tDI-25 iti 6 Economic Development Institute K;;.9 of The World Bank The Restructuring of-Financal Sytem in Latin -Amierica in-n--: Alan- Roe- and Pal . I Pop AN EDI POLICY SEMINAR REPORT *NUMBER 25 Policy Seminar Reports numbers 1 through 7 are 10 Politicas de Transporte en America available from the Economic Development Insti- Latina. Hernan Levy. tute of the World Bank, 1818 H Street, N.W., 11 Managing Financial Adjustment in Washington, D.C. 20433, attention: Ms. Edith Middle-Income Countries. Alan Roe Pena. and Paul A. Popiel. 1 Transport Policies in Francophone 12 Financing Higher Education in Fran- Sub-Saharan Africa: Issues and cophone West Africa. Angela Ransom. Options. Hernan Levy. EDI Catalog 13 The Role of Community Participation no. 705/016. in Development Planning and Project 1F Les politiques de transport en Afrique Management. Michael Bamberger. francophone au sud du Sahara: Problemes et choix. Herndn Levy. EDI 14 Korea's Experience with the Develop- Catalog no. 705/016. ment of Trade and Industry: Lessons for Latin America. Silvio de Franco, 2 Food Policy Seminar. J. J.0Price Alberto Eguren, and David Baughman. Gittinger. EDI Catalog no. 505/003. Also in Spanish. 3 Agricultural Policy and Its Relation- 15 Economic Adjustment in Algeria, ship to Food Policy in Sub-Saharan Egypt, Jordan, Morocco, Pakistan, Africa. Sakwa Bunyasi. EDI Catalog Tunisia, and Turkey. Alan Roe and no. 070/001. Jayanta Roy. 3F La politique agricole et ses rapports 16 Trade Reform and External Adjust- avec la politique alimentaire en ment: The Experiences of Hungary, Afrique subsaharienne. Sakwa Poland, Portugal, Turkey, and Bunyasi. EDI Catalog no. 070/001. Yugoslavia. Alan Roe, Jayanta Roy, 4 Development Policy Analysis. David and Jayshree Sengupta. G. Davies. EDI Catalog no. 420/043. 17 Como Recuperar el Progreso Social en 5 Management Training and Research America Latina. Enrique de la Piedra. for African Development. J. Price 18 Structural Adjustment in Sub-Saharan Gittinger. EDI Catalog no. 430/008. Africa. Cadman Atta Mills. Also in 5F La formation et la recherche en gestion French. pour le developpement de l'Afrique . 19 Issues in Development Management J. Price Gittinger. EDI Catalog in Sub-Saharan Africa. Ladipo no. 430/008. Adamolekun. Also in French. 6 Land and Water Resources Manage- 20 Land and Water Resource Management ment. Jacques J. Kozub, Norman in Asia. Peter Sun. Meyers, and Emmanuel D'Silva. EDT Catalog no. 070/003. 21 Strengthening Local Governments in Sub-Saharan Africa. The World Bank, 7 Export Policies and Administration. the Italian Ministry of Foreign Affairs, Randi B. Greenblatt with Joaquin and the Emilia Romagna Region of A. Cottani and Domingo F. Cavallo. Italy. EDT Catalog no. 400/047. 22 Adjustment with Growth in Latin The following titles may be purchased from local America. Alberto Eguren. distributors of World Bank publications (listed in the back of the book) or, if there is no distributor, 23 Structural Adjustment Programs and from the Publication Sales Unit, World Bank, Economic Stabilization in Central 1818 H Street, N.W., Washington, D.C. 20433. America. Eugenio Diaz-Bonilla. 8 The Political Economy of Reform in 24 The Theory and Practice of Agricultural Sub-Saharan Africa. Ravi Gulhati. Policy. Teresa Curran. 9 Transport Policy Issues in Sub- 25 Restructuring of Financial Systems in Saha*ran Africa. Hern, Levy and Latin America. Alan Roe and Paul Patrick 0. Malone. A. Popiel. AN EDI POLICY SEMINAR REPORT * No. 25 The Restructuring of Financial Systems in Latin America Alan Roe and Paul A. Popiel The World Bank Washington, D.C. Copyright ©5 1990 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing July 1990 The Economic Development Institute (EDI) was established by the World Bank in 1955 to train officials concerned with development planning, policymaking, investment analysis, and project implementation in member developing countries. At present the substance of the EDI's work emphasizes macroeconomic and sectoral economic policy analysis. Through a variety of courses, seminars, and workshops, most of which are given overseas in cooperation with local institutions, the EDI seeks to sharpen analytical skills used in policy analysis and to broaden understanding of the experience of individual countries with economic development. In addition to furthering the EDI's pedagogical objectives, Policy Seminars provide forums for policymakers, academics, and Bank staff to exchange views on current development issues, proposals, and practices. Although the EDI's publications are designed to support its training activities, many are of interest to a much broader audience. EDI materials, including any findings, interpretations, and conclusions, are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World B ank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequences of their use. Because of the informality of this series and to make the publication available with the least possible delay, the typescript has not been prepared and edited as fully as would be the case with a more formal document, and the World Bank accepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to Director, Publications Department, at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to photocopy portions for classroom use is not required, though notification of such use having been made will be appreciated. The backlist of publications by theWorld Bank is shown in the annualIndex ofPublications, which is available from Publications Sales Unit, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, Banque mondiale, 66, avenue d'Iena, 75116 Paris, France. Alan Roe is director of the Warwick Research Institute, University of Warwick United Kingdom. Paul Popiel is senior financial economist in the Trade and Finance Division of the Africa Technical Department, the World Bank. Library of Congress Cataloging-in Publication Data Roe, Alan. The restructuring of financial systems in Latin America / Alan Roe and Paul A. Popiel. p. cm. -- (An EDI policy seminar report ; no. 25) "Synthesizes the discussions ... at the policy roundtable on Managing Financial Systems in Latin America, held at Punta del Este in Dec. 1988"--P. vii. ISBN 0-8213-1616-8 1. Finance--Latin America--Congresses. 2. Economic stabilization-- Latin America--Congresses. 3. Financial institutions--Latin America--Congresses. I. Popiel, Paul A., 1938- . II. Title. HI. Series. HG185.L3R64 1990 336.8--dc20 9041538 CIP EDI Catalog No. 340/033 ISSN 1012-490X Foreword This document is one of a series reporting on policy seminars organized by the Economic Development Institute of the World Bank. Policy seminars provide a forum for an informal exchange of ideas and experiences among policymakers from different countries, leading experts in development, and World Bank staff with respect to major issues of development policy. Policy seminar reports focus on issues raised during seminars that may be of interest to a wider audience. They are not intended to be comprehensive proceedings. They seek, however, to convey the essence of the discussion that took place and to bring out any principal areas of agreement or disagreement that emerged among those participating. Christopher R. Willoughby Director Economic Development Institute of The World Bank iii I Contents Summary vii Preface xiii 1. The Context of Financial Adjustment in Latin America I 2. The Management and Financing of Public Sector Deficits 6 Deficits and Resource Allocation 6 Extemal Deficit Financing 8 Deficit Finance and the Healtl of the Financial Sector 8 3. The Management of Banking Sector Illiquidity and Insolvency 11 Causes 11 Effects 12 Solutions 14 4. Financial Sector Development: Policies, Issues, and Options 19 Public Sector Financial Institutions: The Pros and Cons 19 Development of Money and Capital Markets 20 5. An Overview of Some Main Issues 24 Annex A: Papers Presented at the Seminar 27 Annex B: List of Participants 38 v Summary This paper synthesizes the discussions that took place at the policy roundtable on Managing Financial Systems in Latin America, held in Punta del Este in December 1988. The roundtable aim was to consider some of the main issues and challenges facing policymakers for Latin America's financial systems. In that context, the three day discussions focused mainly on the problems entailed by, or related to, financial stabilization and adjustment. Three general problems, of particular topicality in Latin America today, were discussed: (a) management of the financing of the public deficit; (b) policies to deal with the insolvency and illiquidity of financial intermediaries; and (c) policies, issues, and options for further development of financial systems. As an introduction, an overview of the latest macroeconomic developments in Latin America put these problems into a global context. Hence, this paper comprises four parts and an overview, namely: * the context of Latin American financial stabilization and adjustment; * the management and financing of public sector deficits; * the management of financial institutions' insolvency and illiquidity; * financial system development: policies, issues, and options; and * an overview of some main issues. The Context of Latin American Financial Stabilization and Adjustment Tlhe decade of the 1980s has been an unfavorable period for Latin America. Overall the region has lost ground on the world scene. Its share of world exports has diminished from 15 percent in 1950 to about 5 percent today. With the exception of Africa and the centrally planned economies of Eastem Europe, Latin America is the only region that has been unable to take advantage of the rapid growth in OECD countries in recent years. For most of the countries of the region, the post-1981 adjustment process has been quite rapid with regard to current accounts and balances of payments, reflecting the abrupt reversal of capital inflows. The adjustment in current balances was the corollary of the emergence of net negative transfers of financial resources. The scale of these outflows was important, with the net transfer of financial resources averaging over 5 percent of Latin America's global GDP during the period 1982 to 1985. The result in most countries has been sharp adjustments, with initial decline in income per capita followed by slow growth. The majority of Latin America's financial systems have been adversely affected by this troubled macroeconomic environment, often compounded by financially repressive policies. The combination of credit subsidies, financial market segmentation, high inflation, and regulated, often overvalued, exchange rates with periodic large devaluations has proved extremely damaging and led to loss of financial depth. This unfavorable situation developed at a time when the importance of the financial systems in the domestic economies as mobilizers and allocators of financial resources had increased sharply. As foreign borrowing dried up, the domestic economies depended increasingly upon domestic financial savings. But with alternative assets available in the form of foreign currency holdings and real goods, and with domestic financial systems repressed, the flow of domestic financial savings turned negative in 1982 and continued so for most of the rest of the decade. vii viii The Restructuring of Financial Systems in Latin America The macrofinancial nexus is complex. The major links are between public sector deficits, the exchange rate and interest rates distortions they bring about, and the health of the financial system. The reverse causation runs from weak financial systems, bank failures, and subsequent bail outs to their macroeconomic consequences. Therefore, one major issue facing financial policymakers in Latin America is to improve the management of the financing of public sector deficits in order to relax the related constraints on financial intermediation and find the appropriate balance between financial liberalization and proper regulation and supervision. It is clear that excessive and inappropriate government regulations have retarded growth and the adjustment process, and yet there is a pressing need for a sound regulatory framework. Beyond the liberalization of financial systems and the phases of stabilization and adjustment it entails, there is the imperative need to develop money and capital markets thereby deepening financial systems and enhancing their efficiency at mobilizing and allocating domestic and foreign financial resources. Limitations on the choice of financial assets tend to discourage financial savings and foster capital flight. Banks can be said to channel lower risk financial resources through loans, while capital markets channel higher risk resources through equity and quasi equity. Hence, the degree of development of capital markets is an important determinant of the pace and breadth of investment. Management of the Financing of Public Sector Deficits The roundtable raised three main questions concerning this topic. First, taking a broad view of taxation, can the inflation tax be recommended as part of a taxation system? Second, what are the advantages and disadvantages of using extemal borrowing, as opposed to internal borrowing? Third, what is the relationship between deficit finance and financial policies in general? All real-world tax systems are second best in that they generate some distortional effects. Does it follow that the inflation tax, which also distorts the allocation of resources, should not be ruled out on its face as a component of a second best tax system? It was unanimously recognized that levels of inflation beyond 3040 percent were unacceptable as a means of generating revenue. There was some ambivalence, however, as to whether inflation at below 30 percent was acceptable as such a means. Eventually, it was argued that inflation is a zero sum game. If the govermment is effectively insolvent but continues to spend, the burden of its losses that translate in public deficits have to be allocated to agents within the economic system. Inflation is a mechanism that involves a country feeding upon itself and reduces both its income-generating capacity and its ability to fund expanding levels of government expenditures. To the second question, the general answer was that the best outcome that the region can presently expect from future extemal deficit financing is merely refinancing part of the interest due on the external debt. A sad lesson to be drawn from the record of the past six years is the high degree of vulnerability of the public finances of countries with a high dependence on external financing to variations in debt service associated with exogenous changes in world interest rates. Low administered interest rates, and other financial restrictions aimed at financing public deficit at a lower rate than inflation, have high social costs since they distort the economy in at least four ways. First, they establish a bias in favor of current consumption as opposed to saving. Second, they cause disintermediation, because savers may be encouraged to invest in their own low-yielding instruments. Third, they establish a bias toward excessive capital-intensive projects. Finally, they impose a high cost on the private sector, through the squeeze on private credit, and they hinder the growth of the financial system and impair mobilization and allocation of financial resources. Efforts to offset some of the worse consequences on the private sector credit squeeze through priority credit allocation to specified sectors are of dubious value, because they reduce the flexibility of the financial system and require that it be kept segmented and repressed. The Restructuring of Financial Systems in Latin America ix Management of Financial Institutions' Insolvency and Illiquidity The problem of financial distress, that is of financial institutions' insolvency and illiquidity, is acute in Latin America because of the generally high rate of inflation and high levels of debt. More specifically, the main causes of financial distress are poor macroeconomic and fiscal environment, unfavorable external circumstances, problems in productive sectors of the economy, and mismanagement in the financial sector itself. Macroeconomic policies that lead to boom-to-bust sequences defined as a period of excessive growth in demand associated with large public deficits and, often, with currency overvaluation followed by a period of stringent deflation and large devaluation, is a main cause of financial distress. This is because, during the boom phase, macroeconomic signals induce economic agents to take asset and liability positions that cannot possibly be sustained during the bust phase. Macroeconomic policies that cannot be sustained over a longer period of time are thus a potent cause of financial distress. Notable sectoral causes of financial distress were, among others, the collapse of commodity prices and the too sudden liberalization of trade. Both have an obvious adverse effect on most components of financial institutions' portfolios, which are dependent upon relative prices for their return. Mismanagement is one of the most frequent causes of financial distress, developing a dynamic of its own, often involving the following type of sequence within a financial institution or the financial sector. It may begin with technical mismanagement when, for example, a financial institution grows too quickly, over-concentrates its lending, or when its internal control mechanisms are weak. When an erosion of profits and capital results from these management errors, the next stage is cosmetic mismanagement. At this time, accounts are manipulated, and practices such as systematic rollover or unrealistic collateralization are indulged in, in the hope that circumstances may change and cause everything to correct itself. Once the illusion of optimism that underlies this second stage is progressively eroded, financial institutions' managements may tum to desperate management where they seek increasingly high risk operations in the hope that a few large profits will restore profitability and capital. The analogy here is with the losing gambler doubling his stakes on one last desperate spin of the wheel. There have been cases where this stage of mismanagement gave rise to fraud by all or part of the financial institution's management. Obviously, the initiation of the whole sequence will be more likely if macroeconomic and extemal conditions are such as to create considerable uncertainty about the returns on alternative investment opportunities. Similarly, it is most likely to become well-established where and when regulatory and supervisory controls on banks are weak. Once an economy has got into a situation of financial crisis, the consequences for its performance are both numerous and serious. First, a situation of financial crisis distorts financial resource allocation. Financial institutions with non-performing loans need to direct an increasing part of their new deposits and other resources to roll over these loans in order to bail out borrowers in difficulty, and eventually themselves. This process crowds out new productive investment and further increases the concentration of total lending for higher risk productive activities. Second, it puts pressure on interest rates. Interest rates are likely to rise to very high levels as depositors perceive the higher risks that they are asked to take in a situation where financial institutions are increasingly burdened with bad debts. Conversely, financial institutions in distress are overbidding for new funds to ensure a positive cash flow and adding upward pressure on interest rates. Third, it distorts the fiscal and monetary policies of the country following the various efforts that have to be pursued in order to alleviate the situation. Fourth, a situation of financial crisis impedes financial stabilization and adjustment. Efficiency enhancing reforms, such as deregulation of interest rates, or x The Restructuring of Financial Systems in Latin America deregulation of entry into financial markets-that may otherwise be desirable-become extremely hazardous if undertaken when the financial system is struggling with a situation of widespread insolvency. And finally, a widespread situation of insolvency and illiquidity in the financial system provokes general distrust in this system and also intensifies expectations about accelerating inflation as one of the devices to ensure its correction. Solutions to a widespread situation of insolvency and illiquidity are neither easy nor standard. The roundtable agreed broadly that elements of solution include the establishment of sound and sustainable macro policies to avoid further erroneous signals to resource allocation; the establishment of a clear identification of the other causes of the problem so that they can be eliminated; judgments about the manner in which losses should be allocated as between, for example, depositors, shareholders, and taxpayers; general measures of rehabilitation addressed at arrangements such as the accounting and supervisory practices applying to financial institutions; specific programs of restructuring of particular productive sectors and financial sector companies; and finally, liberalization and other reform measures designed to establish a sounder financial system for the long term. Financial Sector Development: Policies, Issues, and Options The last broad theme of the roundtable concentrated on longer-term institutional developments and the manner in which Latin American countries might seek to extend and strengthen their financial systems to better support longer-term economic development. Within this general theme, the roundtable focused on two sub-issues: problems faced by public financial institutions and development of money and capital markets. It was the roundtable opinion that, as a general premise, public ownership per se produces no special issue of concern; it is well-established in certain successful economies, such as France and Taiwan. However, two classes of problems commonly arise. First, management is often inadequate, in part, but not exclusively, because of the wide variety of noncommercial objectives imposed upon it and, as a consequence, because of the lack of clear incentives and criteria for performance. Second, ill-conceived macroeconomic policies, such as those involving directed credit programs, often use public finance institutions as main channels and so directly hamper their soundness. Development finance institutions might be able to withstand one of these two handicaps for a time, but the co-existence of these problems seems certain to be both debilitating and damaging, leading eventually to financial distress. In most countries of Latin America, the monetary and short-term functions of financial markets have developed better than the longer-term resource mobilization and allocation functions. In large measure, this relative lack of development of the capital markets can be traced to government policies favoring deposit-type instruments and lending arrangements that give too much incentive for the use of shorter-term bank facilities, at the expense of longer-term securities. In particular, the frequently negative interest rates on bank loans not only encourage their excessive use, but also contribute directly to the poor investment selection, weak portfolios, and general financial distress. These policy problems are, in turn, intensified and affect capital market development by a variety of structural features, particularly, the predominance of family-owned local enterprises, a frequently large share in total economic activity coming from foreign-owned subsidiary companies, and an excessive concentration of activity in some companies of the public sector. Further improvements in regulation are still needed in most Latin American countries to give financial intermediaries and investors alike clear roles about their rights and obligations. A balance is needed between the dynamics of an active market and the protection of participants, especially the interest of minority shareholders. The Restructuring of Financial Systems in Latin America xi In relation to the technical aspects of the operation of securities markets, it was noted that most markets in the region are still centered around one fixed point of trading, in the form of an exchange. However, the real time integration of markets to give coverage of areas not served by an exchange was becoming increasingly important in the more active markets, such as those in Brazil and Argentina. On the subject of supply of securities, it was pointed out that one of the major obstacles was the reluctance of smaller, and especially family-based businesses to issue securities and thus expose themselves to outside influences. Different fiscal and other incentives to float companies were discussed. Privatization of public assets was referred to as an important way to boost overall supply of securities. A variety of issues were discussed under the broad heading of demand for securities, including taxation arrangements and the role of institutional investors. Efforts should be made to encourage insurance funds, pension funds, and other institutional investors to structure the maturity of their investment to better match their liabilities and risks, through a broader recourse to capital markets. There was also potential for greater use of mutual investment fund arrangements that have contributed significantly to securities market development in a number of Latin American countries where investor understanding of portfolio investment in securities was relatively limited. I The Context of Financial Adjustment in Latin America In his introductory remarks to the roundtable, Dr. Pascale drew a connection between the first two themes of the conference, namely the management of fiscal deficits and the management of financial crises. First, he noted that the large para-fiscal deficits affecting most Latin American economies after 1980 had arisen in part from the excessively large rediscounts and guarantees provided by central banks and from other measures that had been taken to avoid the consequences of financial crisis in both the productive and the financial sectors of many of the economies. In this sense, the widespread preoccupation with external aspects of the debt crisis had distracted attention from the extremely important fiscal consequences, which in his view justified far more attention than they had been given. Thus, the present roundtable was both timely and important. He continued by noting that these fiscal consequences were vital to economic performance in that they condition both the options of policymakers regarding the policies that they can employ and the effects of these policies. In particular, they had forced the regrettable downgrading of the normal objectives of monetary policy, such as price stability, to enable the authorities to cope with their large fiscal deficits. Regarding the correction of deficits and adjustment, he argued that the choice was not about whether to adjust, since in the absence of a concerted policy a disorderly adjustment of some form would certainly occur with damaging effects on many aspects of performance. The real issue that needed attention was the design and sequencing of adjustment to ensure its reasonable efficiency so that output losses and adverse social consequences could be kept to a minimum. On the subject of the role of the central banks and bank supervision, he noted that the crises of the early 1980s had imposed a high cost on the Latin American economies and their financial systems. He drew several messages from this of which the most important was the need for improved vigilance and supervision on the part of central banks. Even the best managed central banks cannot entirely eliminate the prospects of crisis at certain times, but they are well advised to maintain systems to spot future problems and to have ready contingency plans to deal with these. He noted the example of the United States where the first actions of the new head of the Federal Reserve System included the commissioning of studies of the effects of, and possible responses to, the three worst things that could happen to the U.S. economy, namely a sharp fall of the dollar, the collapse of a major bank, and another crash in the stock market. This form of preparedness was not common in Latin America but it could certainly help to mitigate the consequences of future crises. If Latin American countries use greater consistency and clarity in their policies, however, they would be taking an important step in averting these crises. Several of these themes were further developed in the presentation by Mr. Guy Pfefferman. He began by surveying the recent performance of the Latin American economies, and especially those in the high-debt category like Brazil, Mexico, and Argentina. He noted that the negative net resource transfer from Latin America during 1982-85, which was the corollary of the impressive adjustment in the current account balance of payments, was truly enormous. For the continent as a whole it had averaged 5 percent of GDP during that period, which is double the transfer made by Weimar Germany and equivalent on average to one-quarter of domestic saving. The consequences for living standards and growth rates have been large and extremely painful. At the same time, progress in reducing public sector deficits had been less successful and while most countries had had to cope with increased public outlays for interest payments, few had offset this by reducing non-interest public expenditures. In fact, in most countries a major element of such expenditures, namely public wages and salaries, had increased. Mexico represented the clearest example of this, I 2 The Restructuring of Financial Systems in Latin America with public employment rising by 7 percent per annum between 1980 and 1985, while private employment declined, but Argentina, Brazil, and Peru exhibited some of the same symptoms. Since public revenue raising efforts were generally modest at best, the adjustment still required in relation to public deficits and their financing remained considerable for most countries. Chile and Bolivia had done more than most countries in the continent to curb government expenditures and reduce public deficits. Brazil had enjoyed rapid growth after the end of the recession in 1984, but was nonetheless still able to realize an impressive trade surplus, driven in the most recent year by a 30 percent expansion of exports. Although Brazil remains the most dynamic economy in the region, her dangerously high rate of inflation is undoubtedly distorting the behavior of both the productive and financial sectors. It is an open question as to whether an adjustment program, including a major curtailment of public expenditures, can be implemented to resolve this problem. In Mexico, recent growth performance had been far less impressive, and the small growth spurt before the June 1985 election has not been sustained. By sharp contrast with the case of Brazil, GDP fell by 4 percent in 1986, by 1 percent in 1987, and is likely to be lower again in 1988 because of lower oil prices. Against this and the long-established backdrop of inefficient investment, some positive signs are emerging in the form of a sustained pursuit of trade liberalization symbolized by Mexico's recent accession to the GATT. Mexico's close proximity to the United States and its very small size relative to its northem neighbors-only 3.3 percent in GDP terms-suggest that these moves could help to realize a unique and massive potential for trade expansion. The major mitigating factor is the 1987 shift of policy away from orthodox stabilization policies based on demand reduction to heterodox policies based on wage and price freezes. Although inflation has certairly come down rapidly under the tutelage of this policy, the costs have been high. In particular, the freezing of the nominal exchange rate has led to a real appreciation of one-third since end-1987, and this has caused imports to rise by 50 percent, to non-oil exports slowing down, and to the resumption of speculative capital flight. The large capital inflows from the United States, to the extent that they are not offset by capital flight, are helping to sustain the freeze. However, it cannot continue for much longer without undermining the major export potential of Mexican industry already referred to. At the same time, it will be difficult for the government to end the freeze without reigniting inflation and releasing the pent-up pressures for higher wages and increased public expenditures. In Argentina the post-recession spurt of growth came only in 1986, but this also proved unsustainable, with growth in 1987 of only 2 percent. Inflation remains a major problem with its rate accelerating after the failure of one heterodox adjustment process (the Austral Plan), and decelerating slightly in late 1988 after the introduction of another such package (the Primavera Plan). Despite the inherent difficulties with such programs, Argentinian policymakers, unlike their counterparts in Mexico, have succeeded in sustaining an exchange rate favorable to exporters and encouraging industrialists to seek expansion in exports rather than in the traditional domestic markets. As in several other countries in the continent, the key to real stabilization and resumed growth lies in the reduction of the public deficit. Argentina has an extremely bad record of revenue collection and seriously deficient controls over public spending, both of which need to be reformed if the endemic inflationary problems of the economy are to be resolved. As a final example, Chile is the nearest one gets in Latin America to a model of adjustment with growth. GDP is now a third higher than in 1983 and, by contrast with Brazil, this has been achieved with only moderate inflation. Exports have been expanding assisted by improved copper prices, unemployment is now less than 10 percent, and the absolute volume of external debt has declined from US$19 to US$18 billion. Most significantly, the debt service ratio has fallen from 50 percent in 1982 to only 23 percent by end-1988, which contrasts with the rise in the corresponding ratio in Argentina to its present level of 56 percent. The major problem still facing Chile is the The Restructuring of Financial Systems in Latin America 3 political one, since until a new system of democratic government is firmly established, there is no guarantee that the policies which have generated the recent successes will be sustainable. In his assessment of the Latin American position in the world scene, Pfefferman noted that the continent had lost a good deal of ground in recent years with its share of world exports; for example, declining from 15 percent in 1950 to only one-third of that figure today. A number of key parameters recently quantified by World Bank research were of considerable relevance in assessing how this and other aspects of the continent's economic perfornance might evolve in the next few years. First, a one percentage point change in the OECD growth rate translates into a 0.75 of one percent change in Latin American growth. Since OECD growth prospects are themselves highly conditioned by growth in the United States and by the manner in which the U.S. deficit problem is ultimately resolved, it had to be said that Latin American growth prospects are subject to most of the same uncertainties which affect U.S. prospects. It is possible that 1989 and beyond could see the U.S. economy in a continued "holding pattern" with growth rates remaining close to 3 percent. But several triggers, for example, a collapse of real estate prices, or a further stock market crash, could force a crisis adjustment that would take growth rates down to much lower levels. Second, a 1 point move in LIBOR has an opposing effect in Latin American growth equivalent to two-tenths of one percent. With higher interest rates a more likely short-term prospect than lower rates, this factor alone would suggest generally weaker growth in the near future. Obviously the combination of higher interest rates-as one of the policy devices of the U.S. adjustment-and recession would be an extremely serious one for Latin America and one that cannot entirely be ruled out. Equally, since most of the continent's foreign debt is a claim on the public sector, higher interest rates impact very strongly on the public deficit while lower rates may be reflected in increased public spending. Third, an increase in net capital inflow of US$10 billion translates into a growth increase of about three-tenths of 1 percent. At present, because of the generally low levels of private direct investment, new capital inflows are highly concentrated in the public sector. It is a reasonable hypothesis that measures which sought to reduce this concentration could render new capital inflows a more potent influence on growth prospects than is presently the case. Finally, a one percentage point change in the terms of trade translates into a change of one-tenth of 1 percent in Latin American growth. Clearly this factor also will be conditioned by many of the same uncertainties that affect prospects for economic growth in the OECD. However, since two major countries in the region, namely Mexico and Venezuela, are major oil exporters, the general calculus is rather different for this commodity and these two countries. It was noted that since the marginal production costs for oil were now in the range US$4 per barrel rising to US$8, the prospects for commodity exports and of these countries were relatively weak. In this general context, Guy Pfefferman noted that a safe planning assumption based on projections of production and demand would be for an underlying weakness in oil prices for the next ten years at least. This is obviously not an encouraging prognosis for the two countries mentioned. Other participants, however, noted that the strong prices in other commodities such as copper were providing some countries with advantages of negative real interest rates for the first time in several years. Turning to the main subject matter of the roundtable, he made three main points. First, the relationship between different possible components of adjustment programs are extremely strong and suggest the need for great caution in sequencing reform programs. The continent had already learned the hard way about the damage that could arise from trade or capital account liberalization in the presence of a misaligned real exchange rate. But other combinations of policy could also generate undesired results, and the appropriate coordination of policies, although theoretically possible, was a daunting and most difficult task in practice. Financial markets play a pivotal role in all this because they are affected very directly and quickly by policy distortions and imbalances in the macroeconomic system. Equally, a distressed or inefficient financial sector can seriously impede 4 The Restructuring of Financial Systems in Latin America the proper working-out of a reform program both by failing to ensure an adequate mobilization of saving, and by failing to switch resources into, for example, traded-goods sectors in response to a re-configuration of relative prices. Second, while it was widely accepted that financial deepening was an important part of the development process, the recent experience in Latin America has been one of generally increasing shallowness. In only four of the sixteen countries in the region for example, for which relevant data were available, did the money to GDP ratio rise in the period 1980-86. The 12 countries involved include the larger countries of Mexico, Argentina, and Brazil, where high inflation is almost certainly a part of the explanation. Policies inimical to financial sector development and especially credit subsidies, segmented markets, and regulated overvalued exchange rates are a further part of the story. Other factors were the additional public finance required when international interest rates rose and the debt crisis emerged in the early 1980s, along with real interest rates being pushed to "corrosively" high levels in some countries due, among other things, to the precarious state of banking systems and the enhanced risks perceived by investors. In the highly indebted countries there was emerging evidence that financial systems have been abused, and made substantially more shallow, by the inability to reduce public deficits sufficiently. This issue is taken up further in the next section. Finally, he noted the general inadequacy of capital markets in Latin America, and the biases in the organization of economic activity that stem from this. In particular, banks tend to channel only lower risk resources, and only through loans. A greater role for capital markets, which could channel higher-risk resources through equity and other instruments, could have favorable consequences for capital flows, could generate a greater role for indigenous businesses relative to the multinationals, and could help limit the role of the state in the productive system. Efforts in terms of an educational program, a change in regulatory arrangements, and a reform of the taxation arrangements would all be needed to bring about more active capital markets in most countries of the region. All these issues were debated extensively in later sessions of the roundtable and are discussed further in Section 4 of this paper. The concluding note from this brief overview was not entirely an optimistic one. In essence, it said that the core of the adjustment program for many of the countries of the continent was still needed, and would involve politically difficult reductions in public expenditure and an improved tax effort. However, several middle-of-the-road governments might find it difficult to harness the energy and the political support to implement this. Furthermore, the best prospects for the world economy context in which this had to happen was for growth in the range of only 2 to 3 percent. Not entirely ruled out was a scenario of far lower growth than this, with higher real interest rates and lower commodity prices. Further discussion focused on the management of the external debt problem of major debtor countries with the "ideal" solution to the problem being enunciated as a combination of improved macroeconomic policies, something like a doubling of capital inflows to a level of about US$20 billion per annum, and a degree of debt forgiveness. It was conceded, however, that there was no real appetite for debt relief in the major creditor country, namely the United States, and the prospects for this was, in any case, undermined by the knowledge of the large cumulative capital flight from major debtor countries. Equally it was acknowledged that international financial institutions such as the World Bank cannot find the resources to match the total capital flow requirements of the region. Similarly, while mechanisms such as debt-equity swaps and debt buy- backs are a necessary part of the methods to resolve the debt problem, their size and scope are inevitably limited. In Argentina, for example, while it was now possible to buy back debt at only 18 cents per US$1, the country simply did not have the foreign exchange to do this on a substantial scale. Finally, it was noted that the "populist" route to resolve the debt crisis as, for example, in Peru, was wholly inimical to the sound macro policies required as one leg of the solution to debt. The Restructuring of Financial Systems in Latin America S Peru, as a consequence, had derived zero benefit from the unilaterally declared limit on her debt service in 1984. 2 The Management of Financing of Public Sector Deficits The two presentations and associated discussion in this area addressed three main questions conceming public sector deficits. First, taking a broad view of taxation to include the inflation tax, how can the tax system be operated to minimize the misallocation of resources? And can the inflation tax be recommended to have a place in an optimal system of taxation? Second, what are the advantages and disadvantages of using extemal borrowing as a routine part of deficit financing, relative to intemal debt? Finally, and more broadly, what is the relationship between deficit finance and financial policies more generally? The background to this discussion is one in which central government deficits intemationally have shown substantial rises relative to GDP in the first half of the 1980s relative to the late 1970s. In Latin America this rise has been higher than the world average, with the size of the deficit rising to almost 6 percent of GDP in the 1980s as compared to an earlier figure of only 2.2 percent. This increasing competition for resources coming from the public sector has resulted in substantial increases in real interest rates, especially since it has coincided with a declining volume of world saving. In most countries the enlarged deficits have also been accompanied by an acceleration in inflation, and the Latin American and Caribbean region has suffered particularly badly in this regard with inflation accelerating from a 50 percent average in 1976-80 to 120 percent in 1981-85, and an even higher rate of 126 percent by 1987. Deficits and Resource Allocation At the theoretical level, it was noted that since all real-world tax systems are second best in that they generate some distorting effects, the inflation tax, which certainiy distorts the allocation of resources, cannot be ruled out immediately as a component in a second-best tax system. This implies that at the conceptual level at least, the inflation tax is not necessarily to be regarded as the 'residual" tax in the system. Equally, a rational decision to tax money holdings could be considered in conceptually the same manner as the tax on any other set of goods or services. In the paper by Roque Fernandez, this general line of reasoning led him to enquire to what extent inflation might be regarded as a regular source of deficit financing. Similarly, if the government was able to arrange occasional and unanticipated price-level changes, could the liquidation of debt resulting from this be also regarded as a periodic if not regular form of government financing? These rhetorical questions succeeded in provoking a heated discussion mostly directed to explaining the various negative consequences of high inflation. Some delegate participants likened inflation to a cardinal sin or to a virus that needed to be expunged if a truly free society was to be established and sustained. In particular, the idea of using periodic jumps in inflation to effectively expropriate wealth for the use of the government was felt to be at best a cynical and unacceptable way for any government to behave. Furthermore, it was self-defeating in that the public, not being generally foolish, would quickly adapt their behavior in response to these tactics in such a way as to severely erode the ability of any financial system to properly discharge its function of mobilizing and allocating resources.1 At a less emotive level, it was noted that the precise rate at which inflation is established ought to condition one's judgment about its acceptability as an element of deficit financing policy. At very 1. Some details of the mechanism whereby this erosion would take place are in the paper by Alan Roe and Paul Popiel referred to earlier. 6 The Restructuring of Financial Systems in Latin America 7 low levels of inflation, less than 5 percent say, no one is seriously inconvenienced by inflation and no adjustments of behavior to accommodate it are made. At moderate levels of inflation, of up to 20 percent, for example, some people notice the consequences of inflation and adjust their behavior accordingly. Since not everyone can succeed in doing this, an arbitrary redistribution of income is one consequence. Finally, at very high levels of inflation, everyone notices and adjusts to inflation, and efficient resource allocation is fundamentally disrupted. In particular, in this hyper-inflationary stage, the continued use of the inflation tax to finance large public deficits is certainly self-defeating since it has to result in a progressively higher rate of tax and a lower monetary tax base. The example of Brazil was mentioned as a specific case of the third stage of inflation. There the very high deficit to nominal GNP ratio and associated hyper-inflation had resulted in a massive decline in the money to GDP ratio. The government, which was no longer able to rely on the inflation tax to cover its deficit, had become progressively more dependent on adjustable bonds for its finances, and the amounts outstanding of such bonds were now equivalent to 12 times the monetary base. At the same time, since the financial sector was able to make substantial profits by dealing in these short-term bonds, it was rapidly losing its ability to make conventional business loans, especially those with longer maturities. The banking system, in other words, was becoming little more than a broker for government paper. The only eventual solution to this was now in the realm of politics rather than economics. Thousands of public sector jobs had to be cut, several ministries had to close, and a major privatization program was required to bring down the deficit and resolve the inflation problem. That savage cuts in public expenditures to reduce inflation are possible in Latin America was illustrated by the examples of Bolivia, where inflation was brought down from 24,000 percent to 10 percent, and Guatemala, where inflation of approximately 43 percent was reduced to only 9 percent by 1987. That the political moves to achieve this are far from easy for the policymaker is illustrated by the fact that leading policymakers in these and other countries were subjected to serious personal threat as they sought to unwind the populist expenditure programs that had fueled deficits. Participants at the roundtable unanimously recognized that the higher levels of inflation beyond 30 to 40 percent were unacceptable, and that levels of public deficits and the use of monetary creation that generated such levels of inflation were never to be recommended. However, there was some ambivalence as to whether inflation at around the 20 percent level was acceptable. A few delegates from countries such as Bolivia, which had struggled to reduce inflation, suggested that it was possible to live with inflation at this level. Others, however, noted that in their own countries, inflation, even at these moderate intermediate levels, gave rise to serious financial repression. The strongest statements in this area came from those with substantial experience outside Latin America. Some of them argued that inflation at any level was never to be recommended as a part of sound policy. In particular, Dr. Andrew Sheng drew a useful connection between the successes of the East Asian economies with their outward oriented trade policies and their generally superior performance in controlling inflation. Inflation, it was argued, is essentially a zero sum game. If the government is effectively insolvent but continues to spend, the burden of its losses (that is, the public deficits) have to be allocated to agents within the domestic economic system. In particular, the inflation tax allocates this burden to various groups, including those whose activities are channeled through formal financial markets. These certainly include the activities that generate extemal trade and so earn income for the economy from outside markets. In this sense inflation has the opposite incentive effects to a devaluation (it punishes the exporter) and should certainly be avoided if the country wishes to raise incomes and potential tax revenues from outsiders. Inflation is a recipe for preventing this; it is a financing mechanism that essentially involves a country feeding upon itself, and by so doing substantially reducing both its income-generating powers and the ability to fund expanding levels of government expenditures. 8 The Restructuring of Financial Systems in Latin America External Deficit Financing The paper presented by Dr. Roque Fernandez considered the options for inflationary finance, then reviewed the range of mechanisms available to the Latin American economies to reduce their present burdens of external debt. The general conclusion at the practical level was that the best outcome that the region can presently expect from future external deficit financing is merely the refinancing of part of the interest due. One lesson to be drawn from the sad record of the past six years is the extremely serious consequences for economic growth and investment levels that may be associated with a wrong judgment about the appropriate level of external financing to finance public deficits or for other purposes. A second lesson is the high degree of vulnerability of the public finances of countries with a high dependence on external financing to variations in debt service associated with exogenous changes in world interest rates. Professor Maxwell Fry's presentation further clarified several conceptual issues concerning dependence on external financing. First, it stressed that if foreign loans to the government are monetized, the inflationary consequences of the borrowing are equivalent to domestic borrowing from the central banks and, in particular, there will be some crowding out of private expenditures. The consequences are quite different if the foreign exchange proceeds of the loans are somehow made available to the private sector to relieve demand pressures on domestic expenditures. Second, beyond a certain point, enlarged foreign borrowing is likely to generate strong expectations of higher future taxation to finance the service of the debt. This in turn may result in the same crowding out of private expenditures that is associated with conventional taxation or the inflation tax, since private agents are likely to increase their own savings in order to meet higher expected future tax liabilities. Alternatively, if the strict conditions for this Ricardian equivalence are not satisfied, then it may well result in households moving their savings abroad in an attempt to evade future taxation. In this case, the savings function would shift to the left as external debt increased and interest rates would rise. In both cases the external financing of the deficit would result in some substitution of public for private expenditures. Third, while certain minimum levels of debt may well result in enhanced domestic investment, an increasing debt build-up is likely, as in the case of Latin America, to seriously deter investment through a variety of mechanisms, including the deterrent effect of expected future taxation and the interest rate rises connected with the higher risk premiums that are associated with large debts. At the level of policy, these various arguments establish that over-borrowing is certainly possible, and consequently that policies to restrict capital inflows may certainly be justified. On the basis of his own previous work, Maxwell Fry suggested that the various negative effects of a rising stock of extemal debt on growth begin to outweigh the positive effects of the flow of new debt when the debt to GNP ratio reaches about 50 percent and the debt to export ratio reaches about 2.4. At this point, if not before, some tax or other restrictions on capital inflows are called for to limit the malign effects. Although a number of participants at the roundtable expressed concern about the deterrent effects on foreign investment that this would cause, Maxwell Fry and others stressed that a laissez-faire policy toward foreign borrowing can only be justified theoretically using extremely restrictive assumptions. Deficit Finance and the Health of the Financial Sector The central proposition debated here was the familiar one that a low apparent cost of the finance used to meet a government's deficit is in no way indicative of the true social cost of that finance. It was noted that it is very rare in developing countries including those in Latin America for more than a small part of government deficits to be financed by voluntary sales of securities to the private sector. Instead governments have routinely sought to find ways of ensuring captive sales of The Restructuring of Financial Systems in Latin America 9 government debt or, alternatively, of lowering the relative attractiveness of the competing private sector securities. In the latter case, the desired result is achieved using a variety of instruments, such as transaction taxes, stamp duties, interest rate ceilings, and high reserve requirements on private sector financial institutions. Successful financial restrictions of these and other types together enable a given public deficit to be financed at a lower rate of inflation and with lower nominal interest rates. However, they usually impose a very high cost on the private sector through the squeeze on private credit that is involved, and they do, of course, hinder the growth of the financial sector and impair the functions of mobilizing and allocating resources. Efforts to offset some of the worse consequences of the private sector credit squeeze through policies such as subsidized interest rates or priority credit allocation to specified sectors are, in practice, of dubious value. In particular, selective credit policies and directed credit programs reduce the flexibility of a financial system and require that it is kept segmented and restricted. While there is no evidence that such approaches do anything to ensure the efficient allocation of scarce resources, it is readily apparent from Latin America and elsewhere that they increase the fragility of financial institutions by requiring them to expand their exposure in particular areas beyond the limits of what a commercial risk to return calculation would show to be justified. Hence, directed credit programs are certainly a major cause of the alarning levels of non-performing assets of financial institutions. These are discussed more fully in the next section. Overall, it was suggested that the stultifying of financial sectors in developing countries in the past two decades can be traced back to the desires of many governments to manipulate those sectors so as to be able to finance their own expenditures at the lowest possible interest cost. Obviously this problem has worsened as deficits have expanded in the manner described in Section 1. Specifically, low administered interest rates can distort the economy in at least three ways. First, they establish a bias in favor of current consumption rather than saving. Second, savers may be encouraged to invest in their own low-yielding investments rather than allow their saving to be intermediated through the financial system. Third, it establishes a bias toward excessive capital intensity of projects. Additionally, such policies are likely to have adverse effects on the equitable distribution of income even though the case for them is often argued in terms of the supposed equity advantages. Finally, when the rediscount mechanism of the central bank is used in support of directed credit programs at low interest rates, as is commonly the case, its effective use as part of a mechanism of monetary and inflation control is clearly undermined, since the central bank injects high powered money into the financial system. All this suggests that reserve requirements and other mechanisms of financial restriction are not to be preferred unambiguously to the inflation tax as a method of financing government deficits. Indeed, there is a tradeoff between these two categories of instrument that Maxwell Fry sought to make explicit through the use of a simple algebraic model. He noted that the available empirical evidence suggests that the growth foregone by virtue of sole dependence on inflationary finance becomes extremely large once inflation surpasses a rate of about 20 percent, that it, a figure generally below Latin American norms. Hence, maximizing the revenue from the inflation tax and eschewing the use of alternative financing routes were not to be recommended. The vicious circle whereby excessive reliance on credit expansion to finance a deficit generates inflation, which in turn reduces the size of the financing system, can be avoided by imposing restrictions on the proportion of the government sector borrowing requirements to be met from the banking system. If, for example, this proportion is restricted to 25 percent of total domestic credit, it is possible for an economy with the characteristics defined by his model to extract an inflation tax revenue equivalent to 5.7 percent of GNP. This proportion could in turn be raised if the constant across-the-board rate of reserve requirement was differentiated as between a lower rate for time deposits (10 percent rather than 25 percent) and a higher rate for the non-interest element of the money stock. In this case, inflation tax revenues could be increased to as much as 6.8 percent of GNP. Thus, the 10 The Restructuring of Financial Systems in Latin America principal of intramarginal rather than uniform reserve requirements can be shown to be less distorting and so can result in lower inflation for any given financing requirement. Equally, reserve requirements on capital inflows can eliminate the negative protection on domestic financial intermediaries and counteract some of the negative consequences of unrestricted capital inflows referred to earlier. In the discussion of these issues, the basic propositions of Maxwell Fry's argument were broadly accepted. However, there was again sustained comment that reliance on a high degree of inflation finance, albeit blended with other finance mechanisms, was ultimately unacceptable. Several participants argued that much reduced deficits that would preclude the need for inflationary financing ought to remain the cornerstone of policy, even though the prospects of achieving this at an early date in most of the participating countries might be remote. 3 The Management of Banking Sector Illiquidity and Insolvency This part of the discussion began with the fact that in developing countries generally, bank insolvency has become a widespread phenomenon, even an epidemic, and is likely to have consequences running well into the 1990s unless treated with determined and appropriate measures. These problems are acknowledged to be particularly serious in Latin America because of the generally high rates of inflation and the high levels of debt. In the main presentation by Mauricio Larrain, three main aspects of banking sector crises in the Latin American region were set out and explained, namely what have been the main causes of the crisis, what are its principle effects, and what are some of the main components of appropriate solutions. Causes It was noted that there are several different but related causes of financial crisis. They include poor macroeconomic management; unfavorable external circumstances such as deteriorating commodity prices; various types of mismanagement within the financial sector itself; and specific sectoral problems in various parts of an economy's productive sector. Concerning macroeconomic policies, it was noted that the boom-bust sequence, whereby a period of excessive growth of demand possibly associated with large public deficits and almost certainly with currency overvaluation is followed by a period of stringent deflation and large devaluations, is a sure recipe for financial crisis. The rise in the prices of fixed assets in the non-tradeable sectors of the economy and the generally associated increase in foreign currency liabilities, which has been a feature of the "boom" phase of this cycle in, for example, Chile and Argentina, provide signals for resource allocation which cannot possibly be sustained once the "bust" phase begins. Thus, the insolvency of some of the agents that participate in borrowing during the boom phase is inevitable, while the banks that have supplied their loans need to lose only a relatively small part of their portfolios before their capital base is effectively destroyed. At this later stage, the orthodox monetary policies of a government almost certainly have to be put aside to ensure the survival of the system. But, at an earlier stage, these monetary policies may themselves be part of the problem to the extent that they involve too much direction of credit and a concentration of credit in certain priority sectors which is excessive in relation to any objective assessment of the risks involved. The central proposition here is that the signals coming from macroeconomic management not only influence resource allocation, but also commit economic agents to asset and liability positions which need to have a long-term validity if financial crisis is to be avoided. Macro policies which are incapable of being sustained on a longer-term basis are thus a potent and central cause of such crisis. Regarding mismanagement as a cause of financial crisis, the paper by Aristobulo de Juan laid considerable stress on the dynamics of deterioration associated with mismanagement in the financial sector itself. The basic idea, which struck a chord with many roundtable participants, is that financial crisis often involves the following type of sequence within a bank or a banking sector. It may begin with technical mismanagement when, for example, the bank grows too quickly, concentrates its lending too much, or when its internal control mechanisms are weak. When an erosion of profits and capital result from these errors, the next stage is cosmetic mismanagement, when accounts are manipulated, and practices such as systematic rollover or unrealistic collateralization are indulged in in the hope that circumstances may change and cause everything to turn out right. Once the vague optimism which underlies this second stage is progressively eroded, 11 12 The Restructuring of Financial Systems in Latin America bank managements may turn to desperate management where they seek increasingly risky operations in the hope that a few large killings will restore profitability and capital. The analogy here is with the losing gambler doubling his stakes on one last desperate spin of the wheel. In a few cases this stage of desperate management may give rise to out-and-out fraud by all or a part of the bank management. Obviously the initiation of the whole sequence will be more likely if macroeconomic and external conditions are such as to create considerable uncertainty about the likely retum on alternative investment opportunities. Similarly, it is most likely to become well- established where regulatory and supervisory controls on banks are weak. Concerning the sectoral causes of crisis, it was noted that the collapse in commodity prices will have an obvious negative effect on all parts of a banking sector's portfolio which are dependent for their return on particular prices (for example, oil rig and drilling investments in Texas). In addition, the sudden liberalization of trade policies will have serious consequences for sectors whose profitability formerly depended on high protection. Aside from liberalization in the narrow field of trade policy, the discussion also directed a great deal of attention to liberalization in general, and especially financial sector liberalization, as a major potential cause of financial crisis. The participants observed, for example, that the interest rate fluctuations that characterized the liberalization experiments in the Southern Cone economies were large because the interest rate was effectively absorbing all the perturbations, including those caused by speculation in all the other markets. But these interest rate movements were clearly extremely damaging both for the liquidity and the solvency of significant parts of the productive sectors of the economies concerned. The participants also noted that the problems associated with the Southern Cone reforms arose in part because the banks had neither the knowledge nor the experience about how to manage in the new, liberalized environment. Thus, the likelihood of technical mismanagement of the type described earlier was certainly enhanced by these reforms. Although no general conclusions were drawn from these experiences, it was observed by Chilean and other participants that liberalization certainly ought not to be so abrupt as it had been in some countries in the region in the past, and it certainly needed to be accompanied by serious efforts to revamp bank regulatory arrangements to avoid the more serious financial sector consequences. Economic agents, including bank managements, needed to be allowed to adjust gradually to a new pattern of relative prices. In addition, in the metaphor of one participant, when the football field is made larger and flatter by liberalization, the penalties for breaching the remaining but essential regulations should be greatly intensified and not reduced. Effects Once an economy has got into a situation of financial crisis-defined as one where 50 percent or more of financial institutions are effectively insolvent-the consequences for economic performance are numerous and serious. These were encapsulated in the paper by Aristobulo de Juan as "Growing distortions in resource allocation, upward pressure on interest rates, a corporate culture with no sense of risk or disclosure, and growing losses in the system." Five main points expanding on this succinct summary were discussed in some detail. First, the need for banks with non-performing loans to direct an enlarged part of their new deposit and other resources to rollover facilities for borrowers who are themselves in difficulty (a process referred to as evergreening), will both crowd out new productive investments and further increase the concentration of total lending on high risk activities. While the instinct to buy time both for the bank and the client borrower is understandable, it is also misplaced if there is no serious prospect of the situation recovering. In this case it is little more than an exercise in allocating good money after bad. Furthermore, an economy which gets into this sort of situation assigns a The Restructuring of Financial Systems in Latin America 13 progressively larger part of its saving to projects that have no future and make no contribution to growth. Second, interest rates are likely to rise to very high levels as depositors perceive the higher risks that they are asked to take in a situation where bank portfolios are becoming increasingly burdened with bad debts. These high rates are also sustained from the demand side, since distressed productive sector companies will manifest an insatiable and interest-inelastic demand for credit to ensure a positive cash flow and keep themselves alive, while insolvent banks will be prepared to pay high rates to try to maintain their own liquidity. These high rates, when translated into lending rates, represent another reason why new and sound investments are crowded out and provide one more impetus for technical mismanagement in the banks in the form of the search for a higher rate with its attendant higher risk. Several participants noted that the deposit insurance mechanisms adopted from U.S. practice had generally proved counterproductive in these situations because they had encouraged even more reckless behavior than might otherwise have occurred. Hence, in the recent Chilean reform, deposit insurance had been removed except in the case of very small depositors. Third, the fiscal and monetary policies of the government are likely to be seriously distorted by the various efforts that have to be made to patch up the situation. If non-performing loans are already in the public sector, some direct response from the government is inevitable. It may, for example, need to provide rediscount facilities or subsidies to productive enterprises under its jurisdiction. Where the non-performing loans are in the private sector they are, nonetheless, of concem to government because of its duty to keep the financial sector alive. Hence, these loans may be directly absorbed by government or their consequences patched up by a variety of liquidity support through special discount facilities, special advances, subsidies to the banks, and similar methods. In almost all these cases, the true fiscal deficit of the government is raised, but since quasi-fiscal mechanisms are frequently involved, the true magnitude of the govemment's financing requirement is rarely made transparent. It was noted, for example, in the case of Argentina that the information about the true level and nature of the government's involvement in supporting distressed financial institutions is extremely poor, with the result that neither the true size of the deficit financing problem, or the real situation in the financial institutions, is apparent to more than a few key figures. A fourth effect that harps back to the earlier discussion about liberalization is that efficiency- enhancing reforms, such as the deregulation of interest rates or the deregulation of entry into financial markets that may otherwise be desirable, become extremely hazardous if undertaken when the financial sector is struggling with a situation of widespread insolvency. The experience in Argentina, for example, indicates that many of the new financial institutions set up after the financial sector liberalization experiment in 1977 were established for speculative reasons, to enable the promoters to ensure easy financing for their own industrial businesses, and to generally exploit or bypass the consequences of financial sector distress. The subsequent liquidation of 200 or more institutions had contributed significantly to the public's general sense of distrust and unreliability of the financial sector. A final effect that was discussed related to the fact that financial sector insolvency and distress provokes a general distrust of the financial system and intensifies expectations about accelerating inflation as one of the devices to ensure its correction. Thus, savers in a distressed economy seek outlets for their funds both in nonproductive inflation hedges and in capital flight. This in turn makes the forecasts of higher inflation self-fulfilling, since even with given deficits, the government now has to resort to higher rates of the inflation tax or enhanced financial sector restrictions to meet its financing targets. If deficits are enhanced because of the various bailout mechanisms commonly employed, the problem worsens commensurately. At the very least, a state of financial distress 14 The Restructuring of Financial Systems in Latin America makes the deficit financing problem substantially worse and contributes further to the process of financial disintermediation that is already associated with large deficits (see Section 2). Solutions The participants acknowledged in their discussion that the solutions to a widespread situation of financial sector insolvency are neither easy nor standard. The participants were assisted by both the extended presentations about the experiences of Chile and Malaysia and the paper by Aristobulo de Juan. Moreover, the participants broadly agreed that the components of the solution include the establishment of sound and sustainable macro policies to avoid further signals to resource allocation which are erroneous; the establishment of a clear identification of the other causes of the problem so that these can be systematically eliminated; judgments about the manner in which losses should be allocated as between, for example, depositors, shareholders, and taxpayers; general measures of rehabilitation addressed at arrangements such as the accounting and supervisory practices applying in banks; specific programs of restructuring of particular productive sector and financial sector companies; and finally, liberalization and other reform measures designed to establish a sounder financial sector for the longer term. Three subsidiary matters were also injected into the discussion as a prelude to the consideration of these main aspects of a recovery program. First, it was widely acknowledged that it is extremely difficult for the authorities to face up to the existence and scale of the problem, and, in particular, to the wide scope of the causal factors that have to be addressed if the situation is to be put right. Second, it is no real solution to try to inflate out of the problem. This is the approach that economic agents generally expect the government to adopt, but for reasons already given, is an element of the problem itself rather than part of the cure. Third, it is natural for policymakers to hope and even assume that economic growth will relieve financial distress and avoid the need for serious action. However, the reality is a vicious circle in which the numerous implications of bank insolvency make it highly unlikely that the sustained and high growth rates needed to achieve this happy result will actually be forthcoming. In the presentation on Chile by Dr. Guillermo Ramirez Vilardell it was noted' that the crisis in the financial sector that eventually broke in 1981 was associated with a major recession, the cessation of capital inflows, a situation of chronic over-indebtedness in large numbers of productive sector firms, and an estimated figure for irrecoverable assets in the banking sector equivalent to 200 percent of its capital and reserves. The recovery program was defined with a number of broad criteria in mind. One was that solvency should be generally restored within a period of two to three years, and that the "wait and hope" approaches should be avoided. A second was that recovery measures should not generally transfer the ownership of distressed organizations to the state (although in Chile, the government did take stakes in some banks), and that rescue mechanisms should be arranged to avoid any major rise in quasi-fiscal deficits. The across-the-board measures that were employed included the transitional use of total and expost deposit insurance to quiet fears and avoid runs on the banks.2 Second, loans were made to banks to encourage them to renegotiate their own loans and write off those that were recoverable. This form of recapitalization served to short-circuit the process whereby banks would otherwise have had strong incentives to continue to lend to failed enteiprises. In support of this, the central bank absorbed some of the bank losses associated with exchange rate changes. It also directly purchased some of the remaining bad loans from the banks. 2. One of the background papers made available to the seminar sets out a variety of pros and cons for this type of approach. See M. Hinds, Economic Effects of Financial Crises (PPR Working Paper 104, Washington D.C.: World Bank, October 1988). The Restructuring of Financial Systems in Latin America 15 Although all three of these categories of injection of resources into the banks might appear to be inflationary, in practice this need not be the case, since monetary expansion was in any case being used to finance the losses of insolvent firms. To the extent that the mechanisms established incentives for the closure of some such firms, it was anti-inflationary and a clear step toward a more stable macroeconomic environment. Finally, an innovative arrangement was established whereby the original shareholders of the banks were allowed to retain a stake in ownership even though accumulated bank losses exceeded the equity. In return they were required to commit their share of future profits to repurchase at par from the central bank those loans which that bank had itself acquired from the commercial banks as part of the recapitalization program. This encumbrance on the old shares drove their market price to a low level and thereby opened the way for the issue of new and unencumbered shares. It was noted that this approach was far less costly than a full liquidation of shareholders, not least because it established no real impediments to the introduction of new shareholdings. In practice, a deep change in the ownership of the banks did occur. In relation to specific measures, a central bank scheme to encourage several thousand small investors to invest in the banks was an important part of keeping the financial institutions in private hands. In addition, many of the bank-level reforms advocated in the paper by Aristobulo de Juan were introduced. In particular, strict norms for classifying loans by quality, rules to prevent banks accruing interest on non-performing loans, and improved arrangements for proper valuation of real assets, such as buildings, were all introduced. At the level of the regulatory authorities, far stronger obligations were imposed on bank supervisors, and these are now frequently and widely published, as are warnings to bank depositors, about the absence of any general guarantees for their funds. The result of these various measures had been encouraging. The solvency in the banking system had largely been restored with only 5 to 7 percent of loans now remaining problematic. At the same time, interest rate spreads had come down to the modest level of about 5 percent, and operational costs in banks brought down to a level of only about 2 percent. The intensification of the bank supervision process had given the government the ability to distinguish between well-managed and poorly managed banks. At the same time, the methods used to recapitalize the banks had encouraged them to pursue collectible debts but write off others or sell them to the central bank. In these various ways the balance sheets of banks were quickly able to provide a picture of their true financial health. Furthermore, the costs of the program had been quite modest given the magnitude of the problem from the outset in 1981 and 1982. The annual interest on the bonds issued by the govemment to purchase the bad debts of the banks was estimated to be equivalent to only 1.5 to 2.0 percent of GDP. However, the restoration of confidence in the banks and the domestic currency had together resulted in a considerable increase in the ratio of money holdings to GDP, and the sizeable revenues that the government had been able to collect as a consequence came close to matching the annual interest costs of the bonds. The presentation on the Malaysian financial crisis by Dr. Andrew Sheng began by tracing it to the collapse in the economy's growth rate in 1985 and 1986 in the face of a severe decline in the prices of major commodities, such as crude oil and palm oil, in the latter case to a figure below the costs of production. The resulting contraction in cash flow for many companies coming on the top of a period of booming property and stock market prices resulted in a dramatic decline in the growth of deposits in the banking system from a rate of 20.3 percent in 1984 to an annual rate as low as 3.8 percent by October 1986. Since loan demand failed to slow commensurately, the loan to deposit ratio of the bank rose to an historical high of 98.1 percent by mid-1986 as compared to an average of only 75 percent in the 1970s. The resulting tight liquidity and the collapse of property and share prices brought serious problems for many businesses, which now faced the triple squeeze of low asset prices, increased debt service obligations, and declines in income flows. These problems quickly translated into problems of bad debts and reduced profitability in the financial sector. 16 The Restructuring of Financial Systems in Latin America The Malaysian commercial banks, which had enjoyed a consistent double digit annual growth rate of assets from 1970 to 1984 and had built staffing levels accordingly, had also had little previous experience of bad debts: by 1984 the bad debt provision at only 3.5 percent of loans was similar in size to that encountered in, for example, the major U.K. banks. However, this rapidly rose, and by 1987 was approaching 13 percent. The margin on loans after deducting operating costs and bad debt provision also became negative for both the major commercial banks and the licensed financial companies by 1987: a sharp contrast with the considerable profitability that was achieved prior to 1985. However, public confidence in the financial system began to be eroded in eamest by the failure of a number of illegal deposit-taking institutions, with many directors and staff absconding with funds and at least 26 persons being charged for such offenses. In addition, the deposit-taking cooperatives (DTCs), which together had a membership of one million, began to face large-scale withdrawals as members withdrew funds to meet their own liquidity requirements. Many delays in payment were experienced. Massive runs against certain of the DTCs forced the promulgation of emergency legislation and eventually the suspension of the activities of most of the DTCs. Subsequent investigation revealed that 21 of the 35 DTCs were insolvent, with losses totaling almost 39 percent of their total assets. These and other cases revealed the disregard for good financial practices and persuaded the central bank, Bank Negara, that it needed to address the problem on at least three fronts. First, general monetary measures were needed to try to lessen the pressures of the tight liquidity and other features of the crisis on productive activity. Thus, measures such as lower reserve requirements, a more flexible interest rate policy, and some directives on bank wage costs were used to try to bring down real lending rates of interest. Second, a whole range of regulatory measures were adopted. These covered matters such as statistical reporting, capital adequacy requirements, guidelines about the booking of interest on non-perfonning loans, and greatly increased investigative powers for the regulatory authorities. Finally, a variety of rescue packages were put in place to deal with the immediate problems. In the case of the three worst affected banks, the approach involved a radical change in management, with first a revamping of the board of directors and then the appointment of experienced professionals to serve as chief executives. The principle that non-income earning assets must be supported by non-interest bearing liabilities was implemented by injecting funds to fully cover the losses of the years 1985 and 1986, including full provision for nonrecoverable loans of US$1,203 million. Existing shareholders were asked to subscribe a part of this through a series of rights issues, the Bank Negara injected US$672 million directly, and the balance was met through a series of subordinated loans. This injection had the same effect as that applied in the Chilean case in that it eliminated the incentive to the banks to rollover facilities to bad debtors, it stemmned losses, and it also clearly established the determination of the central bank to defend the viability of the financial system. An interesting feature of the scheme was the arrangement whereby the existing shareholders retained an option to buy back at par some of the shares purchased by the Bank Negara. In the case of the deposit-taking cooperatives, it was accepted that while the shareholders should expect little protection, the hundreds of thousands of small depositors (one in two of all Chinese households by one estimate), had to be protected from loss if faith in the financial system was to be restored. For 11 DTCs whose losses were relatively modest, the problems were largely resolved through the provision of soft loans from the central bank. For 13 other more seriously affected cooperatives, the rescue system was more complex. Although depositors were assured a $1 for $1 payment, approximately 50 percent of this was provided in cash and the balance took the form of an equity holding in a licensed financial institution or publicly listed company. The central bank acquired the net assets and deposit liabilities of 12 of the 13 DTCs through a small holding company. Altogether the rescue of all the ailing DTCs required M$1 billion in loans from the central bank in comparison with their deposit liabilities of M$1.5 billion. The seriousness of the problem The Restructuring of Financial Systems in Latin America 17 as well as the determined response to it are both indicated by the fact that 22 directors of eight DTCs were formally charged with criminal and other offenses. A final important feature of the Malaysian rescue plan was the arrangement to directly support the financial restructuring of productive sector enterprises. This involved the Bank Negara establishing an Enterprise Rehabilitation Fund capitalized at M$500 to help "resolve the overhang of stalled projects, release resources for new investment, create employment and conserve entrepreneurial skills." The operation of the fund, which sought to both assist the recovery of ailing but viable enterprises and the revival of non-performing loans, was made more effective by its ability to mobilize appropriate management expertise as well as seed capital. In particular, decisions about the nature and amount of assistance to particular companies were greatly influenced by "special turnaround groups," comprising experts in the field of manufacturing, trading, agriculture, and property. In other parts of the world, such funds have often floundered by having their decisions determined in an overly bureaucratic manner and by being too ready to provide new funds without firm assurances that the difficult changes needed to facilitate a real turnaround of the enterprises would be forthcoming. The first indications are that the Malaysian example has avoided at least some of these pitfalls. The first judgments about the Malaysian program of financial adjustment suggest that the prompt and decisive action of the Bank Negara described above stabilized the public's confidence in the fmancial system and prepared the environment for a sound recovery. In fact, the strong recovery of rubber, oil, and other commodity prices in 1987 helped the economy to a 5.2 percent growth of GDP that further eased the financial environment. Deposits once again began rising faster than loans and thereby eased the liquidity situation, enabling the central bank to introduce a variety of steps to reduce both the level of interest rates and interest rate margins in the banks. Growth was even higher in 1988, stock market prices began to rise sharply again, and strong sales and prices of certain types of property were also reported. Overall, the evidence is that the decisive reform measures, together with some recovery of fortunes in relation to commodity prices, have been sufficient to reestablish health and vigor in both the financial and the real markets of the economy. In the course of his presentation of the Malaysian experience, Andrew Sheng also drew several conclusions of general relevance to situations of financial sector crisis. One is that enterprise insolvency is often a direct result of government "insolvency" and in particular, where governments have covered financial deficits by printing money or, in other words by issuing promises to pay, the resulting inflation can be a direct cause of the insolvency problems. Furthermore, the government cannot resolve the situation and honor its own promises to pay until such time as it has taken the hard choices to get rid of inflation. Thus, a remedy for financial sector and enterprise insolvency based narrowly on Bagehot's prescription of "lend, lend, lend," is certain to fail because it will merely intensify the inflationary pressures. This, however, was exactly what had occurred in some of the Latin American countries other than Chile and had led to the loss of monetary control referred to in Dr. Pascale's introductory remarks. A second guiding principle was that in a war situation, because you cannot save all the injured, you must save those who have some prospect of helping you win the war. In a situation of financial crisis, this translates into the proposition that it is more important to save the productive enterprises than the banks and other financial institutions. Fortunately, there are normally relatively few banks and senior bankers in a typical developing country, and so their replacement or major re-orientation is a perfectly realistic possibility. This does not imply any lack of understanding of the extremely important role that banks play in the economy. It does, however, suggest that there are clear limits on the extent to which taxpayers' money should be used to subsidize banks that have become seriously unprofitable. It is also important to note that in crisis situations the most desperate of the productive sector enterprises are likely to seek and find the most desperate of the bankers. Thus, a quantitatively dominant part of the insolvency problem is likely to converge on the worst bankers. 18 The Restructuring of Financial Systems in Latin America If this is the case, it is even more important to adopt a very hard line regarding the future of failed banks and those who have managed them. Equally, as other participants noted, appropriate legal/criminal regulations are needed to discourage the reckless lending and borrowing that can so easily contribute to insolvency if it is not preempted. A third point is that the descent into crisis is analogous to a vacuum inside a black hole; it has the capacity to suck everyone into it. This is the major reason why those in the front line of the control process, namely the bank supervisors, have to be given considerable authority and adequate powers. Once supervisors have condoned the nonrecognition of the bad debt situation or some other dubious practice, then they are trapped, and reimposing sound practices upon the banks becomes far more difficult in the future. Thereafter, it is an easy matter for other authorities as well as politicians to get sucked into the process of condoning the unacceptable and allowing the process to slide gradually out of control. A final point is that the central challenge for bank supervisors in their attempts to preempt financial crisis is how to stop banks from creating liabilities based on paper assets. However, this becomes a progressively more difficult task, even for highly competent supervisors, as the efforts to secure a financial system progress. 4 Financial Sector Development: Policies, Issues, and Options The last broad theme of the roundtable concentrated on longer-term institutional developments and the manner in which countries might seek to extend and strengthen their financial systems to better support longer-term economic development, once the immediate challenges of financial crisis were under control. The discussion divides into two sub-issues corresponding broadly to the main presentations by Messrs. Fry and Horch. These two issues are discussed separately below. Public Sector Financial Institutions: The Pros and Cons Maxwell Fry began by noting that public sector financial institutions have proliferated through the developing world with nationalized banking systems and a variety of specialized development financing institution being the most common manifestation. As a general premise, public ownership per se produces no special issues of concern, and it is well established in certain successful economies such as France and Taiwan. However, two classes of problems commonly arise. The first, reiterated by several participants, is that management is often inadequate in part, but not exclusively because of the wide variety of noncommercial objectives imposed on managers and, as a consequence, the lack of clear incentives and criteria for their performance. The second is that ill- conceived macroeconomic policies such as those involving directed credit programs can often use public financial institutions as their channel and so directly hamper the sound workings of such institutions. The system might be able to sustain one of these two handicaps as, for example, in the case of Taiwan, where the overly bureaucratic management practices in the public banks have done little harm to economic performance because of the generally sound stance of macroeconomic policies. However, the coexistence of both of these problems seems certain to be both debilitating and damaging. His own comments and subsequent discussion both centered on two related issues. The first is the tendency for overly specialized financial institutions to be used to implement directed credit programs favorable to particular target sectors or groups. The second is the alarming proportion of non-performing loans on the books of the public financial institutions and especially the Development Finance Institutions (DFIs) in the developing world. Evidently both these issues relate back to the earlier discussion of financial crises where excessive direction of credit was identified as a major cause of such crises. It was also noted that the attitude regarding specialized financial institutions in organizations such as the World Bank, which had been supportive in the 1960s and 1970s, had changed and attitudes toward them were now far more critical. Various studies identified a large set of problems to which such institutions were subject, including limited flexibility, high administrative costs, insufficient attention to economic criteria in loan allocation, poor quality staff and too many of them, weak management, high delinquency rates, and generally poor supervision. The examples of the provincial banks in Argentina, with their poor compliance with regulations, and their very high proportion of non-performing loans, were noted to aptly illustrate several of these points. But a similar story can be told about public banks in Bolivia, Costa Rica, Peru, Venezuela, and Brazil, and in most of these cases these banks effectively preempt the central bank's rediscount facility. In general, the Latin American experience is that bank nationalizations invariably reduce both the efficiency and the flexibility of the banking system. While there was a good deal of agreement with these basic propositions and criticisms of public sector institutions, several comments were made defending, or at least explaining, the need for them. One set of comments, for example, suggested that there are a wide variety of non-bankable 19 20 The Restructuring of Financial Systems in Latin America but worthwhile projects in developing countries, and that public financial institutions are needed in order that these can proceed. In particular, in an economy that is trying to promote new forms of activity and changes in structure, publicly owned financial institutions have a special role which should not be abandoned without serious thought. Even in countries such as Colombia where progress had been made in making public banks more directly competitive with private banks, it was argued to be the case that specialized institutions for such things as agricultural finance were still required. Most of the replies to this line of defense took the position that even where subsidies to certain classes of economic activity were demonstrably justified, it did not follow that specialized financial institutions were also called for to act as the channel for such subsidies. In particular, specialized loans to, for example, small and medium enterprises have been successfully channeled via normnal commercial banks in several countries, and this practice could be applied elsewhere, especially if guarantees to support the more difficult classes of business could be provided. It could also be added that the World Bank's own experience is that commercial banks are more effective than specialized DFIs at channeling funds to a wider group of small and medium enterprises. Setting up a specialized institution for each special class of project was a dangerous practice, because every institution once established will naturally seek to fight for its continued existence even when the major reason for its existence may have disappeared. Difficulties would also arise because of the inherent unprofitability of certain classes of business, such as those involving very large numbers of small transactions, as in housing loans and popular savings schemes in Chile. Another line of argument stressed that a major problem with public financial institutions was the increasing body of special exceptions and selective arrangements imposed on their activities and, associated with this, the generally weak supervision of their activities. In some countries it was further suggested that financial distress had resulted in a gradual but unplanned expansion of governmental involvement in financial institutions as the assets and liabilities of troubled banks were taken over. In Argentina, for example, it was noted that two-thirds of the banks now have some form of governmental involvement, while in Chile, 64 percent of the total assets of the bank are "officialized" in one way or another. Some participants argued that public banks have more advantages than the private banks: savers and depositors are better protected and financial problems are more quickly spotted. However, the latter point was disputed by others who insisted that there was a need for a far greater transparency in the operation of most public banks, and also that, in general, they could be expected to be less aggressive and effective in pursuing bad loans than their counterparts in private ownership. Development of Money and Capital Markets The starting point for this part of the discussion was that in most countries of Latin America the monetary and short-term functions of financial markets have developed much more than the longer- term resource mobilization and allocation functions. In particular, the capital value of securities in issue in even the more developed securities markets in countries such as Argentina and Brazil lags behind that of countries of comparable income levels in the Middle and Far East. In large measure this relative lack of development of the capital markets can be traced to government policies favoring deposit-type instruments and lending arrangements that give too much incentive for the use of shorter-term bank facilities at the expense of longer-term securities. In particular, the frequently negative real interest rates on bank loans not only encourage their excessive use, but also contribute directly to the poor investment selection, weak portfolios, and general financial distress discussed in an earlier section. These policy problems are, in turn, intensified in their effects on capital market development by a variety of structural features such as, the predominance of family-owned local enterprises, a frequently large share in total economic activity coming from foreign-owned The Restructuring of Financial Systems in Latin America 21 subsidiary companies, and an excessive concentration of activity in some countries on the public sector. Given that both foreign firms and state enterprises are likely to face less restriction in their access to borrowing sources, the development of a strong and resilient corporate securities market might reasonably be regarded as one precondition for the preservation and expansion of a significant level of private sector enterprise. Such a market should be viewed predominantly as a means to expand options both for savers and investors and not as a substitute for the established banking sector instruments. For savers it can provide more choice along the spectra of risk, return, security, liquidity, and maturity. For investors, it offers greater choice about the maturity structure of liabilities and about the debt to equity balance. While the majority of participants accepted these broad propositions about the merits of securities markets as almost axiomatic, a small minority raised major doubts. One proposition based on the Argentinian experience was that the process of savings and capital formation seems able to proceed successfully in several countries, such as Japan, without active capital markets. Therefore, the participants asked where was the real evidence that capital markets contributed to this process? One line of reply to this was that it was erroneous to judge capital markets according to whether or not they succeeded in raising aggregate saving and investment. The real issue was their role in improving the mobilization of saving through the financial system and, from there, their role in facilitating a more efficient allocation of resources. Another comment was that while Argentina might be able to "get by" because of the high level of its direct or indirect state involvement in economic activity, it was hard to dispute the merit of and need for a greater availability of risk capital in an economy that wished to delegate more authority to private sector investment and production. Finally, attention was drawn to certain other cultural and economic characteristics of Japan, such as a low and stable inflation rate that has enabled that country to succeed with a credit- based and highly leveraged financial system in a way that other countries could not replicate. In relation to the preconditions required for the emergence and strengthening of securities markets, the paper by Dr. Hans Horch focused at some length on the required legal and regulatory framework, and on issues of market supervision. It also dealt with the problems of generating a sufficient supply of securities to the market, with the parallel problems of demand, and with a variety of taxation and special support measures that have been proposed and implemented in some countries. On the subject of regulation, further improvements were still needed in most Latin American countries to give issuers of securities, intermediaries and investors alike, the clear rules concerning their rights and obligations. A balance was needed between the dynamics of an active market and the protection of participants, especially the interests of minority-outside investors relative to those of the other shareholders and executive management. To this end, the greatest possible transparency of markets was needed as well as rules to ensure the full disclosure of relevant information, while at the same time remembering that disclosure to market participants rather than to the regulatory authority was the preeminent requirement. In commenting on this matter, participants from several countries conceded that a major problem impeding securities market development in their own countries was the absence of any substantial legislation to protect the minority shareholders, the associated absence of independent rating agencies, and, in the worse cases, a total and general lack of respect for minority shareholders. On the subject of the intermediaries involved in the market and their adequate supervision, Hans Horch noted that there were valid altematives to the traditional concept of individual brokerages with unlimited personal liability. In some Latin American countries, this model had already been fully substituted by the emergence of corporate forms of organization with their potential to mobilize larger financial resources and field a more diversified set of professional skills. The further liberalization of markets was now focused centrally on whether banks themselves should be allowed to come to represent an increasing and possibly dominant element in the securities market. 22 The Restructuring of Financial Systems in Latin Anerica There were advantages and disadvantages here and some evidence that the domination of banking intermediaries in countries such as the Federal Republic of Germany had not been conducive to the realization of the full potential of the securities markets. Hence, some restrictions are advisable to prevent dominant positions, and certainly a clear regulatory system is needed to exclude banks from areas of activity where conflicts of interest are likely to arise. At the very least this would require bank participation in the markets to be conducted via separate legal entities that are fully and separately accountable to the securities market authorities. Although there was relatively little discussion of these matters, comments generally endorsed the propositions made in the paper. In particular, participants noted that a relatively major role for banks in the financial intermediation process was a logical and desirable extension of the move toward universal banking that was being implemented in various countries of the region as a response to the generally critical comments about specialized financial institutions presented earlier. In relation to the technical aspects of the operation of securities markets, it was noted that most markets in the region still focused on a fixed point for trading in the form of an exchange. However, the real time integration of markets to give coverage of areas not served by an exchange was becoming increasingly important in the more active markets such as those in Brazil and Argentina. Although it was widely recognized that information technology had also created the potential for a greater international integration of markets, the realization of this at the Latin American regional level was still constrained both by foreign exchange restrictions and a limited degree of uniformity regarding taxation arrangements. Nonetheless, at least one participant drew attention to the greater sophistication of regulatory arrangements that was needed to deal with this, and to the possible need for a World Security Commission to oversee market behavior globally. On the subject of the supply of securities, Hans Horch reminded the seminar of the traditional reluctance of smaller, and especially family-based businesses, to issue securities in their companies, thus opening them up to outside inspection and interference. He also reminded them of the variety of mechanisms available to overcome these problems. The outstanding success of the Amman stock exchange in Jordan was a particularly interesting example in that it had been based on the granting of limited liability status only to companies prepared to offer at least 50 percent of their capital to the public. In the Republic of Korea too, the vigorous expansion in the market can be traced to a government requirement on companies to either issue shares or face credit restrictions, more rigorous tax inspection, and other restrictions on their operating environment. While these more dramatic remedies were not to be generally recommended for Latin America, other important institutional arrangements did warrant serious consideration. One possibility would be to equalize the reporting requirements for all limited liability companies irrespective of whether or not they allowed public ownership of shares. This would remove any concern that going public would necessarily involve the divulging of an increased amount of information. Another important step given the very high inflation prevailing in Latin America would be to remove capital gains tax from the inflation element of any profits on disposal of shares: a system which it was noted already applied in Chile and elsewhere. In discussion, one participant added that family owners of businesses will invariably be willing to surrender ownership if the price they are offered is attractive. However, when economic policies and circumstances result in securities prices being low relative to the underlying real assets, this willingness will obviously be muted. Privatization of public assets was referred to explicitly as an extremely important way to boost the overall supply of securities. Reference was made to the considerable number of flotations of this type in European countries such as Germany and France and, in particular, to the fact that the French privatization since 1986 has raised the capitalization of the Paris Bourse by an amount equivalent to 6 percent of GDP: a proportion greater than the existing total capitalization in Argentina and Brazil. The U.K. experience was even more dramatic. This route to expand the supply of securities clearly had great potential in Latin America, although a very careful selection The Restructuring of Financial Systems in Latin America 23 and preparation of possible candidates for privatization would be required, as would the same level of political commitment as in the main European economies. A variety of issues were discussed under the broad heading of the demand for securities, including taxation arrangements and the role of institutional investors. Concerning taxation, it was noted in the main presentation and confirmed from the floor, that most tax systems fail to come anywhere near to neutrality as between the different possible types of business finance. In particular, debt finance is almost always viewed as a deduction for corporate tax purposes while the dividends on equity normally have to be paid out of after-tax eamings. This represents a serious discouragement to the issuing of equity securities. Furthermore, in some countries, interest on deposit-type instruments are virtually exempted from tax, while interest and dividends on corporate bonds are heavily taxed. This represents an obvious impediment to the take up of securities instruments. Participants from Costa Rica and elsewhere confirmed the related proposition that dividends on equity are quite frequently taxed twice: first as corporate earnings and then as dividends paid to shareholders. Corporate returns have to be high before equity investments can earn a post tax return competitive with that on safe deposit-type instruments. However, in Chile it was noted that recent reforms have eliminated this double taxation. Finally, even within securities markets, it is quite common for government bonds to receive favorable tax treatment in the hands of investors, and so set a floor return that corporate paper often has difficulty in matching. In relation to institutional investors, it was Hans Horch's position that efforts should be made to encourage insurance funds, pension funds, and other institutional investors to structure the maturity and return on their investments to better match their liabilities and risks. This probably did not mean that they should be encouraged to invest primarily in common stocks given the state of securities market development in most Latin American countries. However, there was certainly potential for greater use of mutual investment fund arrangements that have contributed significantly to securities market development in a number of countries where investor understanding of portfolio investment in securities was relatively limited. At least one participant suggested that the institutional investor ought to receive far more attention in discussion of securities market development than was normally given. This was because of the great unexploited potential in this area-improved management of the region's pension funds would release a considerable source of demand-but also because of their disproportionate potential influence. In particular, where information disclosure was poor, institutional investors were likely to have the resources to exploit the gaps in information to their own advantage. 5 An Overview of Some Main Issues The final session of the roundtable involved a discussion that integrated a number of ideas, especially those concerned with the further development of domestic equity and other capital markets, and with the resolution of the shorter-term and current domestic banking problems. The lead paper by Mr. John Mathis stressed that the consensus of opinion on these matters was increasingly that governments should allow market forces to operate far more freely than in the past. The distortions arising from, for example, the governmental use of financial institutions predominantly to service their own financing requirements, and the widespread use of directed credit, lay at the root of the problems both of insolvent banks and insufficiently developed capital markets. The response now required from governments involved both macroeconomic policies to give greater price stability and ensure more flexible responses to changing market signals, and microeconomic policies that can facilitate a broadening of financial structures and the recovery of particular sectors in difficulty. An obvious problem here was that short-run measures to respond to immediate difficulties or crises could all too easily reverse the more substantial longer-term improvements in an economy's structure that the government was targeting. One way to avoid this was to establish the stability of policy objectives as a central plank of government strategy and eliminate the abrupt changes of policy that are a major source of business uncertainty and can contribute to undesirable outcomes such as capital flight. Keeping policies simple and limiting the range of agencies involved in administering particular policies were also ways to minimize the inconsistencies between different policies proved so counterproductive in some of the Latin American economies. This was a point further developed in Guy Pfefferman's concluding remarks to the seminar. He noted that although governments differed greatly in their abilities to operate complex systems of administrative intervention, few could do this with real success. Inflation reduction, implying lower government deficits, reduced reliance on overvalued currencies, and strong restrictions were obvious components of the reforms needed in most countries of the continent. In most countries in the region the root causes of financial problems could be traced to overly large public deficits. In relation to the problems at the level of commercial banks, it was repeated that these had been caused by a combination of over-regulation of interest rates and credit allocation; too much emphasis on debt-financing on the part of bank clients; and unsound financial practices, including inadequate supervision and regulation. In some countries the problems had become particularly severe, with the operating costs allowing for bad debt provision running at the equivalent of 20 percent of deposits, or four times the average for the U.S. banks. In other countries, the proliferation of non-institutional and unregulated markets merely indicated the extent to which the regulated markets had lost touch with market forces. Too many companies in the continent were now devoting their managerial talents to finding ways around the rules rather than focusing on how best to respond to new challenges in competitive markets. Specific policy reforms in most countries certainly included phasing out interest rate and credit controls, encouraging greater competition in banking by relaxing rules for entry, and reducing the government's role in total of banking activities. Better accounting practices, especially in regard to the treatment of bad debts, were also needed in most countries so that bank financial records could be relied upon as transparent and clear statements about the true financial health of the organization. As a part of this, capital requirements needed to be set at adequate levels and proper reserve provision was needed for non-performing loans. Stricter norms for granting credit were also 24 The Restructuring of Financial Systems in Latin America 25 required, as were restrictions on the amount of lending to any single sector, including the government. Furthermore, an independent supervisory body should have the power, and be expected, to audit banks to high standards, and close those found not to follow prudent practices. This work would be supported by the requirement on banks to provide regular evaluation reports on their portfolios with each loan graded in accordance with agreed categories of risk and performance. In relation to the development of equities markets, it was noted that the same factors that render banks vulnerable to excessive debt in their clients also hinder the development of deeper capital markets. Again, interest rate controls and taxation arrangements that favor debt instruments needed to be changed to correct this. Also required for more rapid equity market development is a better and more standardized set of reporting requirements that is enforced and audited on a systematic basis. In terms of institutional-support arrangements, an independent agency to rate the financial standing of public companies is needed as is a securities supervisory authority to foster a free market and impose tough penalties on malpractice, such as insider trading. It was noted that the liquidity of capital market instruments was a particular problem in Latin America and other developing countries. Of some 8,500 stocks listed in the International Finance Corporation (IFC) database on emerging markets, only 300 presently meet the liquidity requirements of international portfolio managers. In part, this problem could be addressed through an improvement in supervisory and reporting arrangements, as well as by extending the number of market participants by, for example, giving greater freedom to insurance and pension funds in the region to manage their own investment decision. However, it could also be facilitated by the use of country funds such as the Korea Fund, which was established in 1984. The IFC now manages 10 such funds intemationally with a total valuation of some US$10 billion. Inappropriate macroeconomic policies, as well as being a prime cause of problems for the commercial banks, also contribute to the sluggish development of capital markets. This is certainly a problem where those policies result in high inflation and associated high interest rates on safe assets, especially where the burden of these on lenders is muted by favorable tax exemptions or other subsidies. Particular mention was made of the difficulties arising in countries where efforts to reduce public deficits had precipitated the need for rescue operations for commercial banks, which had themselves been conducted using inflationary methods. The devices to avoid this, as described in the earlier discussion on Chile and Malaysia, were clearly relevant in this context. The final point on equity markets was that the efforts in some countries such as Mexico to use the privatization of state assets as one part of the move to freer markets, are clearly of great benefit for the faster development of equity markets. These benefits could be enhanced through favorable tax treatment of equities. The U.K. experience was a good example of how these combined actions can greatly extend the participation in the markets of a variety of small savers including the employees of the privatized companies. However, moves of this type made it even more important that arrangements for the protection of minority investors are strengthened as markets gradually expand. The final topic discussed was one that had not previously been touched on during the seminar. This concerned the role of direct foreign investment as a possible factor in broadening and strengthening financial markets. Participants noted that much foreign investment in developing countries is now in the manufacturing sector and is aimed at exporting, particularly to the investor's home country. Because such investment is widely perceived to provide substantial benefits, such as the transfer of technical, managerial, and marketing knowledge, the international competition to attract it is intense. Asian countries have so far done disproportionately well in attracting foreign investment because they have been able to combine reasonably attractive specific investment incentives with a macroeconomic and policy environment that is highly favorable to exporting. If the Latin American economies are to expand their share of investment activity, they will need to 26 The Restructuring of Financial Systems in Latin Anerica compete more effectively in all these areas. In some countries the failure to do so is connected with an excessively long and often contradictory set of policy supports;that makes the overall assistance to the investor more than a little confusing. Again, therefore, there is a strong argument for sound macro policies, relatively simple policies, and above all for consistency and stability in policies. If these basic preconditions can be achieved, foreign investment can provide a further route for the expansion of total equity markets, as well as being a spur to a generally more competitive internal environment, which is the theme developed in Dr. Paul Popiel's paper. This internationalization of the market may also be self-reinforcing if sound regulatory and intermediary arrangements are established, since in some of the emerging markets the movement of equity prices has been shown to have a negative correlation with the Standard and Poor's 500 index. So portfolio managers could well find some of the Latin American markets attractive in terms of increased portfolio diversification and reduced risk. However, none of this will be possible with unsound and unstable macroeconomic policies, with banking sectors that remain virtually insolvent, and with regulatory arrangements that are both lax as well as unfair to minority participants. Annex A Papers Presented at the Seminar (Enquiries concerning the availability of the following papers should be addressed directly to the authors.) Roque Fernandez, Management of Public Deficit Financing-Implications for Stabilization, Adjustment and Development of Financial Systems, Argentina, October 1988. Maxwell Fry, Professor of Economics, University of California, Managing Deficit Finance, California, November 1988. Maxwell Fry, Professor of Economics, University of California, Public Sector Financial Institutions in Economic Development, California, November 1988. Manuel Hinds, Division Chief, Trade, Finance & Public Sector Division, Technical Department, Europe, Middle East & North Africa Region, World Bank, Economic Effects of Financial Crises, WPS 104, Washington D.C., October 1988. Hans Horch, Auditec-Consultoria Financiera e Inversiones S.A., Development of Money and Capital Markets in Latin America, Hamburg, October 1988. Arist6bulo de Juan, Senior Financial Advisor, Financial Policy and Systems Division, Country Economics Department, World Bank, The Management of Situations of Illiquidity and Insolvency in the Banking Sector, Washington D.C., November 1988. John Mathis, Latin America: Policy Options for Domestic Financial Development, Arizona, November 1988. Paul A. Popiel, Senior Financial Economist, Finance, Industry & Energy Division, Economic Development Institute, World Bank, Recent Developments and Innovations in International Financial Markets, Washington D.C., November 1988. Paul A. Popiel, Senior Financial Economist, Finance, Industry & Energy Division, Economic Development Institute, World Bank, Financial Institutions in Distress: Causes and Remedies, Washington D.C., September 1988. Andrew Sheng, Senior Advisor, Bank Negara Malaysia, Financial Adjustment in a Period of Disinflation: The Case of Malaysia, Kuala Lumpur, August 1988. 27 Annex B List of Participants; Argentina Colombia Mr. Leonardo Anidjar Dr. Julio Manuel Ayerbe Consultant President Buenos Aires Corporaci6n Financiera del Valle, S.A. Dr. Pablo de Estrada Dr. Fernando Montes Negret Chief Advisor Deputy Govemor Banco Roberts Economics & Research Dr. Mario Luis Vicens Central Bank of Colombia Director Dr. Francisco J. Ortega Central Bank of Argentina Governor Central Bank of Colombia Bolivia Costa Rica Mr. Roberto Capriles Vice President Mr. Rodrigo Bolafios Zamora Banco Industrial General Manager National Stock Exchange Brazil Mr. Luis Libernan Mr. Paulo Franca General Manager Cabinet Chief Banco Interfin, S.A. Central Bank of Brazil Dr. Eduardo Lizano F. Mr. Antonio Chagas Meirelles President Pentagono DTVM Central Bank of Costa Rica Rfo de Janeiro Ecuador Economic Commission for Latin Mr. Jorge Gallardo Zavala America and the Carribean (CEPAL) Minister of Finances and Dr. Roberto Zahler Public Credit Senior Economist Mr. Joaqufn Martfnez Amador International Commerce Division Executive Vice President United Nations CEPAL Banco Continental President of Association of Ecuadoran Banks Chile Mr. Abelardo Pachano B. Mr. Enrique Goldfarb Sklar Chief General Manager Junta Monetaria Commercial Stock Exchange of Santiago Central Bank of Ecuador Dr. Guillermo Ramfrez Vilardell Bank Supervisor Guatemala Supervisory Agency for Banks and Financial Mr. Frederico Linares Institutions Executive Manager FIASE 28 The Restructuring of Financial Systems in Latin America 29 England Mr. Paul C. Luke Senior Financial Economist Libra Bank PLC London, U.K. Mexico Mr. Angel Palomino Manager Fmancial Analysis Bank of Mexico Spain Dr. Jose Luis Moreno Vice Dean Laguna University Uruguay Dr. Michele Santo Advisor Central Bank of Uruguay Ms. Beatriz Amoroso Advisor Central Bank of Uruguay Venezuela Mr. Jose Paris Coronel Executive Director Instituto Interamericano de Mercados de Capital Observer Asociaci6n Interamericana de Integraci6n (ALADI) Mr. Jaime Quijandrfa Secretary General Adjunct The World Bank Headquarters EDI POLICY SEMINAR REPORTS make CD 1818 H Street, N.W. available summaries of EDIpolicy Sd Washington, D.C. 20433, U.S.A. seminars that are of particular -rn interest to readers concerned with Telephone: (202) 477-1234 public affairs. The reports seek to Facsimile: (202) 477-6391 convey the essence-of the discussions Telex: WUI 64145 WORLDBANK and to bring out the principal areas RCA 248423 WORLDBK of agreement or disagreement O Cable Address: INTBAFRAD among the participants, who repre- WASHINGTONDC sent a wide range of governmental, academic, and professional back- European Office grounds. 66 avenue d'1ena 75116 Paris, France Telephone: (1) 40.69.30.00 Facsimile: (1) 47.20.19.66 g Telex: 842-620628 Tokyo Office - Kokusai Building . 1-1 Marunouchi 3-chome Chiyoda-ku, Tokyo 100, Japan Telephone: (3) 214-5001 Facsimile: (3) 214-3657 Telex: 781-26838 -~~~~~~~~~~~S