is the most crucial element in the worsening condition of a bank. In this connection, a survey by the Office of the Comptroller of the Currency of the United States points out that the most decisive factor accounting for U.S. bank failures was the deterioration of asset quality. Evaluation of asset quality is thus the most important part of on-site examination. More concretely, evaluation of asset quality begins with identifying problem assets, including on- and off-balance sheet items. The Bank of Japan, like the Federal Reserve Board of the United States, classifies problem assets into three categories: loss, doubtful, and substandard. Then the ratio of these classified assets to total assets is computed to evaluate the overall soundness of asset quality. Furthermore, by comparing the capital ratio and the weighted classified asset ratio, examiners evaluate the real financial strength of a bank, problem assets, and the need for additional capital. In evaluating the severity of a bank's problem assets, it is thus important to assess the implication of the problem of asset quality for fincial strength. For example, even when a bank has a high capital to asset ratio of 10 percent, if its weighted classified asset ratio is 10 percent, we judge that the bank's real finncial position is significantly weak because the bank's capital position is estinated to be criically impaired by problem assets. In evaluating asset quality, it is also important to review the concention of credit extended to any given industry or corporat group. In many cases, such credit concentration leads to a substantal deterioration of asset qualty. Risk Managemem Capability The second important component of on-site examination is to evaluate the capability of a bank to manage various risks. The Bank of Japan's supervisory philosophy of risk management is that, based on the principle of self-responsibilty, banks themselves should effectively manage risks. During on-site examinations, the adeqacy of the risk management framework of a bank is closely evaluated. In evaluating nsk management capability, a wide range of items should be checked against the background of financial liberalization and globalization. Because of this, in 1987 the Bank of Japan distrbuted a checklist for risk management to all banking institutions, which describes the risks to be managed and the management items to be checked during on-site examination. Risks sudied during on-site examinations cover a wide range from traditional banking risks, such as credit risk and daily operations ris, to market risks, such as interest rate risk, price-fluctuation nsk, and foreign exchange rsk, and to adminstr;ative and other mangement risks. In summary, risk management capability is evaluated to determine whether a bank effectively controls rislks by establising an Japwwse Bas. Speciudaive Bubble, and Bank Supervision I8S organizational risk management framework, including interal controls, rules, guidelines, position limits, and so on. Protability The third important component of on-site examinations is the evaluation of profitability. This involves measurng the real earnings of a bank after adjusting for windfall gains or losses, and other extraordinary elements; investigating whether financial statements contain any end-of-year window dressing; and forecastng future profits and losses. In addition to these important targets of on-site examination, the Bank of Japan visits some branches to evaluate the accuracy and efficiency of day-to- day business operations and to investigate any financial irregularities By combining all the various evaluations, the Bank of Japan makes aa overall evaluation of the examined bank in terms of safety and soundness. If the overall evaluation is unsatisfactory, the bank is classified as a problem bank and is subject to the Bank of Japan's strict supervisory guidance. conclusion I have today discussed various issues concerning the Bank of Japan's supervision of individual banlks. The dual mission of a central bank is undoubtedly to stabilize the value of the currency, or price stability, as well as to maintain the stability of the financial system. To accomplish this, it is imperative to appropriately supervise individual banks. But, against the background of increasingly difficult busmess circumstances surrounding financial institutions, bank supervision has taken on a high public profile in many countries. Therefore, it has become increasingly significant for centrl banks to secure the safety and soundness of individual banks by detect problems early and tking corrective measures promptly so as to protect depositors and stabilize the financial system. In this regard, safety nets, such as deposit insurance systems and the provision of emergenry liquidity assistance by cental banks, play an important role in securing confidence in the financial system. Safety nets, however, are not a panacea, and they also often encourage the moral hazard problem, as we learned from the experiences of the United States. Thus, such safety nets should work as the 'Very last resort Authorities should first make every effort to supervise aMd gtide individual banks appropriatly. From this viewpoint, central banks in each country should give priority to establish an effective on-site examination framework. At the seventh ational Conference of Bank Supervisors held in Cannes, France, in October 1992, Mr. Corrigan, Chairman of the Basle Committee of Baniing 186 Tokashi Kanzaki Supervision, said in his address to bank supervisors worldwide that all bank supervisors should hang together or hang separately. By that statement, I think he meant that bank supervisors in each country should unite and coordinate worldwide, otherwise we will be defeated one by one. I conclude my presentation by saying that the Bank of Japan would like to hang together with all the participants at this seminar to enhance the effectiveness of bank supervision on a global basis. Japanse Banks, ulative Bubbk, and Bank Superviuion 187 Table 10.1. List of Financial Institions in Japan' Type Fmanciad institutions Number insdtudoons City banks 11 11 Japanes banks2 Regernl banks3 64 64 Regional banks I' 66 66 Trust banks 7 7 Long-tem credit banks 3 3 Rori?nbans 9S 94 Priva fiancial Shinki bmnls 439 354 Fmancial Shoko Chukin Bank 1 I Isainzions for small Credit cooperatives7 396 1 busines Labor credit associations' 48 1 Financial Ncsinchukin Bank 1 1 for agriculturc. Agriclaal cooperativ 3,432 0 and fisheries Fisbhey cooperatives' 1.7tI 0 Insurance companies" 52 0 Odler financial Securities coinies" 209 39 institutions Securities fine companies 3 3 Government banks 2 0 Govermres fmancial instilutons Govcrmert corporations 9 0 Posal savings 1 0 Otrs 2 0 Total 6. 651 1. As of June 1992- 2. Member Banks of die Federation of Bankes Association of rapan 3. Member Banks of the Association of Regional Bank. 4. Member Banks of the second Association of Regional Banks 5. Inludes foregn trust banks. 6. Inludes Zenshine BankL 7. Incs Natonal Fedration of Crefit Cooperatves 8. Incudes National Federation of Labor Credit Associaton 9. As of May 1992. 10. As of April 1992. 188 Takash Klnk Table 10.2. Process of Full Esaminaton- (Te Bank of Japan) (1) Case of Money Center Banks (2) Case of Regional Banks (One month before) Dermination of the (One month before) Determinaion of the bank examined, nodce bank examied, otic utode bank examined to the bank exaingd, colectg inlrfnation colteeting information aid data and data crwo weeks before) Pre-sdy (Two weeks before) Pre-sudY (The fist wek) Ssarr of on-se exa- (The fist wck) Stant of on-sit en- minaion mination Mom Mon Tuc Intevie with tde The lmrview with the presidce executive prside cutive directors, and depart- directors, and depart- met managers ma mnges Wed Wed Thu Thu Appaisal of assets Fri Fri (CTe seod week) Appraisal of assets (Te second wek) Mon Mon Examinnion visit to branch offices Tue Tue Wed Wed Genal examinatio Thu Thu _ _ _ Fri Fri Review ,ce ird week)tb Mon Examintion visit to End on Ut site examintio branch offices TueC Ratng. Reporting and Revw Wed Fri Genral examdon (The fourth week) Mon Tuc Wed Review End of on-site examination Rating. Repor-ing and Review Japnese Banks, Speculative Babble, and Bank Supervision 189 Table 10.3. Main Items of Check listfor Risk Management L Credit risk 11. Interest rate risk 1. On-balance-shcet asset, dometsdc 1. Mimatched Position (1) Prior assment (1) Directors and depament heads' awareness of the risk (2) Follow-up magement (2) Fuctions of the ALM (Asset Liability Managment) commitce and the subcommitte for interest rates forecasting (3) Couniy risk analysis (3) Insullation of datbase and comWterized supporting system for checldng gaps (4) Lending discipline (4) Establishment of gap limits and its authorizadon (5) Computerized supportng system (5) Gap management 2. On-balancshest overseas 2. Dealing in public bonds (1) Country risk analysis (I) Directors' and deparmnent heads' awareness of the risk (2) Management of loans made to debtor (2) Organization and system counries (3) Assessment of commercial risk (3) Establishmet and management of position limits (4) Management of commerial risk and credit (4) Profit management security (5) Global profit management (5) Computerized system support 3. Off-balance-sheet transactions 3. Management of securities in the investmen account (1) Disclosure re s (1) Directors and deparment heads' awareness of the risk (Z) Restictions on the borrower concerning the C2) Authorization fbr semng position limits and for change of business, sales of assets and the purcase and sale of securities mergers C3) Resticions on collateral and financial ratios (3) Managemct of investments in forcign-currency- and the material adverse change cluse denominated securities (4) Components constituting default (4) Management of investments in bonds (5) Management of financial data and (5) Management of investments in stocks authorization table condemn onfolowitg page 190 Tacah! Kaszak Table 10.3. (continued) IlL iquidity risk VI. Systnemic risk (1) Directors' and dqeanment heads' awareness of the (I) Direcors' and deparmnt heads' awarenss of risk die risk (2) Funding neasures reitaed to the increase in (2) Management of risks involved in dhe cllective securities positions direct credit system (3) Fund mangement with regard to off-balance-sheet (3) Risks involved in die on-line cash dispense tr_answuons system (4) Coutmasms for credit crunch (4) Verificaion of contract provision relaxed to the partiipation in large-scale payment system (5) Orgaizanions ad sysmns imvolved in argscalc payment system (6) Supporti system for fiuds transfer WV. Foreign exchange rate risk VIL Daily opeatn risk (1) Direcors' and depatmnent beads' awarenss or the (1) Orgnation and system risk (2) Organizaon and system (2) Cask and docments (3) Establishment and mangement of position limits (3) Management of exceponal transactions (4) Profit management (4) Handling of receivabeWpyables and special deposit (5) Compterized support systcm (5) Measures to prvent troubls with custmers V. EDP rsk VM. Managemnt isk (1) Crin and disaster prvention mneasres and back- (1) OrganizLtion up system (2) System developmem (2) nteal audit (3) Sytem operation (3) Management of affitled companies (4) Personnel nmnagement (4) Capacity of th management (5) EDP inspecdon (5) Quality of emplyes and taining programs Japanese Banta, Specutative Bubble, and Bank Superision 191 Table 10.4. Checklist for Credit Risk Management Checkli Item Prior 1. Analysis of marugementmcredit- * Is the bank aware of the managerial performance of the Assessment worthiness of finns and individuals borrower? * Financial amlysis based on financial statements and trial balance * Executive administrative ability and health * The use of databases for analysis of a firm 2. Assesment of business plans. Is the assessment hosed on the rollowing detiled items. taking usage of funds and repayment business oulook and usage of funds into account? ability of the borrower * Investigation on profitablity of the plan * Investigation on the usage of funds * Investigation on the borrower's financial resoures for repayment * rlear rules for seting the maximum borrowing linit * The use of a hud's allocatin table to judge repayumnt ability 3. Use of firn ratings * Is there a unifonn standard for assessing businss performane such as frun ratings? * Standardized critcria such as the scale, financial conditions, and profitability of the debtor company * Regular reviews of finn ratings 4. Ability to check commercial bills * Does the bank have the ability to check customers' requests to discount nonsdard types of comnecial bills such as acconmodation bills? 5. IWlementation of measures aiming * Arc organizational measures twen to improve the staff's crcdit at improving credit analysis ability analysis ability? Follow-up 1. Follow-up system for borrmwers of * Is them a sufficient follow-p sysem concerning borrwers of large amounts and borowers with large amounts and borrowers with certain problems after loan problems execution? * Risk management focusing especially on bormwers of large amounts * Paying special attention to the main bank's and other banks' behavior toward the borrower * Investigation of die usage of fuinds and monthly monitoring of the financial conditions of problem borrowers 2. Monitoring business performance of * Is financal analbsis aftcr the loan execution conducted. borrowers including that for borrowers with no partclar problems? * Investigation on the usage of funds 3. Investigation of the auge of funds * Is the usage of funds among group companies. e duly among group companies, fimily recognized after thc loan execution? members. and individuals 4. Support by research section * Does a researh section (including related research companies) coUlect useiil information for credit analysis of a firm? 192 Tarwfinaki Chart 10.4. (continued) Caet 1. Man gene and reevaluaoin * Is credit duly secured agins bkrupty of dte borrower? SCut S Standards properly set fbr colaerl an ent real esate as collael * Encourgment of field 'inmvtgatiom * Relar revation of vAhles * Awareness of the collaterI leway securities as collateal * Flexible reevaluatin in the case of a stock ma crash * Mondlly reevaladon of sock marker vales 2. Investigton oa guarators * Same as above (C) credibltyand the reconfirmation * Reconfirmation of inton to be a uaranor and to gim of his wilnes to be a guarantor secuty m a dtird party by giving a comm_me on paper * Assessment of asset condition and auaal iwnme of te ____________________________________ Dguarantor 3. Mamgmncnt of loas in arrears and * Is the tramnt of bad debs appopriate? debls whch need careftl Clear crt for the desigan of credit whichris assessmInt carefil a _esma and for nnaserig of anes to dhe bad office * Distion of a list of loa in aears to bnches * Clear poicies established i respectd of cass of dsomnmr and _banuptcy and prompt rsponse to such eves 4. Colleciginfornationonthe * Is infomarion on inividuls concerning credit systmatcaly anagemetsnof ponal male 1. Inde of crdit sris * Is indendee of crdit aalyushilsk managem se - mgemenrscut n fully nsured? E S Is loan promoions section approyiady cheked? . - ____ 2. Branck managees discretonay * Is dicreton as regards credit extension d-degate kng authoriLy. maximum amount aopriately? of Ioans able to be extended * tdmstrm loan amount revwed every tem * Maximun loan amount revicwed min rsposC to dc condition of individual bormrers 3. Power of head office decisions * Can tde head office guind and conto bmches with clear authority based on its credit analysis abily? * Adequate number of loans to be asesd by each crdit * AbDiy ard experien of credit analysis magr * Functional efficiency of boatds of _m an directors 4. Follow-up of loan cliions * Ar loan conditions strictly observed * Folltw-up conductd through loani condti record? S. Violation of lnding discipine * Is there any violation of lceing disciline? * Check if ay violations inludit folowing have occmurr: -mquest for apprval aietr a crdit extson -negleting to collect wn guarante areemes -credit extended eceedg an officers discretion -stict maagement of de dates, few paymt ddays GOVERNMENT INSURANCE AND FINANCIUL INTERMEDIARIES: ISSUES OF REGULATION, EVALUATION, AND MONITORNG Ross Levine This paper was prepared fer de cormletion of sermnar series d has been icldd n es publication as bacground readng Edior Establishing appropriae financial sector policies is of paramount imrtance to policymakers, because financial iter ies provide services that are necessary for economic growth. If inherent characteristics of the market for fiancial yinrdir services suggest that unreguIated markets will inadequately supply these cmcial services, then governments have a responsibility to consider interventions to ove the provision of financial services. These government interventions themselves, however, often deleteriously distort the behavior of participants m financial markets, so that a delicate balance must be achieved between interventions that ameliorate market filures and the negaive effects of these government interventions. lhis balance will depend on legal, political, and historical conditions and especialy on the abflities of the pnvate and public sectors to monitor intermediaries. After reviewing why financial intemiediary services are essential for economic growth, the paper discusses characteristics of financial markets-often termed "market fafflures'-tbat imply that unregulated markets will not produce a socially optimnl supply of financial intermediary services. Although unrestricted markets may create mechanisms to ameliorate the negative effects of these market failures, the inherent charistics of financial interactions plus the institutional, legal, and political arrangemnts in many countries strongly suggest that these market failures remain important The paper evaluates controversial financial policies in terms of whether these policies mitigate the negative effects of these market failures. Specifically, the paper analyzes government insurance of financial intermedary liabffiies and regulatory restictions on the actvities of banks in terms of how these financial policies affect interediary monitoring of firms, private and public sector monitoring of financial intermediaries, financial stability, competition in financial market, and the availability of financal services. 193 194 RossLevine Each financial policy tends to have positive and negative eff ects. Moreover, economic analysis alone does not provide uneuiivocal answers to fnancial policies regarding government insurance of financial intermediary Liabilities or the proper range of powers that should be granted to banks. Country-specific legal, historical, and political characterstics wil and should shape financial policy choices. Therefore, selectifg a proper mix of regulatory and supervisory strategies to maximize the provisioa of crucial financial services is not unambiguous. Nevertheless, evaluating financial policies in terms of the issues emphasized in this paper allows thorough analyses of financial policies in any country. The paper offers some conclusions regarding government insurance of intermediary liabilities and the powers and latitude that should be granted to banks. These conclusions are conditioned on legal, political, and institutional considerations. Intermediaries and Fmancial Services Government supervision and regulon of filnanial intermediaries need to be reviewed in the context of services provided by financial intermediaries. The pivotal imwran:c of these services in generating economic growth suggests that governments should focus on enacting financial policies, mcluding potentially laissez-faire policies, that encourage financial intermediaries to supply a sufficient quantiy of these serices. A basic service provided by financial intermediaries is that they facilitate transactions, so that businesses and individuals can engage in a broader set of mutually beneficial trades; put succinctly, financial intermediaries encourage commerce. Second, fnancial intermediaries ease the trading, sharing, and pooling of nsk, which sfimulates more efficient resource allocation and faster economic growth. Third, financial intermediaries mobilize resources from disparate savers. Some worthwhile investments require large capital inputs and some enjoy economies of scale. By agglomerating savings from many individuals, financial intermediaries enlarge the set of projects available to society and thereby enhance economic efficiency and development. Fourth, after researching firms, managers, sectors, and business trends, financial intermediaries invest societies savings. The better financial itermediaries are at obtaining and processing information, the better will be the allocation of capital and the faster will be the rate of economic growth. Finally, fiancial intermdiaries evaluate and monitor firm managers and thereby compel managers to act more in the interests of stock and debt holders than would a disparate group of individual shareholders none of whom would find it worthwhile to undertake the large monitoring costs individually. By facilitating the ability of principal claim holders to monitor managers, that is by ameliorating the principal-agent problem, financial intermediaries encourage more efficient resource allocation. Also, by mitigating the principal-agent problem, financial intermediaries encourage greater diversification, since ownership will not have to be concentrated in the hands of a few owners that find it worthwhile to Gbenmenlwurwwc and Fmancfai Intermnedifarir Issu ofRegzmoiz Evahwio, and Monitorn 195 monitor the managers. These five finanil services are crucial for economic development Therefore, if there exist fundamental economic reasons why a free market will not adequatly and appropriatly supply these services, the goverment should consider interventions m the financial sector to improve economic development.1 Reasons for Government Intervention This section describes inherent characteristics of the activities of financial ntermediaries that create a potenially positive role for government intervention in financial markets. Put differently, there exist good economic reasons for believing that free financial markets-financial markets that are unregulated and unrestricted by the government-will frequently not produce an optimal quantity of financial services, so that government supervision and regulation of financial market activities can sometimes improve social welbre. Fmancial regulations, and financial policies more genrally, should be evaluated and compared in terms of whether they ameliorate or aggravate the negative effects of these five market failures. Specifically, financial policies can be evaluated in terms of how they affect the finacial intermediaries egarding the degree of fcial stability, the degree of competition among financial intermediaries, and the spectrm of available financial arrangements. It is also important to determine how financil inr ediaries evaluate and momtor fErms, the borrwers; and how the regulators evaluate financial institutions, the creditors 2 Fimancial policies typically involve tradeoffs among these issues. For example, as argued below, government organized deposit insmuane schemes often reduce the probabiity of bank runs while also reducing incenives for depositors to evaluate and monitor banks. Constructing or reforming a financial regulatory regime, therefore, involves a complex process of choosing a mixture of policies that maximies the provision of financial services. As will become clear, the appropriate mix of policies depends importantly on the efficiency of the country's legal system, as well as on the country's institutional, political, and historical character. 1. Conceptally, any sntu of the role of government in financial marets owes a great debt to Joseph Stiglkz. This paper relies heavily on Stigliz (1993). On the empiical and eretical lins between financial services and growth, see Caprio (1994) and King and Levine (1993ab). 2. The terms evaluating and mmonitoring" should be inteted broadly to include asseg the quality of management, the performance of the intermeda y te accmuracy of disclosed information, the co ons between mangment and firms or other intermediaries. dte quality of the business plan, and changes in the character of the intermediary without the consen of creditors. 196 RoSS Levine Externalities in Monitoring and Mocating Resources to Firms Filancial intermediaries are heavily involved in evaluating firms before they invest m firms. Furthermore, intermediaries monitor firm managers after funding the firL These evaluation and monitoring activities are valuable and costly to undertake. These acdvities are also easily observed by other investors or intermediaries, who can use and benefit from these evaluation and monitoring activities without paying for them; expenditures by one financial intermediary on selecting firms and monitorng managers create extemal benefits for other investors and finacial interediaries. Thus, unless the financial intermediaries that actively evaluate and monitor can design mechanisms to internalize all of the benefits that accrue to these information gatheing activities, the extemality associated with monitoring firms implies that financial intDermdiaries will provide a socially suboptiual amount of evaluation and monitoring services since private returns are lower than the social returns. Inadequate evaluation and monitoring will have negative economic implications for four reasons. - First, suboptimal evaluation and monitoring by intemediarim of firms imply that resources wfll be allocated on the basis of a socially suboprimal amount of information about firms. * Second, suboptimal evaluation and monitoring of firms provide management with excessive independence from firin creditors and hinder the efficient allocation of resources. * Third, suboptimal evaluation and monitorng of firms may deter mvestment by potential creditors and thereby retard economic growth. * Fourth, to the extent that the pnncipal-agent problem is not mitigated by financial intermediaries, fim ownership may become more concentrated (to ease the principal-agent problem) than it would be in the presence of sufficient monitoring services. This could hurt economic efficiency by reducing diversificatdon. An unregulated market may create methods for internalizing some of the externalities associated with financial intermediary monitonng of firms. For example, a firm may pay higher fees and interest rates to financial intermediaries that carefully monitor the firm if the firm believes that this monitoring will be observed by financial market participan and thereby enhance the firm's access to capital markets and other Govnment inswce and Fanciaa Intermediaries: issues Of Regptalion, Evaon, nd Monitoig 197 intermediaries.3 Through this market mechanism, intermediaries that carefully evaluate and monitor firms will internlize more of the social benefits of this information gathering activity and thereby reduce the undersupply of financial intermediary monitoring of finns. Nonetheless, extemalities associated with monitoring firms are unlikely to be elimiated, so that fiancial institutions will tend to undersupply evaluation and monitoring services. Even more important for the purposes of estblishing a useful analytical framework, financial policies may affect the incentives and ability of financial intermediaries to evaluate and monitor firms. Thus, analyses of financial policies should consider the effects of financial policies on incentives to monitor firms even im the absence of preexisting externalites. Given the economic importance of financial intermediary evaluation and monitoring of firms, the effects of financial policies and reglations on incentives for intermdiaries to research and monitor firms should be part of the "checklist" of issues to be aLken into consideration when evaluating the pros and cons of financial policies. Two straightforward, though often unachieved, policy strategies for enhacing fnancial i diay monitormg of firms are worth nofing here. First, make it easier for the private sector to evaluate and monitor finns. Information disclosure laws, cometent accounting stndards, standardized and tansparent financa reporting forms, and an efficient corporate legal system that make the ownership, control, and performance of finms more transparent will faciltate the ability of auditors, crdtors, rating agencies, and finamcial intermedie to evaluate firms. Second, goverment supervision and regulation can be carefully crafted to enhanc incentives and minimize disinCentives for sound financial intermediary monitoring of fms. ernaities in Monitoring Fnancial Institions Monitoring financial institutions also has exteral effects. It is costly and time- consuming to research and evaluate the condition and prospects of complex financial intermediaries. Also important, expenditures by one entity on evaluatng a financial instittion often create benefits for other investors, who do not have to pay the tesearch costs. Instead, the market can observe the behavior of investors and agencies that have carefully evaluated intermediaries. The externalities associated with monitoring financial institutions suggest that, under many sets of insiutonal arrangements, the market will insufficiently monitor and evaluate financial institutions because the private rens from monitoring are 3. In the United States, borrowers expeuicae abnormally positive stoic retuns when they amnunce that thry have renewed loans with their banks. These abnormal retums do not materialize for nonbank debt James (1987) and LDmmcr and McConncll (1989). This allustes the imp ce the market gives to bank monitoring of firms. 198 Ross Levin lower than the social returns.4 This tendency for insufficient monitoring is exacerbated in the case of finaial institutions for at least two reasons: one, financial intermediaries frequendy have many small creditors (for example, depositors), so that the incenives for any mdividual creditor to ndertake thie expensive monitoring costs are small; and two, it is vey costly to evaluate financial intermediaries, so that only vey large, sophisticated claim holders would monitor financial institutions. Thus, free markets will likely produce a socially subopuimal amount of monitoring of financia insdtiions. Unregulated markets may respond to this dearth of creditor monitoring. For exmple, in the case of banks, small depositors would be wary of putig their savigs in banks in which they did not have confidence. Conequetly, banks might respond by designing siple, innovative ways to communicate the safety of their portfolios to savers or creae capal structures where financial intrm es have a few large creditors or owners that are respected by the public and that are exected tD monitor the intrediary objectively. While these mechansms may enance monitoring, they are unlikely to eliminate the undersupply of private sector monitoring of financia intermediaries. Uncertainty about the objectivity of large creditors and the infomaion communicated by the bank would in most cases stl imply socially suboptimal monitoring of fancial institutions. Insufficient monitorig of financial mediaries can negatively efct economic activity for at least five reasons. - First, insufficienLt monorig wil worsen the principal-agent problem of intermediaries, so that financial a managers will not act in the best interests of creditors, and financial inmiay services, therefore, will not be apropriaely supplied to the economy. As explained above, an undersupply of financial services wi tend to slow the rate of economic growth. * Second, insufficient monitoring of financi institutions tends to raise uncerainty about financial institutions. This unertainty will tend to deter investors from entrusting their savings to fiancial institutions. Incased wariness of financial intermediaries will reduce intermediated savings and ivsment, and thereby lower the efficiency with which society allocates resources. * Third, poor information about the managemet and performance of specific financial intermediaries will make it difficult for savers to evaluate and compare 4. Another way to see tbat fee markets wil tend to undermonitor financial intemediaries is to note, that to a sgnific degree, information about th managem and solvency of finanial instituons is a public good: many people can have this infonnaion at the same tine, and it is difficult to exclude others from using thc infbrmation. Goernment lnsuwce and Fuical Iwermffaries: Issues of Regukadon. E luion. and Monitoring 199 financial institutions and funnel their resources to those fmancial intermediaries best able to allocate capital efficiently. * Fourth, insufficient monitoring of financial intermediaries may prevent markets and institutions from arsing or severely linit their activities. For example, the mutual fund industry in the United States would probably not have blossomed in the last fifteen years unless investors could easily compare funds and bave confidence that there are minimal possibilities for fraud. * Fi, insufficient monitoring of financial intermediaries by the market may encourage concentrated ownership (or the emergence of a few large debt holders) of financial intermediaries, so that large owners and creditors find it worthwhile to monitor intermediaries. Concentrated ownership may produce suboptinal diversfication and thereby alter the products offered by ntermediaries and create incentives for financial institutions to behave generously toward large owners and creditors. Thus, insufficient monitoring of financial instimtions can reduce the provision of financial services and thereby loiwer investment, reduce the efficiency of resource allocation, and slow economic development Consequently, the effect of financial policies on the incentives for private investors and instittions to evaluate and monitor financial intermediaries must be carefully considered in evaluating and designing fi=ncial regulations and policies. In general terms, governments should create incentives for-and be sensitive to financial policies that create disincentives for-self-regulatory agencies and private monitoring arngements.5 Legal systms that defMe and enforce property rights efficiently will help creditors, rating agencies, and other institutions monitor fmncal intermdiaries while also facilitating the emergence and fumntioning of self-regulatory bodies. Similarly, sound accouTing standards and information disclosure laws will facilitate the ability of the private sector to monitor financial intermediaries. If self- regulatory and pnvate arrangements do not adequately monitor financial intermediary activities, the governments should: seek to improve private sector monitoring through legal, accounting, and other reforms; review and reform financial sector policies that are impeding or creating disincentives for private sector monitoring of intermediaries; and help monitor financial institutions direcdy. Governments ofren set capital adequacy requiremnts, monitor the asset quality of intermediaries, establsh reglations regarding the liquidity of banks, set limits on large exposures, restrict intermediary financing of intermediary managers, owners, or related parties, monitor wading on insider information, and carefully regulate the entry of new participans. Conducting effective supervision and designing appropriate incentives for self- 5. Regarding incenfives for self-regulain of scurities marlxts, see T. Glaessner (1992). 2zo RossLevie regulation, however, are very difficult Some of these details will be discussed below. The essential point in the present context is that one critical criterion for evaluating fnancial policies is how they affect the monitoring of financial institutions. Eixeralities in Failure Failure of an individual financial intermediary may produce negative effects that extend beyond the creditors of that individual institution. There may be contagion: given poor information about the solvency and performance of financial institutions, investors may interpret the financial weakness of one finmaneal instution as a signal of the poor condition of other financial insttons and withdraw their funds from all intermediaries. The pottial social costs of this "contagion" are not intnalized by individual financial intermediaries when making decisions. Thus, with imperfect information about the financial condition of intermediaries on the part of savers, the social costs of a single financial institution failure wil often be greater than the private costs since one failure could reduce saver confidence in other institutions; there wfll exist external costs to excessive risk taldng any single intermediay because of contagion. Because financial intermediaries stand at the focal point of economic activity, and because they mobilize savings and allocate resources to all sectors of the economy, and form the foundation of the payments system, excessive nsk taking by them will have lr negative costs: * First, fears of contagion and financial instability reduce confidence in financial intermediaries and lower intermediatd savings and the efficient provision of crucial financial services. This would tend to slow economic growth and impede improvements in welfare. * Second, contagion disrupts economic activity. Fcr example, bank runs cause banks to demand payment of existng loans and to stop the issuance of new loans, which reduces investment, induces bankruptcies, raises unemployment, and slows growth. In addition, contagion, by inducing a contraction in credit and the medium of exchange, can often disrupt a country's payments system and thereby impede all forms of commrce. Similarly, runs on mutual funds depress assets prices, while uns on insurnce companies reduce risk sharig. * Third, given the pivotal role of financial services, financial intermediary failure on a large scale can negatively influence long-nm, overall economic performance by disrupting the flow of fnancial services. These disruptions are particularly acute in the case of financial institutions, because bankruptcy of financial institutions is different from bankuptry of most other entities. Govenants Insurance and Fmmcial Imennediaries: Issues ofRegWation, Evaliona and Monitoring 201 Financial intermediaries are primarily involved in the production, assessment, and dissemination of information. This information capital is not easily transftred. Thus, bankruptcy of financial institutions will entail the loss of a very valuable resource-information-because it cannot be transferred in bankruptcy court. The loss of financial intermediary information on a large scale through bankruptcy will significantly diminish the ability of society to allocate resources efficiently for an extended period.6 Thus, if risk taking by individual financial entities has external negative costs becam of contagion, governments should be concerned that financial intermedaies will undertke socially excessive levels of riskL Note, individual financal intermediary failures per se will have very positive effects if failure, or the fear of failure, encourages privately organized insurance and crisis mangement systems, self- reglation, and beter monitoring by creditors. The extemrnal social costs of an individual financial intermediay failure stem from potential contagion poor information may induce individual failures to spread to otherwise healthy intermediaries. The importance of contagion and financial fragility to economic acdvity suggests that contagion and financial stability should be part of our select scheclidW of issues to consider when evaluating financial policies. Potentil policy (and market) responses to exteralities associated with financial intermediary failure fall into four categories. Fist, governments may enact financial policies that encourage prudent risk taking by financia miaries. This may include resticting the activities and investments of fiancial institutions, establishing high, risk-based capital requirments, and even limitng competition. Second, govermnents should be wary of tax systems that create incentives for high debt-equty ratios which may enhance enterprise and financi fragility. Third, governments should avoid financial polices that augment the possibility of contagion. For example, evidence from the United States suggests that finanial policies can encourage the proliferation of an excessively large nmber of underdiversifled banks that are more sensitive to economic shocks and less able to organize private nsurance and self-regulatory mechanisms effectively and coordinate constructively 6. Evidence in the United Staes suggests tbat bank failure, or bank distress more generally. hams client firm because of the intrinsic, ftough nonadable. long-mn relatonships that develop between banks and their customers. Increases in the probability that the bank-bormwer relationship will be negatey disrupted reduces the share price of the borrower, while the rescue of a bank expected to fail increases the stock price of client firms. See Slovin, Sushka, and Polonchek (1993). 202 Ro Levine when faced with problems.7 Fourth, governments may insure investor assets in fincial intermediaries to prevent contagion. Imperfect Competition There are largt fixed costs associated with evaluating firms and monitoring activities. The costs associated with obtaining and maintaining accurate information on firms create mcentives for financial intermediaries to establish long-rmn relationships with firms: itermediaries will be able to recoup the costs of spending resources to acquire and update information over long periods, and firms will be able to access cheaper and more secue financing. Indeed, because information is both imperfect and costly to obtain, a firm's current bank will probably have more information about the finr than other banks, so that if the firm seeks financing from a different bank, this search will probably be viewed as a negative signal about the quality of the firm. Thus, an inherent characterisfic of many financial arrangements-the high fixed costs of acquiring and maintaining accurate information-implies that financial markets are lely to be imperfectly competitive.8 The optimal degree of competition among financial intermediaries is difficult to specify, so that the dict economic implications of imperfectly competitive financial markets produced by the high costs of acquiring information on firms are difficult to quantify. Insufficient competition and contestability will reduce innovation and efficiency. Similarly, to the extent that imperfect competition creates excessively symbiotic links between financial intermediaries and firms, the objectivity with which Intermediaries evaluate firms will deteriorate. By contrast to the extent that an oigopolistic financil system reflects the buildup and maitenance of long-run relationships that encourage an efficient exchange of information and more complete monitoring, then apparently lax competition will reflect good monitoring of firms. One mechanism for limiting competition is through "franchise" value as explained by Caprio and Summers (1994). By restricting entry, officials can create monopoly profits. These monopoly profits increase the value of having a license, the franchise value, and increases the costs of losing that license through bankuptcy. 7. For example, Calomiris (1989) notes that in the Ioos many states in the United States prohibited branch banking. This led to the emergence in those states of many, underdiversifled banks. The lack of diverfication increased exposure to idiosyncratic shocks. The large number of banks made it difflicult for banks: (a) to establish private deposit insurance funds with sound monitoring mechanismsand,b) coordinate effectively when confronted with a drop in depositor confidence. In fact, states that restricted branchg suffered more banling failures than states with fewer, better diversified banks. 8. in a free market, sellers are wiling to sell to any buyer at a stated price. Because of informational asymmetries, this is not true in all financial markets. All firms cannt borrow from banks at a stated intrest rate. Typically. a firm can only borrow from a single bank at a publicized interest rate. Trying to move to anodter bank involves a costly and risky process of establishing a relationship with new bank. Goeniet urnce and Fiwnancial Intennediadies: Issues of Regudation, Evalution, and Monitoring 203 Consequently, financial policies that increase franchise value tend to decrease competition while simultaneously decreasing risk taking. Thus, the effect of financial policies on the level of competition must be carefilly considered and therefore forms one of our "checklist" items. Supervision and regulation need to balance the negative and positive aspects of competition in the financial sector so as to maximize the efficient processing of information while minimizing arrangements that thwart innovation and encourage excessive risk taling. While easy to say and hard to accomplish, it is a balance that should be kept in mind when evaluating financial policies. IncompLete Markets Financial institutions are in the business of obtaining and processing information, but there are informational asymmetries in their business. Information obtained by evaluators is imprfect relative to the information known by the entity being evaluated. Information asynmmetries imply that some financial arrangements will be limted or nonexitnt-even though these arranents would exist in the absence of informational asymmetries. The following examples will help clarify this point. * Credi rationing-If it is difficult for banks to obtain accuate information about the nskiness of firms, raising mterest rates may cause firms with the safest projects to drop out of the loan market, so that the mix of firms demanding loans becomes more risky. Thus, raising interest rates may cause an adverse selection of firms in the loan market (without changing the management of projects). Banks may then keep intests rates lower hn the rate that would clear the market to maintain a safer mix of firms in the pool of firms demanding credit. This low intrest rate will produce an excess demand for loans by firms; the adverse selection problem reduces the issuance of loans below what it would be in the absence of asymmetric information. If it is impossible to monitor the behavior of firms perfe ty, raising interest rates may induce project managers to change their behavior and undertake riskier projects. To mitigate this moral hazard problem, banks may keep interest rates lower than would clear the loan market. Consequently, there will be an excess demand for loans by firms. * Equity issuance-If fim insiders have more information about the firm hn outsiders, then the issuance of new shares by insiders will be perceived as a negative signal by outsiders: insiders willing to sell shares to outsiders must think the price is high. This informational asymmetry between insiders and outsiders will discourage the raising of capital through equity issuance and 204 Rosa Levne therefore reduce the usefulness of the stock market as a vehicle for raising capital and diversifying risk. * Incomplete insurance-Insurance creates incentives for the insured to do less to avoid the insured-against event. If information were fully available and monitoring were costless, the insurance agency could prespecify a comprehensive, complex list of actions and behaviors for the insured that would eliminate this moral hazard problem. But, all actions cannot be monitored. Thus, insurers will provide less than complete insurance to enhance Lhe incentives for the insured to avoid the insured-against event. The result is incomplete insurance because of information asymmetries. Voluntary deposit insurance schemes will be difficult to organize because of informational problems inherent in evaluating and pricing the rskiness of banks. Good banks may not want to join deposit insurace schemes because they are unsure about the asset quality of other banls; good banks do not want to subsidize bad banks and it may be too difficult and costly to evaluate other banks and set risk-based deposit insurance premiums. This adverse selection problem induces a deterioration of voluntary deposit insurance schemes. The above examples illustrate cases of incomplete or missing markets that involve a reduction in financial intermediary services from the level of services that would exist in the absence of information asymmetries. Thus, investment and resource allocatioa may be suboptimal because of credit rationing and built-in incentives against the raising of capital through stock offerings. Similarly, moral problems created by nsurance may imply incomplete insurance markets that yield less risk reduction opportunities. Finally, because of adverse selecdon problems, certain types of private insurane, like voluntry deposit insurance and other self-regulatory insurance schemes may be provided at a suboptimal level. For these reasons, policymakers should consider the effects of fiancial policies and regulations on the types of financial contracts and services offered by financial markets to the public9 There are market mechanisms that rely on the legal minstructure to help reduce these asynmnetries and their negative economic effects. htermediaries and firms build long-run relationships that facilitate information exchange. The ability to use collateral in loan contracts helps extract information from borrowers that reduces informational asymmetries and expands the availability of financial services. Also, efficient legal systems and registries permit creative financial contracting to service the needs of 9. Caprio (1992) analyzes die inerctions of financial reform and asymmetric information. He concludes that die long-run relationsiups that form between financial intemediaries and firms to mitigate informatonal asymmeties must be carefuly considered in desning financial reforms, so that the negative economic implications of asymmetric information are not unecessarily aggravated by reform. Government Insurance and Financial Intemwdiaries: Issues of Regulation, Etaluarion, and Monitoring 205 clients, and sound bankruptcy courts enable intermediaries to seize and dispose of assets of delinquent borrowers quickly and confidently, which further promotes the provision of financial services. Finally, financial systems that allow a single financial intermediary to engage in different financial contracts, like issuing loans and buying equity, may be able to establish relationships with firms that maximize information exchange, thereby reducing the negative effects of asymmetric information. Nonetheless, these informational problems frequently cannot be eliminated. Market Failures-A Summary By studying important characteristics of financial markets, two general points emerge. First, from a policy perspective, the existence and economic importance of market failures suggest that governents should consider selective interventions in fnancial markets. This conclusion is based on the following argument: the market faiures analyzed above suggest that free markets will not produce a socially optimal amount of financial services; and these market failures have negative economic implications. Although the market may respond to these market failures by yielding contractual and fiancial arrangements that mitigate the negative consequences of the five market failures, the inherent charactenstics of financial markets imply that some of these market failures are important in all countries. Second, regardless of the hypothetical importance or unimportance of these market failures in an umegulated environment, countries typically have a complex network of financial policies and regulations. Thus, changes in any particular policy or regulation must be carefully studied. The checklist outlined in this section will help organize analyses of financial policy reforms by providing analysts with a vehicle for comprehensively considering the effects of policy changes on the provision of crucial financial services. Analysts should evaluate whether policy changes on balance aggravate or ameliorate the negative effects of market failures. Specifically, financial reforms should be evaluated in terms of how they affect (a) finmancial intermediary monitoring of fim, (b) private and public sector monitoring of financial intrmediaries, (c) the possibility of contagion and financial stability, (d) the degree of competition, and (e) the spectrum of available financial arrangements. Insurance and Financial Ilntermediaries A basic economic rationale underlying government organized insurance of financial intermediary liabilities stems from externalities associated with financial 206 Ross Levine intermediary failure (for example bank runs).10 Government insurance, however, then creates an entirely new ecouomic rationale for further government involvement in the financial sector: government insurance of investor assets in financial intermediaries reduces the incentives for investors in financial intermediaries to monitor the health of those intermediaries and for mtermedianes to self-regulate one another to prevent failure and contagion. This tends to increase the incentives for and the ability of filaancial intermediaries to undertake more risky activities. This is the moral hazard problem. Frequently, this moral hazard problem cannot be avoided because the public believes that the govermnent would isure their savings in the case of a financial nsis. This expectation alone, even in the absence of explicit government insurance, creates the moral hazard problem, because expectations of government insurance reduce incentives for investors to monitor financial intermediaries. The extent of this moral hazard problem in each country, therefore, depends on public expectations on the role of the government. Since financial sector policy choices depend on the extent of the moral hazard problem, financial sector policies should contain country-specific elements that reflect these differing expectations. Government nsrance There are many diferent ways to organize government insurance. Govermnents may insure all asse, a percentage of assets, or only relatively small asset holders. The design of insurance schemes may also differ. For example, governments may attempt to charge a market price for the insurance they provide based on the nskinss of particular financial institutions. Or, governments may charge a simple fee that is not market or risk based. Furthermore, governments may require financial institutions to insure themselves with private insurers that are authorized and monitored by the government. The exact nature of the insurance scheme will have important implications for the behavior of financial ntermediaries. For simplicity, first consider complete government insurance in the absence of risk-based premiums or risk-based capital requiements to get the blunt, first-order effects of government insurance on the five market failures defined above, and then broaden the discussion to more sophisticated schemes. This paper, however, does not comprehensively review the design of insurance schemes. It identifies a few features associated with insuring saver assets, noting a key political economy issue associated with insurance, and extracting some important strategic considerations concerning the linkages between insuring saver assets and the expansion of banking powers. 10. Talley and Mas (1993) provide aa excellent analysis of deposit insurance. The discussion here draws Liberally from their insights. GaMenmn t insrrace and Finanril beredrwins: Issues of Reguaio, Evaluton, and Monoring 207 Contagion and Financal Stability Insurance of assets in financial intermediaries will tend to reduce the probability that the failure of one institution will spread to other institutions; credible government insurance lowers the probabiity of contagion. Nonetheless, insurance of intermediary liabilities increases financil fglity by generatng incentives and capabilities for excessive' risk taking on the part of intermediaries. This is discussed below while considering the effects of insurance on the monitoring of intermediaries. Evaluon and Monitoring of Intermedaries The most important consequence of insuring investor assets is that it increases the incentives for risk taking by financial in ediaries, a socially subopthal monitoring systm. If investor assets are credibly insured, they wil have fewer incentives to monitor the intermediary, and will provoke risk taking by financial institutions. For example, the capital strength of banks is a way of signaling depositors about the security of the bank. Deposit insurance lowers the benefit of maintining high capital standards to reassure depositors. Thus, banks with deposit insurance have icentives to reduce capital/asset ratios. Furthermore, banks with deposit insurance have greater incentves to lend to riskier clients than banks with no deposit insurance. Bank owners keep most of the benefits from lending to clients with very risky but potentially high return projects, but if the risks do not pay off and the bank fails, some of the losses will be passed to the government insmrance fimd. This incentive to gamble with insured deposits is itensified as the capital position worsens. Thus, the combiation of greater mcentives for financia intermedies to assume risk and lower incentives for pnvate creditors to monitor financial inrmediary behavior implies that government isurance dramatically augments the need for mechanisms to intensity monitoring of inTmediaries.11 As an aside, it is worth noting that govemments can promote better monitoring of financial intermediaries through a combination of mechanisms. Governments can monitor financial interdrie directly. The govenment can restrict the activities and investments of fman intermediaries to promote prudent behavior; require intermediaries to hold well-diversified portfolios to reduce exposure to idiosyncratic shocks; review the owners, mangement ad organization of intermediaries to enhance the soundness of financial insfttions; and use risk-based capital requirements and risk- based insurance premiums to create appropriate incentives for interdiaries. Also, governments may insure only small investors, which would maintain incentives for 11. See Dowd (1993). 208 Ross Leince large investors to monitor intermediaries.12 Besides direct government supervision and reguaion, governments may require and use audits from internationally reputable accouting firms and assessments by rating agencies to engage a diverse network of financial experts to evaluate financial insutiuons. Furthermore, by increasing the francise value of insured intermediaries through reduced competition, governments reduce inceives for risk taing and thereby counterbalance enhanced incentives and opportunities for risk taking created by government insurance. Fmally, the government can also encourage or even require greater private sector participation in monitoring financial institutions to bolster oversight of fmancial institution activities. Some aspects of private msrance will be discussed below. The literature on deposit insurance is enormous and advances a myriad of scbemes. This paper only mentions a few approaches and focuses instead on the fist-order effects of government insurance on financi market behavior. Most import, governet insurance tends to reduce private creditor monitoring of intermediaries. Evauton and Monitoring of Firms Govermuent inrance tends to reduce the intnsity with which insured fiancial intermediaries monitor finns.13 One way in which i iane compete for funlds is by havmg a reputantion of carefully momtorng the fims in which they inves Careful m itoring by the itermediary lowers the probability that the intemiediary will experience losses. Thus, in the absence of insurance, safe ineediaries should be able to raise funds less expensively than intermediaries who do not monitor firms intesively. Investors with assets in insured intermediaries, however, are less concemed about losing their savings than if these savings were not insured. Therefore, in the presence of insurance, intermediaries have less of an incentive to monitor finms carefully, invest in a diversified portfolio of relatively safe firms, and cmmunicate this information credibly to savers. Competition Among Financil Inermediaries Insurance will also influence the level and form of competition. For example, with deposit insurance, depositors view banks as closer substitutes than without deposit 12. Incomplete insuce, however, wil not eliminate contagion under all conditions. For example, the best informed and biggest invests may still view imperfect signs of financial weaknmess in any pardcular instition as sufficient information to withwa funds from other financial insdutions. 13. This assumes that inuance premiums and cap requirements are not risk based and that regulaion and supevson of intenmediares do not fiull compensate for the ienives created by isman G;owra em neluwawc adFwinanolbermeIaies: lsesofRegultiEvauaion,andMonitoring 209 insmrance. Banks will compete in terms of serces and interest rates, but banks will have fewer incentives to tmit information to depositors about the quality of their loans. In the absence of insurance, however, intermediaries would need to convey information about the quality of their assets to attract investors. Insrance may also change the overall level of competition. For example, in an oligopolistic banking system with a few large, well established banks that are able to self-regulate each other and self-organize a deposit insurance system that excludes other banks, the innroduction of government deposit inurance for all banks may increase the level of competition for deposits in the system. Thus, it is difficult to draw clear conclusions about the effect of government insurance on the level and form of competition. Incomplete Markes and Avaiable Fuancl ServWces Mandatory govent insurance may be a mechanism for overcoming a market failure. As discussed above, missing insuance marets often exist because of adverse selection: if the costs of acquiring information and monitoriog other intermediaries and desigmig risk-based inurance fees are very high, safe financial i ries may opt out of privately organized, voluntry insmrance schemes because they do not want to subsidze more risky indiaries. The existce of a large mmber of banks, for example, will tend to aggravate the adverse selection problem by maling monitoring of banks and coordiatig a voluntary, pnvate isurance system more difcult. Thus, voluntary msurance systems will tend to deteriorate when adverse selection is particularly acute, which would not occur with mandatory government insurance. Political Economy and the FallaK of Choice While reducing the potential for contagion, insurance of intermediary liabilities tends to aggravate the suboptimal evaluation and monitoring of insured financial intermediaries. This negative consequence has led some analysts to argue that the costs of insuring investor assets in intemediaries are gater than the benefits, and many ommend abolishing or avoiding government organized insuane. This is typically not an option. Given the huge macroeconomic mplications of financial failure, governents wil typically act to prevent individual financial intermediary failures from speading and becomning systemic failures. Many governments have insured assets when faced with fiamcial failures evea in the absence of preexisting comnitments and even after stating beforehand that the govermment would not insure assets in the case of financial failure. Thus, most governments camot credibly commit to not interfere in the presence of financial failure. The belief by the public that the govermment insures their 210 RAW Levin investments creates the moral hazard problem: there is a reduced incentive to monitor financial institutions on the part of the public because people expect the government will insure their assets in the case of failue. Thus, the policy cnoixe between govermnent insurance and no govermnent insurance is generally irrelevant. The more useful concerns are explicit and implicit insurance and the level and dcesign of the coverage. Different societies expect diferent levels of government insurance. Some societies expect governments to protect the assets of small savers in banks, others expect deeper coverage (for example, insuring large bank accounts), while some societies expect the government to insure a broader set of instiuons such as insurance policies, private pension accounts, and even the reurns on investment company assets. The extent of the moral hazard problem produced by insurance is a by product of these expectations and therefore also depends on political and historical ingredients that vary from country to country. Public expectations about the coverage of govenmment msurance may also change systematically with financial development and even respond to changes in nancial policy. Two examples will help illustrate these points. As a country's fial system develops, households may shift their assets out of insured demand deposis mto uninsured, relatively unregulated money market accounts. As this shift occurs, public expectations concering govermment responsibility toward money market accounts may change. Specifically, expectations may expand to include money market accounts under the umbrela of governent inuance. Thus, the undertnding between the public and government is critical in designng financial policies, becaue this understndig, or social contract, between the public and the govemnment concering what saver assets are insured by the govenment determines the depth and scope of the moral hazard problem. Usmg finncia regulation to limit the scope of government insurance is a complex task. For example, some analysts propose the creation of 'narrow' banks.14 These narrow banks would enjoy 100 percent deposit insurance and would be the only institutions tied to the nation's payments system These nrow banks would also be very resticted and tightly regulated. They could only make loans that were almost risk free; they would not be permitted to assume intest rate risk; and they would face high capital reqirements. Thus, households would have a safe place to save with correspondingly low returns. If savers seek higher returns, they could invest in unmsured financial institutions. This would limit the moral hazard problem created by insrance, because the itermediaries receiving governmn inrance would be tigbtly restcted and supervised. One problem with this scheme is that it may not be compatible with public expectations. While in some contexts the population may adjust its expectatons of the scope of governmen insurance to this new regulatory sucture, this may not occur in all countries. For example, savers may believe the government 14. See dte discudos and chaziom in Talley (1993b). GovernenffnsrnCe andFaoFancal emenndaries: Issues of Regtidon. Edation MandMoniong 211 would also insure intermediaries that are not narrow banks if faced with a financial crisis. This expectation would lower incentives of private creditors to monitor non- narrow bank financial intermediaries carefully. Thus, the narrow bank scheme may not control the moral hazard problem. One generic conclusion that emerges from this analysis is that in counties where relatively broad and deep coverage of financial assets under the umbrella of government insurance is expected, there will be correspondingly greater undersupply of monitoring of financial intermediaries by investors. Therefore, the greater are public expectations of a govemnment safety net, the greater is the need for financial policies that strengthen incentives for monitoring of insured intermediaries. Government Insurance: Implicit or Explicit? Authorities could forgo a formal government operated insurance system. Instead, the government could intervene following a financial failure. This "implicit' insurance has been used in many countries. It must be emphasized, however, that implicit insurance does not avoid the moral hazard problem created by pubhic expectaions of govermment insumane; using implicit instead of explicit insurance does not circumvent the reduction in monitoring of intermediaries by creditors created by public e ion of government insrance. Implicit insurance provides flexibiy in terms of the amount and form of protection since preexisting rules and procedures restrict decisionmaking. Nonetheless, on balance, explicit insurance generally has advantages over implicit insurance. Inplicit insmrance will often not offer the same stability as explicit isurance. Implicit insurance implies ad hoc, unsystematic procedures for coping with failures, does not foster the buildup of an insurance fund that could withsnd potentil financial crises, and therefore will not significantly enhance public confidence in the safety of their assets. Thus, implicit insurance will not mitigate the probability of contagion to the same degree as explicit insurance.15 Explicit insurance seems to offer greater opportunities and encouragement for government to enact forward-loolkng financial policies that bolster private and public sector oversight of fnancial intermediaries than implicit insurance. Although proponents of implicit insurance argue that greater uncertainty concerning government insurance enhances incentives for (a) private creditors to monitor intermediaries and (b) intermediaries to form private insurance and self-regulatory bodies, this argument relies on the assumption that uncertinty surrounding the extent and form of government insurance creates positive incentives for private sector monitoring of intermediaries that 15. It should also be noted tbbe publc expenditure effects from financial failure can be very large (as the savings and loan expeence in the United States demonstrates). Therefore, buildimg an insuraice fuid prior to a fure may mitigate the macroec=omic implicaions of a systemic financial failure. 212 Ross Levine are greater than the negative incentives for private creditor monitoring of intermediaries generated by expectations of goverment insurance. More important, under the premise that the public expects the government to insure assets in the presence of a financial crisis, governments will be abh: to counteract the moral hazard problem better with explicit insurance than with implicit insurance. By explicitly recognizing a social responsibility to insure some class of saver assets, govermnents will be able to design and enact forward-looking financil regulations that augment the monitorng of financial mtermediaries and enhance financial stability better than could be achieved with an implicit insurance system. For example, governments could use risk-based insurance premiums with explicit insurance, while dtis would send confbsing signals with implicit isurance. Moreover, credible, explicit insurance system may be able to limit public expectations regardin the size and set of financial instruments insured by the government For example, with explicit insurance and universal banks, it may be possible to limit the government safety net to small checking accounts. It may be impossible to draw this line ex post with implicit insurance. As mentioned above, isuance tends to reduce the intesity and the incentives with which iermediares monitor firms. Although choosmg explicit or implicit isurn does not appear to affect firm monitoring icentives differentally, explicit government insurance encourages better regulatory strategies. Specifically, in the absence of explicit goverunent insurance, govemments may believe there is not a moral hazard problem. Therefore, officials may not design financial policies appropriately, because they will ignore, or insufficiently weight, the distortions created by expectatons of government insurance. Explicit recognition of public expectations, and therefore govenment responsibilities, will permit and spur more prescient policies. Thus, explicit insurance would have a higher probability of generating regulations to inpel intermediaries to effectively r'nmitor firms than implicit insurance. For similar reasons, explicit insurance may prompt government to consider raising the franchise value of insured intermediaries. This would reduce risk taling and fuirther work to counterbalance the moral hazard created by government insurance. This would be recommended only if there was sufficient competition to spur innovation and efficient provision of financial services. Finally, in terms of the effects on the availability of financial arrangements, there does not appear much difference between explicit and implicit insurance. But using this paper's criteria for assessing the advantages and disadvantages of explicit and implicit insurance, there seem to be substantial advaages to explicit insurance Most important, explicit insurance entils prior recogition of the problems created by public expectations of insurance. Therefore, explicit insurance encourages forward-looking financial policies to mitigate these problems. Govw=nme I,surwawe and Funda Intermediaries: Issws of Regidaon, Evmaion. and Moiorug 213 Vignettes on Private Insurance kn alternative or complement to government organized insurLace is private insurance. For example, a group of banks could create a deposit insurance fund. If credible, Drivate insurance will limit the probability of contagion, stimulate self- regulation, and encourage financial stabiity. Private ins e also has advanages with respect to government insurance. Edward J. Kane has noted that, in the United States, government officials are slow to recognize the existence of a problem and also slow to take action to cope with the problem once it is recognized. Kane argues that the interests of government officials often differ from those of taxpayers. Taxpayers want to cope with bank problems in the least expensive way, while government officials want to project a favorable image of their capacity. Government officials, therefore, focus on minimizig the mwiber of failures recorded on their watch and assigning the blame for failures to others rather than focusing their energies on expeditiously minimiing the aggregate expense to the insurance fund. Private insurance may reduce some of these problems by establishing better incentives for the owners and managers of private insurance funds. Private iurance funds that eliminate fears of contagion will be difficult to organize, however, for a number of reasons. First, losses in a crisis could be larger than the reserves of the private insurance fund, in which case members would face two difficult options: to inject more capital, which could weaken otherwise healthy institutions and contrbute to the contagion; or not inject more capital, which could reduce public confidence in the pnvate insrance scheme and precipitate a spread of the cnsis. Thus, in many cases, private insurance may not be sufficiently credlble to lower ihe probability of contagion substantially. Second, if the public still believes that the government ultmately stands behind these private insurers, the public will not carefully evaluat whether the private msurance system has adeWat funding and staff. Under these conditions, private insurance will not substitu for government insmance in the case of a sufficiently large financial failure. Governnents will still have to choose between implicit and explicit government insurance and the precise fonn of the isur= system if they choose explicit insuance. Third, as mentioned above, adverse selection problems in participating in voluntary private insurance schmes will hinder the functioning of these voluntary schemes. For example, safe banks may not want to subsidize risky bankcs, and it may be very difficult, complex, and costly to set risk- based insurance premiums. Thus, safe badlks may opt out of private ismrance schemes, which they cannot do with government insurance. FinaIly, an alternadve to having banks (or other intermediaries) form their own private insurance system is to use ismurance companies to insure bank deposits. But, in most countries the bankdng system is larger than the insura industry, so that the isurance industry may not have the capacity to underwrite bank deposits. Also, if an insurance company canceled 214 Rows Lerinc the insuranc of an individual bank, this would tend to precipitate a run on that baenk Governments may not wish to give private insurers such powers. Private insuranc±, therefore, may only work in special cases. Where a few banks have very good historical reputations and have established sound mechanisms for self-regulation, the public may have confidence in their ability to self-regulate, insure deposits, and screen other intermediaries. Using examples from different states in the United States in the 1800s, Calomiris (1989) shows that in states where a few banks created self-insurance organizations with strong self-regulatory powers, and where each bankWs liability to the insurance fund was unlimited, there was much tighter monitormg of banks and many fewer bank failures than in odter states. (see footnote 16 on the next page.)16 Instead of viewig pnvate and public insurance as substutes, Kane (1993) views the two methods of insurmg saver assets as complements. He proposes that banks with insured deposits should be required to have iurance with a private company. These private insurance companies would have to be authorized, monitored, and reinsured by the government. While not solvig all problems, Kane argues that this type of instional arrangement would make evaluation and monitoring of banks more responsive to market conditions, because private insuers with their financial capital on the line would have incentives more aligned with those of txpayers than government supervisors who do not have their financial capital eposed. It is iportant to note, however, that even in Kane's proposal, the govemment ultimately stands behind the Liabilities of financial institutions. Bank Powers This section evaluates the costs and benefits of allowing banks to engage in activities that go beyond traditional banking activities, such as taldng deposits and making loans.'7 These broader activities include underwriing and dealing in securities, holding eqty in companies, operating imvestment and trust companes, and participating in the insurance industry. The focus will be on securities market activities and holding equity in nonfinancial firms. 16. Another example is West Germany. In 1966, a purely privae consortia of banks formed a mutual- type, deposit insurance fund. T!x Federal Assoiaion of Gennan Commercial Bans orgaized cross- monitoring of bans to ensure stability of the baig system In 1976. tibs purely private finurance fmd came under public sector regulation but admstration and moniorng are s11 organized and conduted by te private Federal Associaton of German Commercial Bank. 17. Conversely, this section could be viewed as evaluating the costs and benefits of restricting the actvties of finmail conglomerates to only dwose activtics associated with traditonal banking and forcig other activites to take place in distinct legal enuits. Govenmenr Insawe and Fmaitcial Intertediaries: Issus ofRegulation, Evalaidon, and Moaiwring 215 Monitoring of Firms Broadening bankdng powers should have generally positive effects on how the fnancial system evaluates and monitors firms, though there are reasons for caution. Expanding the array of services provided by a single entity may broaden and deepen the relationship between the intermediary and firm and facilitate the flow of information. Also, banks that hold equity in firms and sit on the boards of directors will tend to exert better corporate governance. On the other hand, intrmediaries with broader powers may have a greater tendency man more narrowly defined istitutions to become overexposed to a few firms and thereby lose their ability to objectively monitor corporations. Monitoring Financial Intermediaries The monitoring of banks will become more difficult as the activities of banks become more complicated. More important, complex financial entities that are engaged in a wide range of activities may be more difficult to monitor thm separate entities providing these same services. The complexity of measuring exposure to specific firms, industies, and geographical locations, and of evaluating the riskiness of the intermdiary in general may grow more than proportionally as the permitted activities of the intermediaiy grow. Similarly, broadening financial intermediary powers may encourage the development of larger, more powerful intermediarie The political influence that will likely accompany dtis power may also hinder the ability of officls to force full disclosure of information and supervise intermediaries effectively and objectively. The potential for conflicts of interest and insider manipulations will grow. Monitoring transactions between the fiacial intermediary and significant shareholders, directors, officers, and their important and relevant business concerns will be more difficult, because the intermediary will be involved in more transactions and more complex transactions. Similarly, there will be greater opportnities for conflicts of interest. For example, banks with problem loans to a client may try to extricate themselves by underwriting and selling shares to an investment company or trust account run by the bank itself. Or, banks may issue a bridge loan to support a new equity sale being underwritten by the investment banking arm of the bank and thereby use insured deposits to support lucrative but risky investment banking activities.t8 18. See Edwards and Edwards (1991) for examples of the activities of fincial and industrial groups in Chile. 216 Ross Levine Given the inherent additional difficulties associated widt monitoring financial intermediaries that are engaged in a broad spectrum of activities, establishing an adequate capital base becomes more important. There must be adequate capital to balance the social costs of intermediaries undertaking risky investments. This will also help maximize the incentives for owners to price risk appropriately and for owners to monitor intermediary managers carefully. Furthermore, this capital must be secure, so that m the event of financial problems owners cannot remove capital surreptitiously. The identity of owners also needs to be clear so that observers can monitor insider tradig and investments. Thus, even if industries are permitted to own financial intemediaries, there should be clear documentation of the physical people who ultimately exert control over the financial intermediary. Thus, policymakers weighing the expansion of banking powers should consider the ability of auditors, boards of directors, and government supervisors to monitor the activities of financial conglomerates. Where (a) external auditors are well-qualfied and independent of financial intermediary management, (b) the boards of diectors of fiancial intermediaries exert s-ound corporate governance over management, and (c) the ownership and capital base of financial intrmediaries are clearly defined and secure, broadening the permited powers of financial inrmediaries will have less of a chance of excessively burdening government supervisors. Financial Stability and Contagion Broadening banking powers may inreas or decrease the riskdness of financial intermediaries. The ability to engage in a broader set of financial activities may permit greater mech for diversification. On the other hand, securities trading and investment banldng activities tend to be more isky. If these additional rsks are not diversified away, intermediaes may undertake more risk as their permitted powers expand-9 The emergence of large financial conglomerates through the broadening of bank powers also implies that the failure of any single intermediary may create severe economic disrptions. Thus, stability of each intermediary becomes more important for te stability of the financial system as a whole as financial power becomes concentrated in fwer instiations. Furthennore, broadening the powers of financial intermediaries may expand fte set of financial instuments presumed to be insured by the government and thereby augment financial fragility. In particular, with a compartmentalized financial institntion (banks taking deposits and issuing loans, investment companies purchasing equity shares and bonds, investment banks underwritng security issuance, insurance 19. Note, however, that empirical evidence suggests that security affiliates' operations of bais did not deleteriously affcct the soundns of banks in the United States prior to the Glass-Steagl Act of 1933 that legally separated commercial baking from securities operations (White 1986). Government Insurance and Financial I bermediaries: Issues of Reglaaion. Evaluation, and Monitoring 217 companies writing and selling insurance policies, and so on.), ,vernments may be able to define the set of financial instents insured by the government more narrowly than with ficial conglomerates with broad powers. For example, govermments may be able to credibly insure bank deposits up to US$100,000 but not retums to investments in mutual funds. When all of these financial services are provided by a single financial institution, the government may find it difficult to isolate and insure only specific assets. Thus, complex financial conglomerates will attempt to extend the social safety net to as broad a set of financial instruments as possible in order to attract investors. This may place more and more risky assets under the government's insurance un rella. Extending the social safety net and expanding the moral hazard problem may generate avenues for excessive risk taking without a concomitant increases in the ability of governments to monitor intermediaries. Competition and Markets Broadening the permitted activities of financial intermediaries should increase competition in the short run but may decrease competition in the longer run. By allowing banks to compete against mvestment banks, securities companies, and insurance companies, broadening the scope of permittcd activities would initially increase competition. On the other hand, if economies of scale and scope are important, decompartmentalizing financial powers may eventually encourage the consolidation of financial power in the hands of a few large financial conglomerates. This consolidation of power could work to reduce competition and contestability. Furthermore, such consolidation would concentrate considerable economic and, therefore, political power in the hands of the few individuals that control these financial conglomerates. With broader powers, a financial intermediary will have more financial instruments available to serve clients and may. therefore, be able to obtain more information about clients; economies of scope from bundling several services wfll reduce information acquisition costs. Thus, finmcial intermediaries that enjoy greater flexibility in servicing clients and with more information about clients should be able to overcome-to a greater degree than compartmentalized financial intediaries-the informational asymmetries that cause incomplete or missing markets. With more financial instruments and more information, financial intermediaries should be able to provide better financial services, all else being equal. 218 Ross Levine Conclusions This paper studied five characteristics of financial markets that imply that free markets will produce an undersupply of financial services. Government interventions to improve the provision of financial services, however, often distort financial markets and hinder financial development. Thus, each country must find an appropriate mix of financial policies ffiat optimally balances the beneficial and harmful effects of government interventions. The paper proposes a checklist of five issues to consider when evaluating financial policies. Specifically, how will financial policies affect (1) financial intermediary evaluation and monitoring of firms by financial intermediaries, (2) private and public sector evaluation and monitoring of financial intermediaries, (3) financial stability and the possibility of contagion, (4) the degree of competition among financial institutions, and (5) the spectrum of financial arrangements available to firms and individuals. Using this checklist, the paper evaluates two financial policies: insurance of saver assets in financial intermediaries and the powers and activities in which particular financial intermediaries should be allowed to engage. Although country specific legal, historical, political, and institutional traits should contnbute to the determination of financial policies, the following broad conclusions emerge from the analysis. Governments generally must insure some of the public's savings. Historically, govemments have insured assets when faced with financial failures even in the absence of preexisting commitments and even after stating before the failure that the govemment would not insure assets in the case of a faiure. The public understands this social contract and therefore has fewer incentives to monitor financial institutions that it believes are insured by the govermnent. These expectations of goverment responsibility for insuring saver assets are country specific and depend on political, sociological, and historical developments. Thus, instead of debating whether to have or not to have insurance, the more relevant policy questions involve the extent of coverage and whether this insurance should be provided implicitly or explicitly. Explicit insurance is typically better than implicit insurance. Explicit insurance entails prior recognition that the government will insure some set of saver assets in the case of financial intermediary failure. Thus, expsicit insurance encourages forward- looling financial policies to mitigate the undersupply of monitoring of financial intermediaries exacerbated by government insurance of saver assets. The forward- loolkng policies that may accompany explicit insurance include the buildup of an adequate insurance fumd, risk-based capital requirements, risk-based insurance premiums, restrictions on insured intermediary investments and activities, the coordinated use of self-regulatory bodies, private insurance, trustworthy rating agencies, and reputable accounting firms to help monitor insured intermediaries, and Governrnent Insurance and Financial Intermediaries: Issues of Regulaton, Evaluation, and Monitoring 219 more intensive government supervision of insured institutions. The mix of mechanisms for monitoring insured intermediaries and bolstering their safety depends on the existence and usefulness of self-regulatory bodies, rating agencies, and accounting firms and on the institutional capacity of the government to organize, implement, and enforce particular supervisory strategies. Explicit insurance can often reduce the set of assets insured by the government. Since public insurance of public assets in financial institutions tends to reduce private seixar screening of intermediaries, governments should attempt to limit the extent of this moral hazard by limiting the spectrum of assets insured by the govermnent. Although there is an implicit social contract that some set of saver assets are insured by the government, and although this contract may vary from country to country based on public expectations of the role of government, governments may be able to limit the extent of the coverage by designing and publicizing a credible, simple, and explicit insurance scheme. This cannot be done with implicit insurance. It is dangerous to combine implicit government insurance plus broad banking powers. If policymakers want to expand banldng powers beyond deposit taking and loan making to underwriting and trading securities, buying equity, and managig ivestment companies and trusts, they should make explicit, and socially credible, commitments about which assets are insured by the government. rBroadening the range of financial instrumpnts offered by banks or putting banks into hoiding companies that offer a wide range of instruments may expand the set of financial instruments that the public believes are insured by the government. Financial conglomerates will have great incentives to exploit and extend the perceived social safey net to attract savers. An expansion in expectaions conerning the spectrum of financial assets insured by the govermnent will tend to reduce public monitoring of intermediaries. At the same time, authonties may not feel sufficiently compelled to expand supervisory capabilites because there is no explicit insurance! Thus, unless supervision and regulation of financial conglomerates are particularly comprehensive and effective, broad financial powers combined with an unclear delineation between insured and uninsured instruments will tend to produce incentives that yield great financial instability and suboptimal provision of financial services. Do not expand banking powers if private and public entities are not capable of rigorously monitoring the new finaril conglonmerates. Broadening the powers of financial intermediaries makes monitoring intermediaries both more important and more difficult. Monitoring becomes more important because intermediaries will tend to be larger, so that failure of any single instiution has bigger macroeconomic implications. Monitoring also becomes more important because complex fiancial conglomerates will tend to blur the difference between insured and uninsured assets. 220 RAssLeve This implies that financial intermediary creditors will have fewer incentives to monitor the intermediaries. Monitoring becomes more difficult because the institutions become much more complex. Thus, unless a solid base of private auditors, rating agencies, self-regulatory groups along with government supervision and regulation can adequately monitor fiancial intermediaries, broadening the legal powers of fiancial instituions may produce negative results. Govenumn ethswr and Fmaidat Interediaries: Issues of Regpdaion, Eahuoion, and Monitoring 221 References Baer, Herbert. 1985. "Private Prices, Public Insurance: The Pricing of Federal Deposit Insurance." Economic Perspectives, Federal Reserve Bank of Chicago (SeptemberlOctober): 45-57. Bhatracharya, Sudipto, and Anjan V. Thakor. 1993. Contemporary Banking Theory. Indiana University, School of Business: manuscript. Calomiris, Charles W. 1989. 'Deposit Insurce: Lessons From the Record.' Economic Perspectives, Federal Reserve Bank of Chicago (May/June): 10-30. Calomiris, Charles W., and Charles M. Kahn. 1991. -The Role of Demandable Debt in Strucuring Optimal Bankig Arrangements." American Economic Review 81 (June): 497-513. Caprio, Jr., Gerard. 1992. Po04cy Uncertaint, Informaion Asymeries, and Fiadal Intermeaion. World Bank: Policy Research Working Paper, No. 853, February. Caprio, Jr., Gerard, Izak Atiyas, James A. Hanson, and Associates. 1994. FunaciaL Reform: Theory and Practice. New York: Cambridge University Pess. Caprio, Jr., Gerard, and Lawrence H. Summers. 1994. -Financial Reform: Beyond Laissez Faire." In Dimitri Papadiniitriou ed., Funancing Prosperity in the 21st Centry. New York: McMillan Press. Dowd, Kevin. 1993, "Deposit Insurance: A Skeptical View.' Review. Federal Reserve Bank of St. Louis (January/February): 14-17. Edwards, Sebastan, and Alejandra Cox Edwards. 1991. Monetariwm and Liberalization: The Chilean Experiment. Chicago: University of Chicago Press. Glaessner, Thomas. 1992. 'Extenal Regulation vs Self-Regulation: What is the Right Mix? The Emerging Markets LAC Perspective." World Bank mamnsipt. James, Christopher. 1987. 'Some Evidence on the Uniqueness of Bank Loans." Journal of Financia Economzics 19(2):217-235. 222 RAoss Levine James, Christopher, and Peggy Weir. 1990. "Borrowing Relationships, Intermediation, and the Cost of Issuing Public Securities." Journal of Fiancial Economics 28 Nevember:149-72. Kane, Edward, J., 1985. The Gathering Criis& in Federal Deposit Insurance. Cambridge: MT Press. . 1989. "How Incentive-lncompatible Deposit-Insurance Funds Fail.? National Bureau of Economic Research Working Paper, No. 2836. . 1993. "An Agency-Cost and Bonding Paradigm for Analyzing Deposit- Insurance Reform.? Mimeo. Boston College. Kng, Robert G., and Ross Levine. 1993a. "Finane and Growth: Schumpeter Might Be Right? Quarternl Journd of Economics (forthcoming). King, Robert G., and Ross Levine. 1993b. "Finane, Entrepreneurship, and Growth: Theory and Evidence. " Journl of Monetary Economics (forthcoming). Lunimer, Scott L., and John L McConnell. 1989. Further Evidence on the Bank Lending Process and the Capital Market Response to Bank Loan Agreements." Journal of Finan Economics 25:99-122. Neuberger, Hugh, and Houston H. Stokes. 1974. -Germa Banks and German Growth, 1883-1913: An Empirical View. ' Journal of Economic HIstory 34 (September):710-32. Stiglitz, Joseph E. 1993. "The Role of the State in Finncial Markets. World Bank. Paper presented at the Annual Bank Conference on Development Economics. Sloving, Myron R, Marie E. Sushkia, and John A. Polonchek. 1993. "The Value of Bank Durability: Borrowers as Bank Stakeholders. Journal of Finance 48: 247-66. Talley, Samuel H., 1993a. 'Bank Holding Companies: A Better Structure for Conducting Universal Banking?" In Dimiti Vittas, ed., Finacial Reguation Changing the Rules of the Game. Washington, D.C.: The World Bank. Taley, Samuel H. 1993b. -Are Failproof Banking Systems Feasible? Desirable?" World Bank: Policy Research Working Paper, No. 1095. GovEnment Insurance and Fiuandad Inlmedlaric: Issues of Aeguldoni, Evuhodaon, and Monitorg 223 Talley, Samuel H., and Ignacio Mas. 1993. -The Role of Deposit Insurance. in Dimitri Vittas, ed, Financial Regulation: Changing the Rldes of the Game. Washington, D.C.: The World Bank. Tilly, Richard H., 1966. Financial Institutions and Indstriaiazation in the Rhineland, 1815-1870. Madison: University of Wisconsin Press. Vittas, Dimitri. 1993a. "The Impact of Regulation on Financial IItermediation? In Dimitri Vittas, ed., Fuiancial Regulaion: Changing the Rules of the Game. Washington, D.C.: The World Bank. Vittas, Dimitri ed., 1993b, Financial Regulon: Changing the Rules of the Game. Washington, D.C.: The World Bank. White, Eugene N. 1986. 'Before the Glass-Steagall Act An Analysis of the Investment Banking Activities of National Banks?" Exploraions in Economic HisTory 23 (January):33-45. 12 CAPITAL ACCOUNT LIBERAUZATION: THE PHILIPPINES EXPERIENCE Edgardo P. Zita, Deputy Governor, Central Bank of Phlppines During 1991-92 the Philippines implemented a series of historic changes toward the fuiberalizaion of the exchange regime. These changes have resulted in virUally full convertibility of the current account. At the same time the govenmen took major steps to liberalize the aital account Together, the measures ended over four decades of pervasive exchange controls. Liberalization of the exchange regime constitutes a key element of a broader package of stuctural reforms aimed at enhancing Phppine competiveness in the global marketplace. The stabilizaon of the economy has made it possible to mplement these reforms. This paper descrnbes the foreign exchange reforms, focusing specifically on the caital account It describes the evolution of exchange controls in the Philippines leading up to the present reforms, and it offers some prelim ry ae s of the impact of the changes and considers the policy dhallenges that now fce the autorities. The Evolution of Exchange Controls in the Philippines The govenmment first imposed comprehensive import and exchange controls in 1949 in order to conserve available foreign exchange for the reconstucton and rehabilitation of its war-damaged economy. Capital controls were mosVy in the form of outright quantiative restrictions implemented hbrough a stringe prior approval system administered by the central bank. Major liberalizaton began in the 1960s when the governmt lifted the requiremnt for prior central bank approval of foreign exchange transactions, inchlding capital transactions. Thus, authorized agent banks were allowed to servicforeign exchange ements at the prevailig market rate without cental bank clearance. The only notable restiction during thLis period of decontrol was the mandatory surrender reqirement, albeit at the prevailing market rate, for aU foreign exhange recepts. Initially, the llation measures of the 1960s had a strongly positive efect on the balance of payments. Rapid increases in import volumes and deterioratig terms of trade, however, eventualy led to an unsustaiable balance of paymens. Bdo the decade was out, the central bank was forced to reimpose coaols on certain foreign 225 226 Edgando P. Ziatlda exchange transactions. Such controls were a central theme of the 1970s and the 1980s with only slight modifications during that time. In the capital account, with the issuance of central bank (CB) Circular No. 289 (February 21, 1970) and subsequent amendments, all foreign borrowings were subjected to prior central bank approval. In general, the circular and its amendments established criteria to channel foreign borrowings to idenfified priority undertakings. Guidelines also established acceptable borrowing terms. To ensure enforcement of these controls and to ensure inward remittance of borrowing proceeds, the government enacted a system of debt registration. Without registration, foreign borrowings could not be serviced. The authorities adopted a similar restictive approach for foreign investments. All investments made after March 15, 1973 required registration with the central bank. Priority investments received preferential treatment with respect to the timetable for allowing repatriation of profits, dividends, and capital, while policies were tightly controlled and generally discouraged outward foreign investments. An important restriction on the foreign exchange activity of local banks was also placed in the form of a requirement for a balanced foreign exchange position.. Not all initiatives in this period were restrictive in nature; however, in July 1970 the authorities introduced the Philippine Foreign Currency Deposit System (CD). Nevertheless, permissible deposits by residents were initially quite restricted. In particular, deposits in foreign currency notes and travelers checks were not allowed. ([This was eased later in 1987.) Furthermore, loans that could be extended to residents were restricted. An offshore banking system was established in 1976. This allowed foreign banks to have a small window of operations in the Philippines at a time when entry of foreign banks remained blocked. The permisslble activities were, however, initially very resticted, being confined to foreign currency transactions that over time could margin. Despite the genernlly restrictve exzh=nge control system, the economy still succumbed to a serious balance of payments crisis that culminated in the declaration of a debt moratorium in October 1983. Essenially, a series of adverse external developments brought about the crisis, including LWO major oil shocks and an interest rate shock compounded by inadequate policy adjustments, especially m the fiscal area and in the exchange rate. During the ensuing crisis, exchange restrictions intensified to ensure the availability of foreign exchange for critical imports. Principal repayments on commercial and bilateral debt were suspended. In November 1983 the central bank also required all agent banks to surrender their foreign exchange receipts into a common pool. The central bank administered the pool and rationed payments in accordance with an administratively set priority imports including oil, cereals, and a few other critical items. Cawipa Acont Libeafzon: The Phiflppines Etrpeienc 2 By October 1984 a measure of stabiliy had been restored, and the central bank was able to once again hberalize the exchange regime using a an International Monetary Fund MP) supported standby arrangemen; through the lifting of emergency exchange contols. The reopening of the exchange market also ushered in the independent float of the peso. Thus, this system, which remains today, eliminated the central bank's role in determining the exchange rate, which it had done by announcing an interbank guiding rate and imposing a trading band. The banks could now feely quote their interbank buying and selling rates and establish a market clearing rate, albeit within relatively restricted trading rules. These rules primarily limited interbank trading to on-floor transactons that could last for only tirty minutes each trdig day. The previous trading day's completed transactions formed the basis for the reference rate announced by the Bankers Association of the Phdippines. These restrictive rules enable the central bank to intervene in the exchange market when necessary without requiring a huge war chest of reserves. A senes of rescheduligs and financing subsequently regularized debt service arrears to official bilateral creditors and commercial banks. Moreover, trade liberalization resumed in 1986 when the Import Liberalization Program was continued in two phases. Pbase I, underten from April 1986 to April 1988, hiberalizd a total of 1,229 import items. Meanwhile, another 533 items were deregulated under Phase IL Currently, about 94 percent of the total number of tariff lines have been hberalized. In additon, indirect tax reforms were introduced to eliminate most of the discriminatory aspects of the domestic tax strucure against imports. Fmally, the reforms signficantly trimmed down the scope of exemptions, thus decreasmg the variability of the tariff structure. Despite the virtual completion of liberalizaion on the trade front and the regularization of the debt service, no significant measures were taken since 1984 to further hlberalize the country's restrictive nontrade payments regime, nor were efforts made to remove the coercive elements of foreign exchange supply that tended to penalize foreign exchange earners. Moreover, trading rules that made official intervention in the market easier and more effective at the expense of thin tading contied to hamper interbank foreign exchange trading. All these features prevented the attainment of a truly market-clearing exchange rate. Addressing these lagging elements would form the basis for the major foreign exchange liberalization measures of 1991-92. Foreig Ehange Liberaliation: 1991-92 The Philippine economy was considerably set back by the debt crisis of 1983. Nevertheless, the long road to economic recovery commenced in 1987 under the Aquino govenment. But the recovery process was not smooth, and it lost momentum 228 Edgwo p. Thkda in late 1989. This, in part, was attributed to adverse political developments and a series of natmral disasters, but it was also clear that, notwithstanding some definite improvements, export growth potental had not been fully realized. Moreover, inadequate savings to support much-needed investments hampered sustainable economic growth . These two critical lagging elements-exports and investmens-were prime motivators for continuing the exchange liberalization process to eliminate remaining anti-export bias obstacles to foreign investment arismg from regulatory consiraints. The widely shared view that further liberalization was necessary facilitated the formation of a working political consensus in support of further decontrol against traditonal resistance. The improving economic conditions in 1991 also provided a good early oportnity for launching foreign exchange reforms. A new program with the IMF provided the framework for fiscal, monetary, and other financial policies that would enbance credibility in the market. The balance of payments had also strengthened considerably, with reserves building up to more comfortable levels. Debt negotiations progressed well. On the political front, a measunr of nornalcy had been restored, following the last coup attempt in 1989, with scheduled elections set in 1992. While some doubted that the new administation would be filly committed to exchange reform, all indications suggested that such reforms were broadly supported by potentWi candidates Nevertheless, to address initial fears that a too drstic lementation of liberalization measures would quickly trigger a foreign exchange crisis, the reforms were staggered throughout 1991-92. Liberaliation Measures Table 12.1 summanizes the series of liberalization measures in both current and capital accounts. These measures can also be grouped according to the following desired effects. First, exporters of goods and services were given fall freedom to dispose of their foreign exchange receipts as they saw fit, hence avoiding being a captive foreign exchange supply source. This reform is an acid test of the market clearing property of any given exchange rate and would work to ensure the best exchange rate to exporters, all other things being equal. Second, restrictions on payment modes oher then through letters of credit were Liberalized for export and import transactions. In addition, all central bank prior approval requirments on export transactions were lifed. Finally, reportng requirements were simplified. These various measures were all expected to reduce transaction costs on foreign trade. Third, quantitative restrictions on the amount of foreign exchange that may be purchased from banls for service payments were lifted. This was expected to further CapialAccoumrLberaiiwion ThePhilippinesExperience 229 reduce inpediments to business operations and encourage the general public to voluntarily convert dleir foreign exchange. Fourth, free trade of gold was allowed. This would encourage domestic production of gold, which is a major natural resource endowment. Fifth, exporters of goods and services, as well as domestic producers, resident and foreign currency deposit units (FCDU) depositors, and banks were allowed access to short-tem foreign cmrency loans for interbank transacdons from domestic FCDU rithout need of central bank prior approval. Short-term trade facility credits could also be granted to pnvate companies without prior central bank approval. This would be especially helpful in lowering worldng capital financing costs of exporters given the much lower interest rates on foreign currency loans. Exchange nsk would be minimal, as foreign currency earners would be naturally hedged. Sixth, filll and immediate repatriation of foreign investments, including profit remittances, was allowed to further encourage inward investments. In addition, all types of inward investments could be made without need for pnor central bank approval. Moreover, only investments whose fiutre fcreign exchange ents for capital repatriation and profit remittance were intended to be sourced from the banking systems (other than FCDU) needed to comply with cental bank regisadon and mandatory surrender requirements. This latter reform legitimized informal investments and broadened further available project financing options. Seventh, outward foreign investments could now be made without prior central bank approval if sourced from foreign currency deposits or nonbanks. Outward investments could also be sourced from the baning system for up to US$1 million per investor per year. This easing was expected to enhance export cability by faciLitating the establishment overseas of appropriate marketng lnks and alnc, facilitating portfolio diversification, and strengthening technology transfer potential. Eighth, commercial banks' long and short foreign exchange positions were now subject only to limits imposed for prudential reasons. Continuous interbnk foreign exchange trading was also permitted and was actually provided through the launching of the Philippine Dealing System, an on-line electronic trading system. The previous regulation only allowed on-floor trading and for only thirty mimntes each day. These changes were expected to lead to an exchange rate that efficiently reflected the views of major participants acting on all available information In some respect the current liberalization effort has been similar to previous efforts insofar as certain measures have worked to lift quantitative restrictions on foreign exchange demand. This has clearly been the case with the liberalization of service payments and foreign investment remittances. However, the main body of reforms-and this is really the distictive innovation of the present liberalaion episode-involved exchange liberalizaton acting chiefly through enhancement of the foreign exchange supply side. This has been achieved for foreign exchange earners either through lowermg transaction costs, lowering financing costs, ensuring efficient 230 Edgwrlo P. Zialda foreign exchange pricing, broadening financing options, or increasing flexibility of capital expenditure decisions. Thus, the current liberalization effort is much bolder and more encompassing than past efforts, although restrictions still remain. Remaining Restrictions In current account transactions, invisible payments have now been completely liberalized: all quantitative restrictions have been lfted and central bank prior approval requirements have been removed. This is also virtually the case for import payments. However, in the case of service payments, it is still not possible to waLlc up to a bank and ask to buy foreign exchange with no Westions asked. One would have to show to the agent bank proof of the legitimacy of the transaction. Some restrctions still apply on key capital accounts. Proceeds of foreign investments and loans still have to be surrendered to the baning system, albeit at the prevailing market rate, in the context of registration with the central bank of a loan or investment Without registration, neither future debt servicing, capital repatriation, nor profit remittance can be sourced from the banling system. However, and this is the major change at this time, they can still be legitimately sourced from FCDU accounts and nonbank sources. Furthermore, the hiberalized treatment of outward foreign investment is still subject to a quantitative limit of US$1 million per investor per year if funds are to be sourced from the banking system, although no limits are applicable if fimds are to be sourced from outside the banking system. These remaining restrictions need to be reviewed further and can be liberalized dependig on the situation. In particular, there may be a case for fully liberalizg outward investments to help offset the large inflows of foreign exchange remittances into the formal economy. These inflows, now taldng place, are proving difficult to absorb into the domestic economy in the short run. Moreover, it may be necessary to reconsider the need to require surrender of investment and loan proceeds for these to be later eligible for servicing. These transactions could be permitted if transaction costs were further reduced. The authorities are also seriously considering liberalizing the entry of foreign financial institmtions to further enbance the efficiency gains arising from the liberalization of capital accounts. Such a measure can deepen the domestic capital market by mobilizing more foreign capital efficiently and by speeding up innovation in financial technology. The Early Results of Liberalization The exchange reforms are quite recent, with many important reforms only completed in the second half of 1992. Thus, it may be too early to firmly determine CapitatAccount Liberalization: The Plhilippines Experinffe 231 the efficacy of recent measures. But a variety of economLic indicators suggest that the changes are making a difference compared to pre-reform trends. The most visible impact is on service transactions. Major invisible receipts, notably travel receipts, combined worker remittances, personal transfers, FCDU withdrawals in pesos, and the residual other invisible receipts, have surged since late 1991. This can be clearly seen in figure 12.1. The comprehensive exchange reforms have created incentives in the form of market clearing exchange rates and better assurance that foreign exchange will be readily accessible in the future ff it is needed for residents to bring foreign exchange into the formal economy. The enhanced confidence in the economy, as suggested by trends in service flow transactions, seems to be supported by analysis of the developments on the balance sheets of the nonbanlcing sector. Figure 12.2 shows the overseas deposits of residents, as reported in Inttional Financial Statistics (IFS), as a ratio of the total resident deposits (n pesos and peso equivalent of foreign currency deposits). The overseas deposit ratio shows a sharp drop in relation to total deposits from 1992. This trend may suggest a fundamental portfolio shift by the nonbank private sector in favor of resident financial assets. This point, however, cannot be too strongly made because there are too few observations and because the relative reduction in overseas deposits could also have been temporarily influenced by nominal and real interest rate differentials favoring peso assets (figure 12.3). Nonetheless, such substantial differentals existed in pre-liberalization periods without making any apparent strong impact on the asset portfolio distrbution of nonbank residents. There is clearly a need to carefully monitor this development, especially because it may significantly affect monetay policy settings. By looking at the balance sheet of the foreign currency deposit unit, one might see the beginnings of a significant increase in FCDU resources driven largely by a marked upward shift in resident foreign currency deposits in 1992 (figure 12.4). Two factors could account for this change. First, developments may reflect the reflow of ffight capital, albeit not all the way into the economy, because such capital is still parked in foreign currency in FCDU rather than being inwardly remitted for pesos. Second, the relaxation of the export surrender requirement has led to the diversion of part of the export proceeds to foreign currency deposit (FCD) accounts, thus reflecting exporters' portfolio preferences. By closely matching the expansion in FCD resources, FCDU loans to residents are also accelerating. This is reflected mainly by the inceased access of exporters as a result of recent liberalization measures. The combined trends of more demand for resident loans and increased deposit-based funding from residents would have significant implications for monetary and credit policies as the FCDU system expands and plays a bigger role in financial intermediation. More generally, the foreign asset holding behavior of local banks appears to have shifted significantly in 1992 as a result of market liberalization. Specifically, the 232 Edgardo P. Zialcita relaxation of rules on net open foreign exchange may have led to a decline in liquid foreign asset holdings by commercial banks by about US$0.6 to US$0.7 billion, compared with the average for the period 1988-91. This behavioral change may well have been a shock factor in strengthening the peso in 1992. On the other hand, the balance on foreign currency transactions in relation to the volume of foreign currency tansactions, as determined by the value of private imports of goods and services, appearto have remained stable, notwithstanding market liberalization. There are still other interesting developments that probably reflect the impact of reforms. The volume of interbank foreign exchange trading has escalated sharply, beginnig in 1991 (figure 12.5). In such a free environment the central bank has been participating heavily as gauged by large shares accounted for by the central bank in total interbank transactions (figure 12.6). This, in turn, has led to the rapid accumulation of central bank reserves, as shown in figure 12.7. This high degree of activity has been largely motivated by the desire to slow down the tendency of the exchange rate to appreciate so sharply the reforms began. The concern about a potential upsurge in foreign exchange expenditures that could lead to a foreign exchange crisis as a result of lifting the controls has largely proven to be unfounded. To a large extent total import demand has been controlled through tight financial policies. However, it may also be observed that import propensity had not jumped up because of liberalization. Figure 12.7 shows trend developments in service payments, excluding interest, consumer imports, and non- consumer imports, respectively. as scaled by the GNP. There appear to be no major disruptions from past trends. For investments, a definite recovery has taken place in 1991-92 from the low in 1990 (figure 12.8). The return of portfolio-type investments, including net bank interbranch investments, has been the main factor underpinning such recovery, because net foreign direct investments, including net bank interbranch investments, have remained stable since 1989. Direct investments have consistently served as a solid core of foreign investment activity even during periods of adversity. Nevertheless, the nagnitude of direct investnents has not yet shown clear signs of more rapid growth, which has been disappointing. Part of the expansion in net inward investments of all types in 1992 have been somewhat offset by significantly higher outward investments. This is a rational consequence of the relaxation of outward investment rules and could represent a one- time adjustment. This trend, however, will have to be carefully monitored in the futre to ensure that developments do not assume abnormal proportions. Export developments continue to be a source of concern. For the moment exports have not sustained a clear growth acceleration. Recent developments are turning out to be rather disappoindng, with the exception of a slight improvement in export market shares (figure 12.9). Recent real appreciation of the exchange rate and weak international market demand, however, has likely dampened stronger export Capi:alAccoumtLiberaization: TihePhilippinesEFperience 233 response. More time will be needed to make a firm judgment on how liberalization efforts have affected export developments. Some Policy Issues A key policy dilemma facing Philippine authorities is the observed sharp appreciation of the exchange rate in both nominal and real terms. This may have implications on the continuation of exchange liberalization. A major factor in the recent nominal appreciation is the sharp improvement over previous trends in invisible receipts as compared side-by-side with sluggish economic growth. Substantial inflation in the Philippines, however, has been the prime cause of real appreciation, at least relative to major trading partners and competitors, despite marked progress in 1992. Not surprisingly, pressures from the export sector have been mounting. The cental bank has not been indifferent to such pressures and simply taken refuge in the equilbrating mechanism of the exchange rate. As a result, pan of the total potential appreciation has been resisted through direct intervention in the marketplace. Such intervention, however, has had to be sterilized in light of the requirements of the mrrent monetary program agreement with the IMF. Sterilized intervention has, in urn, created upward pressure on domestic interest rates, which has likely drawn in volatile short-term placements from abroad. Moreover, given the large stock of domestic debt to begn with and sharp increments due to sterilization, market intervention is going to have significant negative fiscal consequences. Thus far, the central bank has resisted administrative measures that would restrain, for example, the problematic inflow of volatile short-term capital. As a practical matter the volume of such inflows does not yet appear to have reached troublesome proportions. Moreover, on policy grounds, the current thining is that this approach not only departs from recent liberalization trends and sends the wrong signals to the market, but it may also become distortionary as private agents attempt to circumvent new restrictions. Moreover, a liberalized current account regime is bound to limit the effectiveness of capital controls. Another approach would be to strengthen further the fiscal stance. The near term prospects of this, however, are not very encouraging in light of competing demands, notably the need to develop infrastructure and power investments and to consider the current weakness of domestic demand. Perhaps the way forward is some measure of flexibility in monetary policy that could well be justified by increased demand for money as a result of lower inflationary expectations and increased confidence brought about by institutional reform. Inflation has declined sharply in the Philippines during 1991-92; the latest inflation in December 1992 was at 8.2 percent compared with a peak of over 20 percent in mid-1991. 234 Edgardo P. 7Ziaciba Inflation, however, remains substantial relative to major trading partners and competitors. At this stage, while the need is there to further tighten inflation differentials, it is not obvious that the remedy lies mainly in further tightening monetary policy, especially in the presence of capital inflows. If anything, the lifting of exchange controls forces the authorities to look deeper at the causes of inflation, especially the role of structural factors. Recent developments also lay exposed the lack of depth of domestic capital and financial markets and showed their inability to absorb increased inflow of private savings, denominated in foreign exchange, and to recycle that efficiently into the economy. As a result, some of these resources are leaking out of the economy in the form of capital outflows. Over the medium to long term,, however, a stronger, more versatile capital market is urgently needed to provide the financial services and products to recycle these savings. Toward this objective, instead of offering special fiscal incentives, the govermment's most useful contribution would be to ensure macroeconomic stability and to reduce regulatory constraints that hamper or distort incentives. Capital Accowt Loibuztion: The PNiippines Eaperiece 235 Table 12.1. Philippines: Foreign Excange Liberalization Measures, 1991-92 Trade and trde related Nontrade and services lmtenaionoflnandal Ruling trasactions related transactions transactions 1991, July 2. 1991 Allowed commodity CB Circular exporters to retail 2 No. 1291 percent of dteir fore earnings in Special Foreign Currency Deposit Accounts (SFCDA) to be used for certain specified pur- poses. October 15,1991 Extnded repatriation pe- CB Circular riod for export proceeds No. 1317 from 60 to 90 days. Deember 11. 1991 Allowed exporters to use CB Circular foreign currency deposit No. 1317 units (FCDU) loans for up to 70 percent of the value of export letter of credit, purchase order or suppliers crediL 1992 January 3, 1992 Allowed domestic pur- Lifted mandatory surren- Allowed fill and immediate CB Circular chases and sales of gold der requirement of fore repatriation and remittance No. 1318 without priorCentral Ba except for 15 ypes of for all ypes of foreign approval. nontrade fore earners inveslments. Increased limits on the Allowed outward invest- amount of fore that ments, provided funds are Authorized Agent Banks widtdrawn for FCDU or are (AABs) may sdl to resi- not among those required to dents without prior CB be sold to AABs. approval CB Circular Increased retention rate No. 1319 for export proceeds to 40 ocrccnt. January 30, 1992 Bank's long and short forex CB Circular positions pemitted but may No. 1327 not exceed 25 pr -M and 15 percent. rep .vely. of their unimpaired capital. table contnues onfollwingpqge 236 Edgardo P. Ziakita Table 12.1. (continued) Trade and rnde ' elted Nontrode and services Inrenatfon ljfiancia Ruling transwations relaxed transactions trunsactions April 1.1992 Allowed fiunds retained CB Circular in SFCDAs to be used No. 1334 freely for any purpose. April28, 1992 Lifted restrictions on off- CB Circular floor fbrex tading air-n-g No. 1338 foregn banks with tde launching of dte Philippine Dealing System (PDS). Allowed service exporters to use FCDU loans for up to 70 percent of their epected forex receipts. July 28. [992 Allowed use of all codes CB Cicular of export payments No. 1338 (except imercompany overseas accounts OA. which still requires prior CB approval) and import payments and to domestic producer, with prior CB approval. Lengthened period for imward remiance of export procceds from 90 to 180 days. Lifted prior CB approval requirement on certain tansatons (e.g., on- dollar exports, and self- funded and consigned imports). Simpliied reportorial and procedural require- ments. table continues on folowing page CapitaAlccowtLibemrizaio,r The Philippines Erperience 237 Table 12.1. (conmnued) Trade and trade related Nontode and services Inernwuionalfinandal Ru/lAg transcions reated transactions transactions August 3,' 1992 Allowed all forex trading to be conducted through the PDS because all banks had their systems on line. August21, 1992 Allowed FCDUs to grant CB Cirular foreign currency loans without No. 1351 prior CB approval to exporters, domestic producers/manufac- turers, and FCDU depositors. August24, 1992 Allowed free domestic Lifted all quantive re- Allowed inward forcign CB Circular and international trade of strictions on amount of investments not to be registered No. 1353 gold. fbrex that can be and proceeds not to be purchased from banks. surmendered except if bank funds are to be used sui- sequendy for capital repa- tniation and dividend re- Allowed exporters to Lifted mandatory sur- Allowed outward investments retain 100 percent of render requirement for all without prior CB approval in their export carnings. forex receipts. accounts less than USS1 million per investor per year. Liberalized further Allowed exporters, domestic allowable export pay- manufactrers, and FCU ments. depositors to receive shor-term foreign currency loans without prior CB approval. However. Lifted prior CB approval medium-and long-term loans as on all export vans- well as short-term loans to the actions, public sector stil require prior approval of the CB. With or without prior approval, foreign currency loans mnay not be registered and proceeds thereof nay not be surrendered to ihe banking system unless bank; funds are to be used subsequently for debt servicing. 238 Edgardo P. 7lakta Figure 12.1. Selected Invisible Receipts, 1988-92 IA~~~~~~~~~~~~~~I 1.2 0.9~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~S 1.2 *I ------*---- - --'' 's 1.1 0.7 / 0.6 - - - - 0.8 - . 0.3 0.2 198 198 199 199 199 8 9 0 1 2 1- -0- - Remiiimces FCDU -U- QuanmraesTvel Oters - -*-Totalrecipts Figwe 12.2. Total Foreign Exchanges: Overseas and Foreign Currency Deposit, 1988-92 (as percem of non-gold reserves MGRM 7 - 600.-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ A~~~~ 300j _-Fcr- -EA Deps --E-QuarmnOveas -FCD to NGR 8o NGR Depo199 r9 N11R CapitalAccout Liberaizaian: The Philippine Experience 239 Figure 12.3. Foreign Exchange Deposirs: Overseas and Foreign Currency Deposit, 1988-92 (as percent of total deposits) 32- 24 2N 22 - - 20. 18. ;~ - -~ C-E- -;_ 14 t, 10f - 1988 1989 1990 1991 1992 -*--- Foreign Exchange to Total - -ED- - Quarters Ovecseas to Total -±-FCD Total Deposits Depos Deposits Figue 12.4. InterestDierentidls, I9&-92 (peso vs dollar TBhil rates) lA -------cf \ _ \/ 22 20 , 1o I *-u . . . . . X ... . 16 14 . 12 t *-O 1 10 1 2 > 1988 1989 1990 1991 1992 -O-Real ---- Nominal 240 Edgardo P. Zialcita Figure 12.5. Assets and Deposit Liabilities, 1988-92 i5 - -TA..~~ 4 k._A' 4.5~~~~~~~~~ 3> ^__^__s_ ~ ~ ~ ~ ~ ~ ~ . _5. 5' 4 ~ ~ ~ ~ ~ ~ ~~~~- ji 1. . . . . . . . . . . . . . . . . 1988 1999 1990 1991 1992 1- - -- Local Depos Liabdiitie -U- CLahms on Resident - -A- - Tota Asset Fire 12.6. Total Interbank Foreign Exchange Trading Volume, 1988-92 1.4 . LAX 1.3 1.2 0.9 0.6 0.4 /-~ 0.3 0.2 *98/ 199 1991 199:! 1989 ~1989 1990 MI9 1992 GapitalAccowutLiberalmion. T7hePhilippines xperience 241 Figure 12.7. Central Bank Foreign Exchange Volune, 1988-92 (ratio to roraf volume) 100 X~ )I 30 20 10 0 I 4 ;I I I 19BB 19S9 1990 1991 1992 Figure 12.8. International Reserves of the Baning System, 1988-92 9 LS 7 6 S 4 3 2 0 193 1919 1990 1991 1992 UCenr EBnks CCamzcdRiBk 242 Edgardo P. ziadita Figure 12.9. Trends in Selected Major Foreign Exchange Outflows, 1988-92 (in percentage of GNP) 30 28 A - 26 A A --,' A 24 v -- A A_-A, 'A 22 A 18~~~~~~~~~~~~~~~~~~~~~~~~~i 20, ;_ ,A , 16 14 12 10 8 6 U -0- 4 - 01 1 1 I I I I I I I I I I I I I I I I 1988 1989 1990 1991 1992 |*--Scmxr p0MM= -N-ConRm= no - - A- - No=cnsmune -mor I iS~~ma 13 INDIA: COMMERCIAL BANK REFORM Suman K. Bery Suman K. Bery is special consudantgo de Reserve Ba* of nda on leavefrom the World Bank. The vzews and opiuions exressed in thu paper are emirely the responsibility of the author and should not be auributed to eher of hie organations with which he is associated. Editor India is currently engaged in a wide-rangig program of reform and deregulation. Aldthugh initial efforts began in the early 1980s, the pace and intensity of reforms has increased markedly since July 1991. In mid-1991 India was in the throes of both poitial and economc crisis. A mid-term Parliamentry elecdon had produced no clear majority and ex-Prime Minister Rajiv Gandhi had been assassinated in the course of the campaign. Concurrenly, the country was racked by a severe balance of payments crisis. The proximate causes of this crisis were the Gulf war with its impact on oil prices and workers' remmances, as well as te umoil in the former Sovie Union, a major trading partner for India and a source of low-cost oil. The combination of polftical and economic uncertainty in tun provoked unease on th part of foreign commercial lenders and rating agencies, causmg a dryig up of commercial credit, prompting recourse to official balance of payments support from the intenational agencies. While these external elements served to provoke it, the crisis was also a sign that the inefficiencies generated by the existig policy framework were no longer affordable. Accordingly, the adjustiment program, developed under the leadership of Prime Minier Narasimba Rao and Finance Miister Manmohan Singh, has consisted of both stabilization and structral elements. The stabilization effort focused on fiscal consolidation, almost entrely at the centrl government level, largely consisting of expenditure cuts1 The structral measures were oriented largely to deregulation of private industry, foreign exchang 1. India bas a federal finamcial system. In fmancial year 1991-92 (April to March) the consolidated deficit of the entire public sector has been estimated at 10 perces of GDP, of which the deficit of the central (or Union) govenment was 6.5 percenL The targe fica deficit for 1994-95 is 6 percct of GDP. Sine March 1993 the exchange rate of the Indian rupee has been unified and market-detmined, with the Resee Bank intvening to smooth out temporary imbaances. The rae bas remained steady at around Rs 31.37 per US dollar. 243 244 Sumwan IC Bery and trade, including liberalization of domestic and foreign entry into previously licensed sectors, the repiacement of a large number of quantitative restrictions (QRs) on imports by import tariffs, and the replacement of complex administrative allocation of foreign exchange by a unified market-determined exchange rate, and virmually full current account convertibility- The package was designed to stimulate export growth, and more broadly to improve the efficiency of a highly protected industrial sector. Reform of the domestic financial sector did not form part of the initial set of reforms, but events quickly moved it to the forefront. The mcoming govermnent commissioned a high-level committee to review future directions for the financial system. This committee, referred to as the Narasimham Committee, Government of India, issued its report in November 1991, and was highly critical of the weak financial position of Indian depository institutions that are mostly owned by the government Second, with the diffusion of Bank for International Setdements (BIS) capital stndards and norms, foreign regulators required Indian banks operating abroad to meet international standards of accouning and capital adequacy. Third, in April 1992 irregularities were uncovered in the portfolio management and treasury activities of a number of commercial banks, both domestic and foreign These irregularities revealed a large-scale breakdown in the internal controls of banks, as well as systematic violation of Reserve Bank regulations. The financial system, and the commercial banks at its heart, were seen to be operating a rogue existence, notwithstanding an elaborate panoply of regulations, internal and external audits, and central bank supervision. For these reasons financial sector restructuring, hberlization and deregulation have become more prominent on the reform agenda, and have spawned a vigorous internal debate on the evolution of the system (Bery 1994; Government of India 1993a; Narasimlhain 1993; Rangarajan 1993). In this paper I will briefly sketch past policies toward the Indian financial system, concentradng particularly on those policies that affect the commercial banks, which comprise around two-thirds of the formal financial system. I will then descnibe the agenda of change which began frst to be ardculated in the mid-1980s and implemented at that time, but which has received further impetus from the forces descrbed above. Finally I will descrbe recent measures taken by the authorities, and present some of the issues open for debate at this time. Key indicators of the Indian economy and financial system indicate that, notwithstanding the denouement in 1991, the 1980s was a relatively successful decade for India, with average GDP growth of 5.3 percent and average inflation rates of around 8 percent. The table also indicates the steady increase in the depth and institutionalization of the fincial system, as reflected in a rising M3/GDP ratio, increase in the number of bank branches (particularly in rural areas), and a reduction in the cuarrency rato. The draw on foreign savings (reflected in the difference between the savings and investment rates) was around 1.5 percent of GDP, resulting in a steady rise in the extrnal debt ratio. India: Commercla Bamk Refom 245 The Historical Context Indian baning and finance were well developed even in the colonial period, sufficiently so to attract the auention of many academic analysts, including Keynes. There was a significant presence of both foreign and domestic banks, and a well- developed stock market. The financial system, then as now, was centered in Bombay, and the formal financial system was surrounded by and well integrated with traditional or informal financial markets of considerable reach and efficiency. The Reserve Bank of India (RBI), the Indian central bank, was established in 1935 along the pattern of the Bank of England. Prior to independence, commercial banking was largely oriented toward financing of both internal and external trade. The first two decades after independence (1947) were spent in creating the legislative framework appropriate for banking in a newly independent nation, one that had committed itself to planned development. The colonial banking system was seen as inadequate in crucial respects: its focus was entirely short-term, providing working capital and trade finance, and its penetr-aion into the rural areas was pitifl. The landmark Banking Regulation Act was passed in 1949, and provided the legal foundation for considerable control by the Reserve Bank of India over the pattern of credit flows by banks. Efforts were made to encourage te largest bank, the Inperial Bank, to extend its branch network into the rural areas, but it did not find this remunerative, leading ultimately to its nationalization in 1955 and rechristening as the State Bank of India. Term fmancing institutions for agricultre and indstry were established in the public sector, supported by rediscount facilities and capital from the RBI. Life insurance business was nationalized in 1956. The privately owned domestic banks did not cover themselves with glory in the post-independence period. In addition to several bank failures, there was growing criticism of the conceDrtition of economic power in the hands of industial houses which dominated the major banks. There were increasing efforts in the 1960s to curb 'excessive' credit preemption by favored industrial customers. This period saw the itroduction of the Credit Authorization Scheme, which initiated the review of large credit accounts by the Reserve Bank itself (Reserve Bank of India, 1985). This was followed by an attempt at defiing a National Credit Plan: what came to be known as 'social bandking'. T.he agenda underlying social banking was the extension of banking to rural areas and reorientation of credit flows away from big business toward the neglected sectors such as artisans and self-employed. 246 Swnan K. Bery State control of the commercial banking system greatly intensified in 1969, with the nationalization of the 14 largest domestic private banks.2 A programming and monitoring system was put in place to realize the goals set before the banks to reorient credit flows, expand the branch network of banks, particularly to the rural and semi- urban areas, and to increase the mobilization of financial savings through the banking system (Rangarajan and Jadhav, 1992). Banks were set quantitative targets in all these areas, and their attention (and that of the regulators) was devoted to performance on these criteria, rater than financial soundness. Target allocation to designated priority sectors was initially 33 percent, and rose over time to 40 percent. Such directed credit targets also applied to privately owned domestic conmercial banks, and in modified form to foreign banks. These trends were paralleled by a rise in the preemption of credit by the central government to finance its fiscal deficit. The main device used was the statutory liquidity ratio (SLR), as an obligatory investment requirement in government bonds related to the size of a commercial bank's liabilities. Such portfolio requirements were also imposed on insurance companies and provident funds. The existence of this large pool of captive investors, mainly under public ownership, permitted government to place debt at below-market rates. This led to the decline of voluntry holding of goverment paper by the public at large, and prevented the development of a liquid secondary market in such paper. In time, the scale of government funding requirements exceeded the capacity of this captive market, and the government resorted to icreased direct borrowing from the RBI CLe. mnetisation of the deficit). In order to neutalise the impact of this expansion in high powered money, the RBI was compelled to rase reserve requirements, referred to as the cash reserve requirement (CRR), which represented a further control on bank asset portfolio composition. The proportion of incremental demand and time liabilities of scheduled commercial banks preempted by the CRR and the SLR rose from 40 percent in 1980-81 to 63.5 percent in 1990-91 (Government of 1nda 1993a). These asset restrictions were additional to the 40 percent of loans and advances that had to be assigned to priority sector borrowers, usually at susidized interest rates. Furthermore, lending to large industial borrowers (in both the public and private sectors) took place through officially sanctioned consortia of banks with rigid rules of entry and exit to such consortia. By the begining of the 1990s therefore, the margin to manoeuvre available to Indian commerci banks had become extremely small. The structue of the industry was however protected by a de facto ban on entry of private 2. In addition six banks were nationalized in 1980. This group of 20 banks, wholly owned by the Government of India, are referred to as the nationalized bans. With the merger of two nationalized banks in 1993, their number now stands at 19. The State Bank goup includes the State Bank of India and seven associate banks. The 19 nationalized banks and the eight baniks in the State Bank group are collecively referred to as the public sector commercial banks. India: Commercial Bank Reform 247 domestic banks, and only limited entry of new foreign banks, or expansion of existing foreign banks. The Narasimham Critique As noted earlier, the most recent wave of liberalization has been guided by the Narasimbam Report. The Report made the strategic judgement that the problems of Indian banking were not fundamentally attributable to public ownership, but rather to the managerial and policy environment within which banks had operated. It was forthright in drawing attention to the high cost, poor service, low profitability, poor loan recovery and weak capital position of virtually all publicly owned banks. The main explanations provided for this state of affairs were excessive ce ion and political interference, which had served to undermine the sense of institutional autonomy, pride and accountability in the banks. To correct this situation the Report recommended a series of measures to tighten accounting rules, raise capital requirements, reduce asset restrictions of various kinds particularly CRR, SLR, and priority sector credit, to strengthen the legal basis for enforcing foreclosure, and to liberalize the entry of both domestic and foreign banks. The Report was somewhat circumspect in recommending full liberalization of interest rates, arguimg instead for administered maximum deposit rates at real levels and minimum lending rates, with only gradual decontrol. On the managerial side, the report made a strong plea for full autonomy for the mgement of the commercial banks. In particular it argued that supervision and control of the commercial banlk be entrusted entirely to the Reserve Bank of India, rather than being shared between the Central Bank and the Ministry of Finance (the owner), as is currently the case. Overall it argued that 'issues of competitive efficiency and profitability are.ownership neutral. It is how the institutions function or are allowed to fuicton that is more important" (Government of India, 1991). As would be clear from the above synopsis, the Narasimbam proposals represented a clear departure from the prior pattern of Indian banking. While an overall vision of the future Indian banlkng system was sketched, issues of appropriate sequencing were not highlighted; thus the restructuring agenda and the liberalization agenda were presented as occurring concurrently. A number of policy actions have been taken since the report was submitted, while as noted a debate has been stimulated on others. Issues of strategy and sequencing are now more sbarply defined, although final positions on certain key issues (e.g. the treatment of contminated assets, and the treatment of excess staff) remain unresolved. Accordingly, we next tum to actions already taken, and issues still to be addressed. 248 Suman X. Bery Asset Classification, Provisioning and Capital Adequacy Norms The first step, appropriately enough, was to establish the true condition of commercial bank balance sheets, especially of the public sector banks. To this end more objective and stringent standards of income recognition and asset classification were announced in April 1992, to be incorporated in the accounts of commercial banks for the financial year 1992-93 which ended in March 1993. Based on these standards, explicit prudential provisioning norms were laid down. In regard to income recognition, the Reserve Bank instructed banks to treat a credit facility (loan or cash credit account) as 'past due' when interest has not been paid 30 days from the due date. A "nonperforming asset' was defimed as a credit facility in respect of which interest remained unpaid for a period of four quarters (one year) from the date it had become 'past due' during the year ending March 31, 1993. This definition will become progressively tighter: to three quarters in the financial year ending March 1994 and two quarters in the year ending March 1995. Banks are not allowed to book interest income on nonperforming assets. With regard to asset classification, banrs will now be required to classify assets into just four categories: standard assets, sub-standard, doubtful. and loss. This system replaced an earlier system of eight health codes, which was heavily reliant on subjective judgments of bank officials. Under the new classification scheme, a 'sub- standard asset' is one which has been classified as nonerforming for a period not exceeding two years. Banks are required to make nonspecific provisions of 10 percent against assets classified as sub-standard. After an asset has been classified as sub- standard for', NO years, its classification changes to 'doubtful'. Doubtful assets attract 100 percent provisioning for the unsecured portion of the advance, while provisions rise progressively to 50 percent of the value of the secured portion over three years following the classification as doubtful. An asset is considered a 'loss' asset when it is so identified by the bank, its auditors (either internal or external), or the RBI inspectors. Concurrent with these regulations, capital adequacy requirements linked to risk- weighted assets have been imposed. A four percent capital adequacy ratio is to be attained by March 31, 1993, rising to 8 percent by March 31, 1996 (Mar4h 31, 1994 for Indian banks with overseas branches). Foreign-owned banks operating in India were to attain the 8 percent standard by March 31, 1993. As has been prescribed by the Basle Committee, two forms of capital have been distinguished: Tier I consisting of common equity and unencumbered reserves and Tier II consisting of long-term subordinated debt and various forms of hidden reserves. The total of Tier II elements may not exceed Tier I or core capital. India: Commercia Bank RLeform 249 Implementation of Norms To facilitate the introduction of ese new standards, the RBI provided the banks with certain transitional reliefs. As minimum provisioning standards were being specified for te first time, banks were allowed to apportion minimum provisioning identified for end-March 1993 as follows: at least 30 percent to be taken into the 1992- 93 accounts, and the remainn 70 percent in the 1993-94 accounts. Such phasing was not however permitted for provisions against loss assets which had to be folly provided for in the 1992-93 accounts. Second, primarily as a matter of administrative convenience, advances with an outstanding balance of Rs 25,000, consisting of a very large number of borrower accounts but only about 12 percent of total credit outstanding, were exempted from the need for detailed classification for provisioning purposes, although a classification into performing or nonperforming was needed for income recognition purposes. For the first year banls were allowed to make a minimum general aggregate provision of 2.5 percent of total loans ousandings in this category for 1992-93, and then a minimum of 5 percent in 1993-94. The impact of the new accounting norms was reflected in the accounts of the banks for 1992-93. Even after these transitional provisions, as well as te relatively generous definition of non-performing assets in the 'irst year, the accounts revealed what were officially acknowledged to be 'significant weaknesses' in the financial position of the public-sector bangs. For loans above Rs 25,000, the share of nonperforning loans in total loans ranged between 8 to 10 percent m the better managed banks to as high as 35 to 45 percent in ihe worst ones. For the public sector banks as a whole, non-performing loans were estimated at around 21 percent of the total loan portfolio, both domestic and foreign (Government of India, 1993). Furthermore, the introduction of tougher income recognition standards and minimum provisioning requiments had a dramatic effect on the publisbed profits of the public sector banls: of the 28 banks 13 were forced to report losses as against only two the previous year. Combined losses amounted to Rs 33.7 billion (approximately US$1 billion) as against a modest profit in the previous year. In regard to minimum capital, it was estimated that the public sector banks as a group faced a capital shortfall of approximately Rs 25 billion in 1992-93, and could require additional capital amounting to Rs 170 billion to atain the 8 percent level by 1996. Rehabilitation and Restructuring With the financial position of the public sector banks more objecdively established, the task of rehabilitation and restructuring has begun to be taken in hand. Achievement of mimmum capital levels was given first prionty. Whfle presenting the Union (Central Government) Budget for 1993-94 in Febnuary 1993, the Finance 250 &mKan.K Bey Minister articulated the broad policies that would be foliowed in replenishing capital and restoring viability to the public sector scheduted commercial banks (Government of India 1993c). This framework been followed since and consists of these principal elements: The government would subscribe capital to help banks meet capital requirements mandated for the first capital adequacy test date of March 31, 1993, to a principal value of Rs 57 billion. A further Rs 56 billion has been provided in the 1994-95 budget. * The capital would carry a counterpart obligation for investment in government bonds. It has been subsequently decided that the bonds would be of finite tenor (twelve years, with six years grace) and carry a coupon of ten percent, somewhat below prevailing market rates. The bonds are marketable, but not eligible for SITR. * The State Bank of India as well as such nationalized banks financially able to do so would be allowed to access the capital markets to raise fresh equity, subject to the constraint that government retain majority ownership and control. In the case of the nationalk,A banks (but not the State Bank of India) such dilution of government equity entails legislative action which has recently been initiated. Specific commitments would be sought from each bank to ensure that its management was taking adequate steps to prevent recumence of the erosion of portfolio quality and of capital. This has since taken the form of "memoranda of undersanding" executed between the Reserve Bank and the boards of each of the nationalized banks, setting out agreed performance targets as a quid pro quo for the infusion of government capital. Events in the intervening year have more or less followed the above course. The tougher accounting standards have thrown into sharp relief the range of performance across the public sector banks. Fortunately the stronger banks are the larger ones in terms of deposits and advances; however there exists a handful of weaker banks whose longer-term viability as independent units is questionable. One of the weakest has recently been merged with one of the stronger nationalized banks. Memoranda of understanding have been concluded with-each of the nationalized banks prior to the infusion of capital. The approach that is being followed could in this sense be considered 'case-by-case& rater th generic, reflecting the diferent financial and managerial context of each bank. The process has served to sdmulate greater attention within the banks on issues of loan recovery and profitability, and to highlight the very India: Commercial Bank Rfom 251 different cost structures of the banks, particularly with regard to numbers employed and payroll costs. Banks are slowly beginning to exploit the greater flexibility available to them to rationalize their branch structure in urban and semi-urban areas; closure of rural branches however remains subject to Reserve Bank review. The banks have been able to negotiate an agreement with their unions to permit computeization on a substantial scale. The government has provided a blanket assurance that either on account of computerization or on account of other restructuring there will be no involuntary retrenchment; nonetheless the banking unions have initated token industrial action to protest against branch closures. On the financial side, the State Bank of India was successful in floatng a large public issue (Rs 24 billion approximately US$775 million) of both equity and debt in the domestic market, establishing a precedent that the stronger nationalized banks will be able to follow once the necessary legislative changes are in effect, and once their balance sheets have been strengthened. Important first steps have therefore been taken, but full recovery and restoration of profitability will reqre sustaned action on a number of fronts which are now discussed. Asset Recovery and Exchange The high proportion of nonperforning loan assets in the portfolios of the public sector banks are partly attibutable to the weak legal famework in India for enforcement of contracts and realisation of collateral, and the delays in prosecuting default cases through the civil courts. Earlier committees had suggested that special tribunals be set up to handle cases of default. This proposal was endorsed by the Narasimham Committee and accepted by the government. Such special tnbunals are about to commence operations in the major cities shortly, and should help improve the prospects for recovering interest due or collecting on pledged collateral. The Committee also argued, however, that: ...the impact of the setting up of the Special Tnbunals will be felt by the banks only over a period of time.. .In the meanwhile it is necessary to work out an arrangement to deal with portion of the portfolio of banks which has already become bad and doubtful and whose recovery is being hampered by the slow legal process. To continue to keep such assets in banks' balance sheets would not be desirable even ... substantial provisions are made there against... It would be far more appropriate if these assets were taken off the balance sheet of-banks and institutions, so that the funds realised through this process can be recycled into more productive assets ...(Government of India, 1991). Accordingly, the Narasimham Committee proposed the establishment of an Assets Reconstruction Fund (ARF), to purchase doubtful assets from the banks at market value, not the book value. It was felt that this would serve to maximize 252 Swnw K Bery recovery, give the banks immediate liquidity, and to concentrate managerial energies on the future rather than the pasL While such recovery agencies have worked successfully in other countries, the ARF concept has not been accepted by either government or the Reserve Bank. Given the large geographic expanse of the country, there is doubt that any new body could quicldy develop the reach to enforce collection. There is also the concern that absolving banks from responsibility for collection will weaken their own credit discipline in the future, and may even contaminate the behaviour of borrowers in good standing. Finally, there are significant problems foreseen in capitalising the ARF, over and above the existing capital needs of the banks (Government of India, 1993a). Accordingly, current policy is to permit the originating banks retain responsibility for recovery, aided by the institution of the Special Tnbunals. The issue of loan recovery is much broader than the creation of a one-time ARF, and is deeply affected by both the overall policy framework governing so-called sick (i.e. bankrupt) companies and the politicization of the banking system so openly referred to by the Narasimham Committee. As has been pointed out by yet anoher govermnent appointed committee dealing with industrial sickness and corporate restructuring, the Goswami Committee (Government of India 1993b), a series of policies, institutions and interventions militate against corporate restructuring and liquidaton in India. These policies are often referred to as 'exit policies'. The net outcome for debtors is that funds remain congealed, good money is poured after bad, and the firm's residual value is often lost through delay, fraud or political intervention. As measures to curb dishonest and repeated default, the Fmamce Minister in his speech presenting the 1994-95 budget (Government of India 1994) announced that the RBI would circulate among banks and financial institutions names of defaulting borrowers above a certain limit, as well as publishing a list of defaulting borrowers in cases where suits bave been filed by banks and financial institutions. The efficacy of these measures is as yet unproven, but they should be seen as part of an overall package of measures and incentives designed to deter default. Asset Composition and Profitability While recognition and resolution of past losses is the indispensable starting point for bank rehabilitaton, the process will not be sustinable without a restoration of profitability. There is litde prospect of the commercial banks approaching the markets for Tier I or Tier It capital without assurance that the pricing and policy framework will permit an adequate return to creditors and shareholders. The Narasimham Report dwelt extensively on the reasons for declining profitability of the public sector banks, and the measures that would be needed to reverse this decline. The major policy causes cited by the Committee were the massive preemption of funds by the government through the SLR and the CRR, both remunerated at well below the Indir Commercal Rant R-fonn 253 opportunity cost of funds, and the directed lending to the priority sectors, also at below market rates. Action has begun to be taken. Along with the reduction in the fiscal deficit of the central govermnent, the combined burden of the CRR and the SLR has declined from 63.5 percent of incremental deposits in 1991-92 to 25 percent in early 1993. The Narasimham Committee set the goal of an average SLR of 25 percent over the medium term. This goal has been accepted by the Reserve Bank, which aims to reach this level by March 1996. The implied portfolio shifts are large. If the commercial banks were to avail of the full reduction in mandated SLR, that is, if they do not elect to hold governent securities in excess of the minimum, and put all the released resources into advances, the incremental credit-deposit ratio could be extremely high. One can question the capacity of the existng banks to handle such an expansion judiciously at a time when their own credit assessment capacities are weak, and when, in response to hlberalization in the real sector, the quality of credit rsks themselves will be rapidly changing. The appropriate response is twofold. The first, obviously, is strengthened supervision. The second, as argued more fullly below, is a rapid shift to market determination of government security rates, to provide a 'safe haven' for funds before they can be prudently deployed. The authorities have pardally followed the latter course. While there remain elements of guidance, rates offered on government securities have become more attractive, and auctions are now widely used to place the debt. The combination of high yields on government debt, weak capital positions of banks, the zero risk-weight assigned to government securities in assessing capital adequacy and the continuing industrial recession have led to a major shift at the margin from advances to investments in government paper, such that the banks are now volunarily holding SLR-eligible securities well in excess of their requirements. Tbus the first response of banks to greater portfolio freedom has been a 'flight to quality' rather than unsound lending, which is all to the good. Such voluntary holding of goverment paper would in turn make possible use of indirect tools for managing monetary policy, as discussed below. Interest Rates, Government Debt, Exchange Rates Deregulation of interest rates will be among the most important, but delicate tasks in the process of financial sector reform. The present strucure is itself the product of gradual deregulation from the excwdingly rigid and compartmentalized structure that existed till the mid-1980s. At that time the role of interest rates in deposit mobilization was well recognized, but there was little allocative role assigned to them on the asset side of the balance sheet A shift in diection was first recommended in 254 Smo K. BRery the report of the Committee to Review the Monetary System by the Chalkavarty Committee in 1985, .which noted a number of weaknesses in the interest rate structure that had emerged (Reserve Bank of India, 1985). At the macro level, the low yields on treasury bills and dated securities were leading to excessive monetization of public debt.3 At the same time, they were constraining the development of the capital market, while reducing the profitability of the commercial banks. At the micro level, concessionality on interest rates had permitted projects of doubtful viability to be undertaken. Insulating banks from price competition had also no. served to improve customer service. The Chakravarty Committee proposed a cautious liberalization of the then prevailing structure, within the famework of an administered interest rate system. The proposed reforms were to be based on across-the-board increases in the rates offered by government on its debt obligations. The Committee recommended that the discount on short-term Treasury bills (91 days) be set to provide a yield marginally positive in real terms, and that the rest of the yield structure be adjusted accordingly to return expected real yields of one per cent to three percent per anmum, depending on maturity. The primary purpose of this increase in yields was to make government securities sufficiently attrActive to induce voluntary holding of government debt by the public at large, so as to reduce the monetization of the public debt, while concurrently boosting commercial bank profitability. These adjustnents in government security yields were to provide the foundation for the interest rate structure offered by the commercial banks. The Committee recommended that the RBI specify a maximm deposit rate for fixed deposits of five years and above, and a floor tending rate for the non-concessional advances of banks. This approach was intended to prevent unequal competition between banks of different sizes, and also to ensure a minimum spread between the deposit rates and lending rates which was broad enough to provide a basis for viable banking operations, and yet narrow enough to prevent laxity in bank administration. On these grounds, the Committee recommended a three percentage point spread between the maximum rate on deposits and the basic minimum lending rate. The Committee further recommended that the RBI specify the one-year deposit rate at a suitable level so as to ensure adequate incentives for mobilization of longer-term deposits by the banks. With regard to priority sector lending, the Committee recommended a move to just two concessional rates: the minimum lending rate, and one concessional rate below this minimum rate. The general evolution of interest rate policy in recent years has been broadly consistent with the above scheme. The Narasimham Committee on the issue of interest rates noted that while interest rates on government debt had been progressively increased in recent years, they were still not at levels to attract voluntary subscribers. While the medium-term objective was to move to market determination of interest 3. Securites paying coupon interest wih inial matunrities n excess of one year. India: Commercial Rnk Reform 255 rates, it advocated a cautious approach to deregulation, insisting correctly that macroeconomic stability and the health of financial intermediaries were essential preconditions for full market determination of interest rates. It therefore concluded that for the time being the RBI should remain the authority to determine the level and structure of interest rates. It suggested that a rediscount or refinance rate ('Bank Rate') be used as the anchor for the structure of interest rates, and that the structure be calibrated along the lines recommended by the Chakravarty Committee. The present situation is as follows. The RBI specifies a ceiling deposit rate approximately 2 percent above the current inflation rate, applicable to fixed deposits of maturities between 46 days and three years and above. Currently, this rate is at 10 percent. Commercial banks are free to set intermediate rates below this ceiling, but in practice these rates are established in accordance with guidelines of the Indian Banks' Association. The RBI prescribes outright the rate of interest to be offered on savings deposits, currently at 5 percent Banks are subject to no rate restrictions on certificates of deposit which are a wholesale instrument subject to a high minimum size. In the case of advances, banks are stipulated a minimum rate for advances in excess of Rs 200,000, currently set at 15 percent; there are currently tbree concessional rates below this, linked to size of advance rather than to sector of beneficiary as was previously the case. Apart from a highly concessional rate of 4 percent offered to the extremely indigent, the two other specified rates are 12 percent for advances below Rs 25,000 and 15 percent fixed for advances above Rs 200,000. In addition exporters enjoy access to short-term preshipment and postshipment credt at 13 percent These rates for exporters are in tu underpinned by access to refinancing facilities from the RBI, although this refinance is not currently being availed of to any significant extent. There has been evolutionary change in rate determination on government securities. At the time of the Chakravarty report all government security rates were determined unilaterally by the RB in consultation with govermnent. Market demand factors were cautiously introduced in 1986 with the auction of 182-day Treasury DiIJs. This instrument was replaced in 1992 by a 364 day bill, also placed by auction, and, in conmnon with the 182-day bill, not held in portfolio by the RBI. In early 1993, the RBI introduced a 91 day bill also placed by auction, but which is being held by the RBI in its portfolio. This instument could potentially be used for future open-market operations. Since April 1992, the RBI has also moved to auction of the entire primary issue of Central Government dated securities, although with RBI participation in the auctions. The one category of debt which is still placed at fixed, preannounced rates is the debt of State governments, although here too there has been a move toward narket levels: the rate offered in 1993-94 was 13.5 percent for a ten-year security. This yield constitutes an effective cap on the term-structure of SIR-eligible securities, and the RBI's effort in the auctions for other securities is to maintain a wellbehaved yield curve with respect to this cap. The RBI has also in the last year introduced repurchase 256 Sun= X Bey operations with banks and fiancial institutions against collateral of dated securities, in an effort to influence conditions in the short-term money market. It is the declared intention of the government to move to market rates of interest on government debt, while the RBI also has declared its intention to move to indirect instruments of monetary control. The need for flexible monetary instruments will be all the greater with the increasing trade and financial openness of the economy, and market determied exchange rates. Thus far, mterest rate and monetary policies have been set largely with domestic conditions in mind. Movement in these directions requires that the primary financing needs of govermment be met outside the central bank, with the central bank intervening in the secondary market to achieve its monetary policy goals. A major step in this direction has been the announcement in the Budget Speech for 1994-95 (Goverment of India 1994) that the government will progressively reduce its recourse to direct funding from the Reserve Bank so-called ad hoc Treasury bills with the intention to phase out this insument altogether m the next tbree years. This move, taken together with the reduction in SLR r, will lay the policy basis for healthy primary and secondary markets in government securities. These initiatives will need to be accompanied by a vigorous regulatory and institional effort, comparable to that which has been underway in the equity markets over the last two years. Indeed the regulatory challenge may be eveen greater, since fraudulent broker dealings in government securities were at the heart of the financial irregularities uncovered in 1992. While a move to market determination might involve a temporary crease in rates, the long-term impact of wider ownership and more efficient markets should be to lower the cost of any given quantum of borrowing. While the Union Budget may be in a position to absorb such a teWmporary increase in rates, the State governments, who are at an earlier stage of their fiscal adjustment, may not. Accordingly, special arrangements will need to be devised to ensure that the constrais facing State govemments do not impede progress in the overall market for government debt. The secondary market yield on government paper should in ime become the reference point for other interest rates in the system, including deposit rates offered by banks. Provided that the capital position of baiks has been strengthened, effective supervision is in place and a modicum of competition has been introduced in the system, the moment would then be ripe for complete deregulation of interest rates. Directed Credit The scope for subsidizing interest rates to preferred sectors will clearly narrow in a market-based interest environment. A rather different set of issues is raised by the priority sector concept itself, i.e., directed porffolio shares, not necessarily at subsidized rates. Charges of discrimination by banks against certain sectors arise in all societies. The case for government intervention is usually justified by the considerable regulatory protection given to banks. By the same token regulators in industrial Indlr Commercal Bank Reform 257 countries typically resist explicit portfolio requirements on the argument that these lower credit standards and dilute accountability. A market response, sometimes stimulated by policy, is to foster specdalized institutions who are more adept at assessing credit risk and recovery in individual niche markets. Considerable action has been taken in recent years to reduce the number of concessional rates. Concessionality is now libked to size of advance rather than to recipient sector. The Narasinham Committee recommended that the share of priority sector advances be reduced from 40 percent to 10 percent and apply to a more restricted group of sectors. No reduction in the 40 percent target has yet been approved; however the list of eligible priority sectors has recently been widened. Sice the incremental lending capacity of banks will rise sharply as the SLR declines, there is a danger that the expansion in lending to these sectors would be in excess of their absorptive capacity, and could perpetate the problems of poor credit quality associated with such lending. Regulation, Supervision and Governance As noted above, the focus of supervision in the post-nationalization period was either on achievement of quantitative targets, or on monitoring large individual credit accounts, but not partidcularly on the financial soundness of banks. The Narasinham Committee urged a refocusing of the supervision effort on prudential issues. It also recommended a shift in the organizational structure of supervision, from the present arrangement as a department of the RBI to a quasi-autonomous board under the Reserve Bank, but with responsibility for all depository institutions. The new structure has been designed and will become effective shordy. Supervision is not intended to subste for management: modem bank transactions are vast in volume and multifrious in scope, and cannot be controlled, let alone run by a central bank. Indeed, the focus of most supervision in industrial countries is largely on the intrnal systems of banks, and not on their individual credit decisions, and is increasingly direted at identifying and addressing systemic risks. This raises perhaps the most difficult and controversial issue: what will be needed to improve the managerial accountability of the nationalized banks. As noted, the judgment of the Narasimnm Committee was that public ownership of the bank was not the issue, but rather bureaucratic and political interference. The Committee offered a range of solutions, the most important of which was to concentrate regulatory powers in the Reserve Bank of India and the new upervisory board, while limiting the role of the Banldng Division of the Ministry of Finance. The Committee also made various suggestions on the appointment and powers of chairmen and bank boards, and on areas for greater autonomy for banks. There has been no consolidated response from the authorities to these proposals, other than to reject any immediate diminution 258 Swnan K Bery in the role of the Ministry of Finance. There has, however, been an official airing of the very deep-seated and substantial managerial difficulties of the banks, indicating that changes in their governance are perhaps forthcoming (Govermment of India 1993a). The dilution of govermment shareholding in the nationalized banks, as and when it occurs, will involve changes in the composition of bank boards, although the right of appointment of the chairman will remain in government hands. Adjustment in the Banks When the commercial banks were nationalized they were permitted to retain their individual identities in the interests of stimulatink limited competition. In practice, a number of centrally determined policies, sucb as central recruintment of staff, rotation of bank chairpersons across banks, policies assigning lead roles to individual banks in particular geographic areas and in lending consortia, and powerful industry-wide trade unions have all served to dilute rather than intensify compeution. Despite these pressures toward homogenization, differences in managerial raditions of individual banks are noticeable in their financial results. However all of them are, by interaional standards, overmanned and technologically antiquated. Further, the quality of their internal controls, communications and humn resources was greatly overstretched by the rapid expansion of the last twenty years. A major task of institutional upgrading hs lies ahead. While the agenda will clearly differ from bank to bank, mannng levels, human resources, technology, iiternal controls, and improved risk management are clearly common areas for all banks. The Narasinbam Committee was of the view that this process would be best facilitated by giving full operational autonomy to the management of each bank. While managerial autonomy is no doubt desirable, it is questionable whether a completely laissez-faire approach would work for the weaker banks. More fundamentally, the banks will need to rethiink their role in the financial marketplace. Given the cost structure of banks (in large part the result of regulatory constraints) the better credit risks will continue to look elsewhere for their funding needs. As access to external funding has widened, this process wil accelerate. In common with banks worldwide, fee-based income wiU need to become a larger part of the total, and more creative ways found of using the extensive branch network which is the banks' major compettive advange. In order to increase the competitive pressure on the public sector banks, the Reserve Bank of India in 1993 announced new guidelines for the entry of new domestic private sector banks. A minimum paid-up capital of Rs 1 billion bas been established, with new banks required to meet the 8 percent capital requirement from the begining. These banks will be required to meet the same priority sector lending goals as existing banks. Several private and semi-official promoters have been granted in principle approval and actual operating licenses have started to be awarded. india Commercial Bank Reform 259 Conclusion This paper has concentrated largely on the challenges facing the Indian public sector commercial banks, and changes in the environment within which they operate. While such a narrow focus may be criticized, the justification is that these remain the core institutions of the Indian financial system, with the greatest capacity to retard or assist the broader programme of real sector adjustment and growth that is underway. The fundamental preconditions for a healthy financial system are clearly in place in India: high saving rates and low inflationary expectations. However, the role assigned to the banks has been radically altered. From institutions designed to channel resources to public investment and to ligbhly regulated private uses, the banks are now being asked to become globally competitive, and to develop risk assessment capacities that may have rusted. The sequencing of reforms so far has followed the "new orthodoxy". Real sector reform has preceded financial sector reform, so that the future pattern of real re;turs is clearer; fal adjustment and disinflation was to provide the foundation for liberalization, although this process has received a setback in the last year; diagnosis and clean up are preceding new freedom. Perhaps the only departure is that entry of new banks is being permitted early in the process. The major issues thus revolve around the management of the process and the pace of change. As the banks take up te task of globalization, though, it is as well to renmeber the persistence of the populist agenda that has colored past policy. It is only if those demands are seen to be met by the new policy framework, that the recent changes will be politically susminable. Equally, policymakers have so far had the luxury of treating reform in each segment of the financial system in isolation: the capital markets, the banking system, government debt markets and foreign exchange markets. A surge m capital inflows in the last six months have demonstrated the linkges between these areas. Lberalization of borrowing abroad has affected the offtake from the domestic banking system; capital inflows have eliminated short-term exchange risk, leading to attractive arbitrage possibilities. These new developments are straining the existing famework of economic management and will entail the development of new sUills and capacities on the part of policymakers. 0x 260 Swnum K. Bery References Bery, Suman K. 1994. "Financial Rehabilitation of Public Sector Banks: Conceptual and Policy Aspects." Economic and Political Weekly XIX, 5 (Jan 29, 1994):251- 55. Government of India. 1991. Report of the Commiuee on the Financial System. New Delhi. Government of India. 1993a. Public Sector Commercial Banks and Financial Sector Reform: Rebuilding for a Beuer Future. Ministry of Finance, Department of Economic Affairs. New Delhi. Govemment of India. 1993b. Report of the Committee on Industrial Sickness and Economic Restiauring. Ministry of Finance. New Delhi. Government of India. 1993c. Speech of Minister of Finance Presening Central Government's Budget for 1993-94. Government of Idia. 1994. Speech of Minister of Finance Presenting Central Government's Budget for 1994-95. Patel, I.G. 1993. "Some Reflections on Financial Liberalisation." Indian Institate of Bankers. Bombay. Narasimham, M. 1993. "Financial Sector Reforms: The Unfinished Agenda.' A.D. Shroff Memorial Trust. Bombay. Rangarajan C. and Narendra Jadhav. 1992. "Issues in Finanial Sector Reform." ln Bimal Jalan (ed.) The Indian Econony: Problems and Prospects. New Delhi: Viling. Rangarajan, C. 1993. 'Banking and Finance: Monetary and Fiscal Policies.' Reserve Bank of bndia Bullin, (July):987-90. Reserve Bank of India. 1985. Report of the Committee to Reiew the Working of the Monetary System. Bombay. Reseve Bank Of India. 1993. Annual Report. Bombay Varma, Jayanth. 1992. "Commercial Baning: New Vistas, New Priorities" (Mimeo). Indian Institute of Mgemet. Ahmedabad. 14 FINANCLAL SECTOR REFORMS AND LIBERALIZATION IN THE REPUBLIC OF KOREA: CURRENT STATUS AND PROSPECTS Choon Taik Chwng Recent Trends in Korean Financial Policy Since the early 1960s the Korean government has made an enormous effort to accelerate its economic development. In order to industrialize in as short a time as possible, Korea adopted a government directed strategy for its economic growth, investing heavily in basic and export-oriented industries. Although Korea has its share of failures in some key sectors, its industial strategy was instumental in and credited for phenomenal growth in a short time. Consequently, such growth resulted in the expansion of aggregate output and a more complex industrial structure. Following the growth of the Korean economy in both size and structure, the direction of economic polcy in Korea shifted from a basically government-led scheme to a free market system during the early 1980s. In terms of internatonal trade, Korea began to make a serious effort to open its market At the same time, domestic companies were encouraged to enhance their competitiveness in the world market. With widespread consensus among Korean economists for less government intervention, the process toward an open market system gained needed momentum in the last two decades. Subsequent govaement efforts to remove trade barriers resulted in substantial increases in both internatonal trade and financial transactions. Since the 1980s, the Korean governnmnt has embarked on financial leralization as part of a comprehensive economic liberalization program. The govermnent allowed the domestic banking market to open up to foreign banks in order to introduce more competition in the industry. The liberalization of the foreign exchange market and capital flow subsequently followed in the latter half of the 1980s. An Overview of Korea's Financial Liberalization Process During the 1960s the government exercised strict control over assets, liabilities, and management of individual banks to extend loans to industries judged to be of strategic importance to the nation's economy. 261 262 C/won Taik C/ang This system of preferential loans was probably seen as quite useful in thc early stage of the economic development. But as the nation's economy grew and became more complex, many of the defects of financial repression became apparent. As long as the interest rates remained low, the excess demand for loans was seen as inevitable. This, in turn, led to extraordinary growth in the money supply and inflationary pressure. The policy loans encouraged investment beyond optimal levels and decreased efficiency in the allocation of resources. The low interest rates paid by banks on deposits discouraged the growth of domestic savings, which led to a greater dependence on the inflow of foreign capital to finance investment. Govermment regulations on the banking industry also retarded the development of the banking system. The securities market and other fmancial institutions were unable to grow out of their infant stage in the 1960s. By contrast, financial repression contributed to the rapid growth of the unorganized market or the curb market. It was difficult to regulate these unorganized money markets. The government realized that an underdeveloped financial sector could impede economic growth and that policy reforms to liberalize the financial market were desperately needed. One of the most serious problems in Korea's financial market was the government control of interest rates. In fact, real interest rates often turned out to be negative because of the government's tight control. To solve this problem, the government drastically raised nominal interest rates in 1965. This brought an unprecedented increase in domestic savings and contributed to the efficient allocation of resources. However, high domestic interest rates led to a sudden increase in international borrowing, and because of the negative-margin interest rate structe, the profitability of the banks deteriorated drastically. Because of these and other side effects, the government lowered interest rates at the end of the 1960s and returned to completely controlling the interest rate policy. PartiaW Liberalization in the 1970s The strict banking regulations of the 1960s brought about a dual structure in the Korean financial market. The market was divided into a regulated market and an unregulated market. Remarkable discrepancy between the interest rates existed in the parallel markets. To absorb the curb market, the most important step would have been to abolish interest rate controls. Nevertheless, the Korean government decided to establish nonbank fiancial institutions (NBFIs). With NBFIs the organized market was not only _xpanded, but it was also segmented and regulated in a discriminatory way. The government reduced regulations on the NBFIs, including the easing of interest rate control, while it continued to hold banks under tight control. The regulatory differentiation on these two financial markets resulted in the unbalanced growth of financial institutions. Various types of Flnancial Sector Reforms and Liberaizarion in The Republic of Korea. Current StatNs and Prospects 263 NBFIs were established in the 1960s. Since then they have grown much faster than the banking industry. During the 1970s the government adopted a set of new industrial policies to promote heavy and chemical industries (HCIs). This change required large investment capital for HCI projects. To provide needed capital, the banks were directed to make loans to HCIs and other strategic industries on a preferential basis. This caused a serious distortion in the allocation of funds and resulted in a huge amount of nonperforming loans, excessive and inefficient investment, and a stifled banking system. Integration of Domestic Financial Markets in the 1980s By the late 1970s it became clear that the government-led economic policy had caused serious adverse effects, such as aggravated inflation, stagnated economic growth, inefficient resource allocation, and sectoral imbalances. As a result the government changed its economic management strategy from government led to private sector led. Accordingly, policy reform required financial liberalization and restoration of the market mechanism in the fincial industry. Financial reform commenced in 1981 and financial liberalization was adopted. Detailed regulations governing the organizational, budgetuy, branching, and business practices of banks were relaxed. The government privatized commercial banks and gradually reduced policy loans. In addition, the scope of operations expanded. For banks, this included certificates of deposit, credit cards, commercial papers, government securities, factoring, and trust business. NBFIs also expanded their range of operations. The overlap and expansion of financial services stimulated competition among fnancial institutions, thereby upgrading their efficiency. To promote competition, new entres were allowed into the financial market. Two nationwide commercial banks opened in 1982 and 1983, and short-term finance companies and mutual savmgs and finance companies were established. During 1988- 89, the government approved the further establishment of commercial banks specializing in small and medium-sizes firms, securities companies, leasing companies, and life insurance companies. By June 1982, most preferential interest rates applicable to various policy loans were abolished, making it easier to scale down policy loans. Tn 1984 the government reduced the interest rates gap between banks and the NBFIs and allowed financial intermediaries within a given range to determine their own lending rates according to the creditworthiness of borrowers. In earlier efforts of interest rate deregulation the Korean government had adopted the strategy of introducing new financial products with interest rates that moved closely with market rates. This strategy seemed to have worked well at gradually providing an environment for wider-scale interest rate 264 Caoon Tadk Anmg deregulation at a later stage when macroeconomic and other conditions for deregulation matured. But these measures sometimes produced outcomes that did not meet expectations, because the government was so worried about an interest rate (cost of capital) burden on the business sector, and thus intervened whenever interest rate levels for new financial products became unpalatable. Low and stable inflation since 1983 and high national savings in excess of domestic investment narrowed the gap between the regulated and free market rates. Against this background, the government initiateu ae plan for intest rate deregulation in December 1988. Most bank and nonbank lending rates and some long- term deposit rates were decontrolled except for rates on some policy loans and short- term deposit rates for the fear of excessive competition among financial intermediaries. A fbw months after the deregulation, however, the government and business sector became so concerned about the drastic rise in interest rates that the government again intervened, giving tacit consent to a collusion on interest rates by financial instiutons. As a result, bank lending rates and most rates in the primary securities market again became very rigid and unresponsive to market conditions. Thus, the first attempt at interest rate deregulation ended as a futile exercise. Financial reform contributed to the integration of domestic financial markets. Unlike the financial reforns in the 1960s and 1970s, financial reform in the 1980s harmonized differential regulations between banks and NBFIs and integrated financial markets that had been segmented. The government has been more concerned with preventing fuher shrikage of the banking sector while still keeping it under control. Meanwhile, restrictions still existed on foreign participation in many businesses within the domestic financial markets and on international capital movements during the 1980s. Current Status and the Issues of Reform in the Korean Financial Industry In the process of Korea's economic development, the main role played by Korea's financial industry was to successfully support growth in the real sectors of the economy, even with the acute shortage of domestic capital. At the early stage of hie economic development, Korea's banks functioned as quasi-government organizations in supplying capital from the intemational sources. They were managed more like public entities, and the development of the fiancial industry was delayed. At the same time, it was necessary for the government to intrvene in the financial system to allocate funds for promoting the growth of the economy. As such, no financial institution in Korea has gone bankrupt. Despite the lack of a deposit insurance system, there was high public trust in the financial instiutions. In this respect. Korea's financial system is different from that of the developed countries where financial institutions operate in a free market with minimm govermnent intervention. Financial Sector Reforms and Liberalization in Die Rcpublic of Korea: Current Sratus and Prospects 265 Having accomplished the initial economic goal, the financial industry is in a position to be fostered to meet the challenges of the new world economic order. To this end, financial institutions must first be allowed to operate under market mechanism to achieve higher efficiency in the allocation of financial resources, while government influence can gradually be reduced. Second, Korea must strengthen the competitiveness of the financial industry through continued innovation and scientific management. Third, Korea's financial industry must improve and modernize its micro- structure while training professionals to handle new financial techniques and products. Korea's Financial Deregulation and Market Opening The Korean government has pursued deregulation, liberalization, and intenationalizaton of its financial market. Since the early 1980s it has supported the privatization of the banking sector and the self-regulation of financial institutions. Last year a four-stage interest rate deregulation plan was announced, and foreign exchange management was converted from a positive to a negative system to encourage full-scale overseas activities. In particular, the government has augmented the competitiveness of the markets by applying an equal treatment policy to foreign financial institutions. The insuance market was fully opened in 1988, introducing twenty-seven foreign insurance companies to the Korean market. Since January 1992, the stock market was opened to direct foreign investment, and since then thirty-eight foreign securities firms have entered Korea. As of October 1992, about US$1.7 billion in foreign capital have been invested in the equity market. In the meantime, Korean companies gained permission to issue securities in overseas markets, and the limit on direct foreign investment in local stocks has been raised. The Korean govermnent's efforts toward financial liberalization are reflected in the Uruguay Round negotiation on financiaL services and in Korea-U.S. fnancial relations. The Korean government has actively participated in the Uruguay Round since it began. In January 1991 Korea became the twelfth participant from more than one hundred participants to submit the offer list to the General Agreement on Tariffs and Trade (GATT). During this last stage of negotiations, Korea plans to make further endeavors to lower global barriers in financial services. With regard to Korea-U.S. financial relations, the Korean government has held the Korea-U.S. financial policy taLks since 1990, and many issues raised by the United States have been resolved through this bilateral negotiation. Participation in the trust business, the elimination of ceilings on operating funds, and restrictions on branch establishments are some of the latest achievements of these talks. A detailed list of measures includes about twenty-four items (see the appendix). As a result of these actions the business environment for foreign banks has improved. For instance, the return on assets of foreign banks rose to 1.6 percent in 266 C/oon Taik Clung 1991, which is over four times the 0.4 percent for Korea's five major commercial banks. The T:orean government continues to make an effort to foster its financial industry in the spirit of the deregulation and opening policy through the Uruguay Round and the financial policy discussions. The Road Ahead Considering the current international econonic environment, such as globalization of the international financial and capital market and the progress of financial services negotiations in the Uruguay Round, it will be very difficult for the Korean government to maintain the current level of liberalization of its financial market and capital movement. Furthermore, because the Korean economy is growing rapidly and the restricted and oligopolistic structure of the Korean financial industry has resulted in inadequate services and high profits, foreign financial institutions may see many opportunities for effective and rewarding market penetration. The United States initiated financial policy talks with the Korean government in February 1990 to press fiuther financial sector liberalization, regulatory transparency, and national treatnent of U.S. fiancial institutions. All these factors will lead to Korea's rapidly and furer liberalized financial system in the near future. The government is now focusing more attention on a comprehensive liberalization of the financial system The government's plan to liberalize interest rates, initiated in December 1988, was largely abandoned several months later because of the sharp rise in interest rates. It was not until November 1991 that the government reintroduced a comprehensive interest rate liberalization plan to be implemented in four phases. This plan is a gradual process, with deregulation of most short-term deposits and govermnent instuments deferred until the latter two phases. Complete liealization is not envisaged until 1997 at the earliest. In early 1992 the government initiated a more comprehensive reform program and announced its three-stage Blueprint for Liberalization and Opening of the Financial Industry (see iable 14.1). The first stage, which was already implemented, includes measures to facilitate the Korean currency funding for foreign banks, to improve transparency in bank supervisory regulation, and to widen the range for foreign exchange rate fluctuations. The second stage, announced in June 1992, included plans to open the investment tust industry, relax restrictions on foreign exchange positions of banks, and internationalize the Korean currency. The third stage deals with long- term and structural issues in the development of Korea's financial industry. These include deregulating interest rate, relaxing lending regulations, opening a bond market, and liberalizing capital transactions, to mention a few. The Korean government plans to finalize the blueprint for these measures by the end of 1992. By continuing with such measures the Korean government plans to liberalize and open up Korea's financial industry to the level of developed countries in Fmsdaa Sector Reforms and Liberalizarlon n The Republic of Korea: Current Satus and Prospects 267 the near future. However, as the experience of some Latin America countries shows, what the Korean government worries about is the fact that hasty liberalization could invite undesirable side effects and destabilize the economy. As such, implementation depends on certain macroeconomic conditions, including narrowing the interest rate differential between domestic and international rates, restoring the balance in the current accounts of payment, and achieving lower inflation. 268 Chon Ta& C,ung Table 14.1. Blueprint for Liberalization and Opening of the Financial Industry Implemenitation schedule Items First stage 1. Expand CD issuance quotas, extend maturities, and lower (1992-93) minimum, denominations. 2. Improve the transparency of regulations. 3. Extend maturities on the short-term call market. 4. Allow securities companies to hedge fiuds against exchange rate risks. 5. Allow over-the-counter trading of domestic bonds. 6. Liberalize over-the-counter option transactions. Second stage 1. Allow more branches of foreign securities companies. (1994-96) 2. Open securities investment trust businesses. 3. Allow Koreans individual access in the foreign securities market. Macroeconomic conditions: Current account in balance, Korean- international interest rate differential narrowed to half that of 1992, below 7 percent. Third stage I- Complete the proposed libelization of interest rates. (1997-onward) 2. Open the domestic bond marker. 3. Eliminate the mandatory purchase of monetary stabilization bonds by foreign banlks. 4. Permit foreign banks to establish subsidiaries. 5. Expand limits on foreign investment in the stock market. Macroeconomic conditions: Current account in surplus, interest rate differential narrowed to 2 to 3 percent, inflation of 5 percent. rwncudl Seaor Refo msand Lieralization in Ihe Republic of Korea: Cwurenr Status and Prospects 269 Appendix Liberaization Measures Inplemented by the Government through Financial Policy Talks (FPT) Banking Measures taken to expand access to won (W) funding for foreign banks: * The W 12 billion ceiling on local capital was eliminated (April 1991). * The swap facility (preferential arrangement for foreign banks) reductions by 10 percent annually were suspended (June 1991). * The limit on CD issuance was raised on three occasions: to 175 percent of local capital (June 1991). to 200 percent of local capital (October 1991), and to 225 percent of local capitl (May 1992). * The scope of eligible trust business was expended to include specified money trust and non-money trust (May 1991). a The call market was integrated to equalize interest rates between Korean and foreign banks (May 1991). - The blind brokerage system was introduced to the call market (December 1991). * The establishment of ATMs on outside walls of branch buildings was permitted (Janary 1991). Measures taken to expand treatment of foreign banks: * Differential criteria on the acquisition of the trust business license were abolished (August 1991). The criteria included requirements for three years of operation in Korea, US$10 billion in head office capital, and employment of trust business specialists. * Differential criteria on the establishment of multiple branches were abolished (June 1991). The criteria included requirements for the bank's minimum of 10 years of operation in Korea, assets over W 300 billion, and ranking among the world's top 100 banks. * The multiple branch network was recognized as a single entity to simplify procedures on and management of ceilings, Enancial statements, and various approvals (December 1991). * A channel for dialogue between domestic and foreign banks was established for foreign bank participation in Giro/ATM networks (September 1991). 270 Cawon Talk Chung Measures related to liberalization of foreign exchange operations: * Underlying documentation requirements accompanying foreign exchange transactions were eased (July 1991). As a result, for spot transactions between the won and foreign cirrencies, instead of original documents the application for funds transfer may be presented as evidence of the transaction. For spot transactions between foreign currencies, no documentation is required as long as it is executed prior to the maturing of the spot transaction. * To ease restrictions on and widen the scope of foreign exchange operations, the requirement to maintain foreign exchange overbought position from 2 percent of the previous month's average purchase was gradually relaxed to the square position (July 1991, September 1992). * The multiple currency basket system for exchange rate determination was converted to the market avenge exchange rate system (MAERS) (March 1990). The range for daily exchange rate fluctuation was widened from within 0.4 percent of the previous day's rate to 0.6 percent (September 1991) and to 0.8 percent (July 1992). * The Foreign Exchange Management Act was converted from a positive to a negative system (December 1991). Securities Market Operations and Foreign Investment - Establishment of branches and joint ventures by foreign securities firms was permitted (January 1991), and ten brancbes and one joint venture have been subsequently approved. * Membership to the Korea Stock Exchange of foreign securites finms was allowed (June 1991). * The stock market was opened to direct investment by foreigners (January 1992) * The limit on foreign investment in stocks of foreign-invested firms or Korean companies with overseas securities issues was raised to less than 25 percent (July 1992). * The definition of foreign investor was widened to include foreign governments, pension fumds, and unit trust (January 1992). * The ratio of equity participation of foreign securities firms in domestic securities firms was increased from 40 percent to 50 percent (January 1992) * The four-stage interest rate deregulation plan was announced (August 1991). 15 CLOSING ADDRESS J. B. Swmarlin, Minister of Flnance, Indonesia Distinguihed fellow speakers and guests: As one who has been very much involved in the economic and financial reform in my own country, Indonesia, I am delighted to have been invited to give the closing address at this seminar. The quality of discussions reflects the caliber of this distinguished group of policymakers brought together this week from South America and Asia. Our common interest on this occasion lies in the challenge we all face in grappling with the complex and dynamic process of economic and financial reform. In seeking to chart our individual paths, there is much that can be leamed from the different approaches of other countries. Through their joint sponsorship of this seminar as a follow-up to the seminar held last May in Santiago, the World Bank and Bank Indonesia are to be warmly congratulated for facilitating this furtier sharing and learning process. You have already heard from previous speakers something of the detailed circumstances in their individual countries. I would like to focus on some of the broad lessons that not only ring true from Indonesia's experience, but also, I believe, have relevance for all of us here today. My starting point concerns the issue of economic stabilization, which I am sure we all agree is an absolute precondition of economic reform. An economy gripped by instability or inflationary pressures is highly unlikely to produce the circumstances conducive to successful reform. Excessive inflation grossly distorts the price incentives that are central to securing fundamental structral adjustment. Advances in the real economy are also held back as resources, especially scarce entrepreneurial talent, are diverted from efforts to boost efficiency and productivity to more lucrative speculative activities aimed at simply beating ination. Finally, high rates of inflation and instability are typically accompanied by high levels of capital flight, depriving the economy of the financial resources that could ease the reform process. Experience also teaches us that while stabilization measures, such as credit restriction or currency devaluation may symbolize resolving of a problem, their effectiveness is likely to be short-lived in the absence of steps to remedy more fundamental constraints. Structural adjustments, ranging from fiscal reform, trade 277 272 1. . Swvrfin liberalization, and deregulation to the creation of sound capital markets and financial intermediaries, are equally essential. Indonesia provides a good example of stabilization going hand-in-hand with structural reform. Stabilization measures constituted our immediate survival kit in the wake of the severe economic reversals brought about by the oil price crash and the international currency realignments of the mid-1980s. Yet, even as we struggled to stabilize the economy, we had already embarked upon the structural reforms necessary to secure renewed growth. Tax reform in 1983, some three years before the collapse of oil prices, laid the foundation for subsequent enhanced non-oil revenues. And once stability had been restored, we were able to move ahead swiftly with a wide-ranging program of economic restructuring. As a result, we have ushered in a transformation of our economy that has seen diversified manufacturing activity and exports, instead of oil as the primary engine of growth. This change has resulted in an annual average real GDP growth of 6.3 percent since 1986, well above our target of 5 percent while inflation has been kept within single digit figures. A second broad lesson from our experience is that countries need to be both competitive and creditworthy if they are to attract the risk capital needed to supplement goverment= funds for development. The urgent need to attract private risk capital and the increasingly stiff global competition for these funds reinforces the importance we in Indonesia have atached to structural and institutional reforms, removal of the ngidities and inefficiencies associated with a high-cost command economy, and providing scope for competition and market forces to work both in the financial and real sectors of our economy. it also explains our moves to implement swift remedial action before our hard- earned favorable credit rating with the interational lending community is in any danger of being compromised. For example, this occurred in 1990 and 1991 when investors turned increasingly to overseas commercial credit to fund new developments, including a number of very large state-related projects with long gestation periods and uncertain payback. This combination of factors not only brought additional pressure on our balance of payments but also threatened to erode creditworthiness for all our borrowing needs. Our policy response came in the form of rescheduling a mmber of costly projects, and setting up a high-level govermment team charged with coordinating all public sector related borrowing commitments. I am pleased to report that these responses have succeeded in restoring overseas commercial borrowing to manageable levels while helping reduce balance of payments pressures. Risk capital is, of course, also averse to high inflationary pressures, which is why in 1990 Indonesia embarked on measures to control growing inflation within the domestic economy. The very success of earlier reforms was propelling development at too rapid a pace. Domestically, flourishing aggregate demand, fueled by rapid credit growth, had pushed inflation to a disturbingly high rate. Tighter control of the money Closing Adress 273 supply was our remedy for reducing inflation to 4.9 percent rate in 1992, around half the 9.5 percent in each of the previous two years. Success in controlling inflation has enabled us more recently to cautiously ease the monetary policy, which we hope will boost business activity in the days ahead. Of course, measures of stabilization, adjustment, and reform are not ends in themselves but constitute the means for achieving longer-term goals. Although other nations may articulate them slightly differently, I suspect that our trilogy of goals in Indonesia-growth, stability, and equitable distribution of development gains-are not significantly different from those in your home countries. While each of these goals depends on the others, there are, I believe, particular lessons to be learned from experiences related to achieving equitable development. Important among these are the need to effectively communicate the benefits of reform at all levels of society. This is especially important where, for whatever reason, economic reform would otherwise give interests that are opposed to change the ability to negate what has already been achieved and thwart further attempts at reform. Just as important, if not more so, is the need for all sectors of society, the poor as well as the economically strong, to feel a genuine sense of sharing in the benefts of reform. Without doubt the most significant achievement of our development strategy in Indonesia has been the attainment of a substantial reduction in poverty. Whereas just twenty years ago some 60 percent of the population, nearly 70 million Indonesians, lived in absolute poverty, by 1990 the estimated figure had fallen to only 15 percent. Continued reduction in poverty is a major priority for our future development efforts. in addition to the broad lessons, there are issues related to the sequence and timing of economic and financial reforms. Developing countries are not short of advice from experts about the benefits of structural adjustment and the precise reforms that should be undertaken to achieve these benefits. What is often less clear, and where advice can be contradictory, is the timing and sequence in which reforms should be undertaken. While these are obviously crucial questions, it is doubtful that there are any definitive answers. I would suggest, however, that many critical lessons are to be found in the cumulative experiences of the countries represented at this seminar. The first lesson learned is the need for a case by case approach, which takes into account and reflects the full range of political, economic, social, and cultural realities in a country, together with its abilities to absorb and digest change. The very distinct differences in culture, national resources, infrastructure, institutions, and stages of development that exist between one country and another preclude the likelihood of a single prescription being equally applicable to all cases. The second lesson is that successful reform does not come about overnight. Transition takes time to achieve: time to plan and implement the adjustments necessary for reform, and time to sort out the mistakes that are sometimes associated with the tremendous undertaldng involved in securing fundamental change. In 274 1. B. Sumarlin Indonesia, we needed about twelve to thirteen years simply to lay the basic principles for economic reconstruction. Only after that we could seriously begin to forge closer integration with the international economy and use to our advantage the processes of globalized production and marketing. All this has taken time, almost twenty-five years, and we are not finished yet. In contrast with many other developing countries, our endeavors have benefited from political stability, which has facilitated a gradual or evolutionary approach to the timing of change. Gradualism has the further advantage of helping progressively win new constituencies for reform. As business people begin to recognize the benefits of a low-cost economy, new voices appear in support of further reform. The third lesson related to timing is that, paradoxically, government intervention is a necessary and vital component of the reform process. Centralized direction and control is essential if tansition to a more open economy is to proceed in an orderly fashion. A market economy will not be secured simply by allowing people the freedom to manage it in any way they deem expedient. Central coordination and direction are required for course corrections that may be needed from time to time to keep the transition process oL track. Just as important, deregulation does not mean no regulation. Prudential safeguards, which frequently can only be implemented and supervised by government, are necessary to ensure the soundness of key institutions and enhance protection of the public interest. This is particularly true for the financial sector if public trust and confidence are to be m-aintained. Having secured expansion and internationalization of Indonesia's banking system and capital marlcets through previous reforms, a major focus of more recent measures has been the strengthening of prudential oversight and supervision of our inancial sector. Let me conclude with a final lesson from Indonesia's reform experience and, I suspect, that of other nations. That is, reform programs are likely to be less effective [tan they might be in the absence of balance in the development process. In Indonesia we strive for a balance in all aspects of development: between different sectors of the economy, between public and private sector interests and goals, between new operational freedoms for business and prudental regulation, and between the needs of the individual and those of society as a whole. This, I believe, is well evidenced by the example of Indonesia's experience. It is throuigh this balanced, evolutionary approach that we see our surest route for achieving our development goals. Distinguished participants, in conclusion, as I commented at the outset of my remarks, the path of economic and financial reform is a complex and difficult route. I very much hope and believe that this seminar will prove to have been helpful to each of us as we continue to grapple with these complexities in the days ahead. We appreciate the efforts of the sponsors of this seminar, the World Bank and Bank Indonesia, to provide us the opporuinity to share our experiences and to learn from one another. Thank you for your kind attention. Distributors of World Bank Publications ARGENTINA ECWT, ARAB REPUBICOF KENYA SAUDI ARABIA, QATAR Carl_Hhsch,SRL AlAhm. Able BookSu WA." LW JaurblkSwe Cels!Guorm AIGI W QeiummHoumIdaumalSl PAIDDR3196 RPlMs 165.4th M-O. 4W445 cal pm. losiNG Rly.dk 471 133 3BuenosAkr NAro The MiddlefiniCbman SINcAPORETA!WAN Olrig. dil Llbrolntadmdonl 41,Shelf'Stut NOU,KEPUBUC OF MYANMAIBRUNEI Albert 40 Cao PMa Ko Bok COOSO Goww Ada Pudfuc IM LWL 1062 suau Abu P0.Bx 101, Kwawhum Golden Wha! DaUdlaig FINLAD Seoul 41. aJmg PddI% #0445 AUSTRALIA.PAPUA NEW GUINEk Akalbasn lKirJ.ppa Sam 1334 FIl. SOLOMON ISLANDS. P.O Box 128 KoranStck lok Ca VANUATU, AND WESTERN SAMOA SFO0101 Huldalig P.SoUl SOUTH APRICA.BOTSWANA D.A. lmadonServIce Yiddo Feagtmoms 64 WideonmeRoad FRANCE Sod OidedUnldvety Pro Mikla31sS2 Wom Bank PblIaftm SourAfAika Vktor 66iarawm.d'I MALAtSIA P.Q Om 1141 7S116 Pas U nh_SySM.IayoCopUV Cap. Tm800 AUSTRIA olm.pwpUiIId Carold and Co. GERMANY P.Ox.31127.JaPeafl Ferauht4dm eider Cuu_m31, UWN-Verag S170D KualaLapur atadSubcrtloaServik A-1IMI WI=. PoppIlUdorISAII..55 P.Q BN4V1 531153sm MEXCO UANGLADIESH mC Joua 2O24 MiTa lnduslrln Daarn GREECE Apead.llcetalS 3 AudstDASPapaolrouSA. 14080 U IaI_pu^Mm D. SPAIN HoeS. Rod 16 S.5mm Str. - Mudl-P,u.a U SA Dhawuoodi RIMA. 10662 Adm NETHERLANDS Ceauie'7 Dbaka 12De Di Louelboou/1C.Publlla 2J/OlMd HONG KONGMACAO PA Box2 BELCIUM AsaROD LW. 74W AE H _mkbluu LEsmslntaadobaaAEDC oJADOLannoy 44UW8WyndhwmSftM cafIdeCnts3 Ay. du RPAZ2 Wlnzdngcmdre NEVZEALAND 06 inS. 1060 Banus 7*1mw EUNZ L Cabal Hg.mgl PCgvate.1dlU914 31LANAANDTHMALDIVES BLAZIN Market Lake House Boohdap Publickcoe.Teua[RINICd_oSmkia1 HUNGAR Auddand P.O. Box24 um P1tI..toGomdk.20I POUSSo.IorMERWEL,mmy 10r . .Sla .paA. 0409 SaPaulo?,SP OESwh IUtI7-1 NIGERIA G.dlawMawal l-f11t7Bod Un llwdtyPtLnUS Coloeabo2 CANADA IlIhOUsm mgJmkhmo Le Dffuaur INDIA PrAlmt qSOP9S SBWEDIN 2SABouLde&*rJmgm Allied Publahms Maid. Ltd. tad. Furkzle BoucvilkQAc 751 Moutgod Aba Feckbvum%~ ,4_B56 MdrM-60 2 NOWAY RqwlpptanllBoxm1G Nave lI_ed. 54a3 rna RuaPAbwllngCo INXUE A B D 1294 AR;wna Raud ILIndiraLbidlad P.rBxw43luen Fw "nmwr Oauawa.Oaulo JulanBUoraluur20 N4120Me6 Wu-W UoAB IKU53WI PAlX i181 P.O.Uoul3O jakarta 1 FASSAN Sl.n o255dm. CSW MmrtllokAgency ChIna mKmndal&omak IRAN 6 !i;Shalwah.Qudd.qAzu SWI1IUAND PuaikldngHomu KowbabPumbUda PA s1au1 729 hrap*ehv S,DaFoaSI Dorge P.QA DU 157Fl 1 llPot Beft TehT Vmptm_le3112 Eru - a(Mn Lan COLOMBU IRElAND EdbM.IDrSIOIS Infolm Lad. GohlaffetSupplla.,scy Aplbdo34 reru uidec ApautadoAo 34W7 5HaRead LINal Lblalaot Bqpo4aDE Dublin 2 SwnlodaAbame_ns MEUJFFINSDpmlS COTE DlVOE ISRAEL im_Au DBokaer CH xmI Centre d'idltmet di D on YOraotUtlirm LAd. Sult 17%C1ylamid W AhicahstCWM r. LSIsm CGSdSmha T1owe THAIAND 4B..54U TedAwhC61560 AphlaAvau.FLV dm1. C6 = tnlDRSo I Akid j4 Flows CeE1md 306S ITALY mkLgmw wo CYPUS LcmaCom.nlukuudsrnt SPA CatwofApplld DamnS VIRDusDICIaeII11,F POLAND TRIDAD STOBAGO Cypma CoUCn OPflJSR b_Pdltuad fd mSydnIlcStudiesUnit 6.DogezrSbt.ethod S50n18Fr ULP.kns gwa S PD.lau2006 OOi W Wammaw Cusp. Nina. JAMAC T,dad We lardi Ian ladle PubIbbei Lad. Fraehawib.edr DENMARK 20601 Hop, load Irs Jooanmi UNiTD INDOMM Sauhmd.Uttwttr ICpto6 ULOna3 MiklckoLb. PRoamcarAlIItl _ 02416W.w P..lbos3 DK-IO FrudalbagC JAPAN AIoqHampsMmGU3e2F; E_s tukSerek POrTUCAL Erhad DOMINCAN IEPUKIIC HW _o3ClwmeB ukp4k 123 Uv=aPamtpl EdhoraTelleC.porA. Tokyo RmsDvCavo71174 M5ASWE Resauraisa lamb I Camics 30 1200DLhboe Logoimambm (Pt) LUdL AputdodceCoemo2190MZn ToI. Bo.Axdbemnds Saab DDoii- PBuST 12S Souwbn