Open economy : tools for policymakers in developing countries

FILE COPY / (a Report No.:11615 Type: (PUB) Title: THE OPEN ECONOMY: TOOLS FOR Author: DORNBUSCH, RUDIGER Ext.: 0 Room: Dept.: EDI SERIES BOOKSTORE AUGUST 1991 tools for __ policymakers I n developing IC-5 countries nl tedited by EDI Series in Economic Development   The Open Economy 6'^ EDI Series O;,q in Economic Development  EDI Series in Economic Development Maxwell L. Brown, Farm Budgets: From Farm Income Analysis to Agricultural Project Analysis. Johns Hopkins University Press, 1979. James E. Austin, Agroindustrial Project Analysis. Johns Hopkins University Press, 1981. William Diamond and V. S. Raghavan, editors, Aspects of Development Bank Management. Johns Hopkins University Press, 1982. J. Price Gittinger, Economic Analysis of Agricultural Projects. 2d ed. Johns Hopkins University Press, 1982. Gerald M. Meier, editor, Pricing Policy for Development Management. Johns Hopkins University Press, 1983. J. D. Von Pischke, Dale W Adams, and Gordon Donald, editors. Rural Financial Markets in Developing Countries. Johns Hopkins University Press, 1983. J. Price Gittinger, Compounding and Discounting Tables for Project Analysis. 2d ed. Johns Hopkins University Press, 1984. K. C. Sivaramakrishnan and Leslie Green, Metropolitan Management: The Asian Experience. Oxford University Press, 1986. Hans A. Adler, Economic Appraisal of Transport Projects: A Manual with Case Studies. Revised and expanded edition. Johns Hopkins University Press, 1987. Philip H. Coombs and Jacques Hallak, Cost Analysis in Education: A Toolfor Policy and Planning. Johns Hopkins University Press, 1987. J. Price Gittinger, Joanne Leslie, and Caroline Hoisington, editors, Food Policy: Integrating Supply, Distribution, and Consumption. Johns Hopkins University Press, 1987. Gabriel J. Roth, The Private Provision of Public Services. Oxford University Press, 1987.  The Open Economy Tools for Policymakers in Developing Countries edited by Rudiger Dornbusch and F. Leslie C. H. Helmers Published for The World Bank Oxford University Press iii  Oxford University Press NEW YORK OXFORD LONDON GLASGOW TORONTO MELBOURNE WELLINGTON HONGKONG TOKYO KUALA LUMPUR SINGAPORE JAKARTA DELHI BOMBAY CALCUTTA MADRAS KARACHI NAIROBI DAR ES SALAAM CAPE TOWN i 1988 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. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Manufactured in the United States of America First printing May 1988 Fourth printing August 1991 The findings, interpretations, and conclusions expressed in this study are the results of research supported by the World Bank, but they 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. Library of Congress Cataloging-in-Publication Data The Open economy: tools for policymakers in developing countries / edited by Rudiger Dornbusch and F. Leslie C. H. Helmers. p. cm-(EDI series in economic development) Includes bibliographies and index. ISBN 0-19-520656-8 ISBN 0-19-520709-2 (pbk.) 1. Developing countries-Economic policy-Case studies. I. Dornbusch, Rudiger, 1942- . II. Helmers, F. L. C. H. (Frederik Leslie Cornelius Hazlewood), 1929- . III. Series. HC59.7.0529 1987 338.9'0091724-dcl9 88-5176 iv  Contents Preface ix 1. Introduction F Leslie C. H. Helmers, The World Bank 1. Policy Issues and the Main Policy Tools 2 2. Country Studies 7 3. Not Policy Miracles 9 Note 9 2. The Real Exchange Rate 10 E Leslie C. H. Helmers 1. The Real Exchange Rate as the Price of Foreign Exchange 11 2. Analysis of the Real Exchange Rate 12 3. Expenditure Reducing and Expenditure Switching 16 4. Floating and Fixed Exchange Rates 19 5. Effects of Changes in the Real Exchange Rate 26 Notes 32 Part I. Policy Issues and the Main Policy Tools 3. Balance of Payments Issues 37 Rudiger Dornbusch, Massachusetts Institute of Technology 1. An Overview of Linkages 37 2. Basic Accounting Concepts 42 3. Policy Instruments 45 4. Determinants of Growth 49 5. Some Mistakes To Be Avoided 51 Notes 52 4. External Shocks and Domestic Policy Responses 54 Paul Krugman, Massachusetts Institute of Technology 1. Anatomy of External Shocks 54 2. Policy Response to External Shocks 64 v  vi Contents 3. Additional Problems 74 4. Conclusions 78 5. Overvaluation and Trade Balance 80 Rudiger Dornbusch 1. Effects of Overvaluation 80 2. Definitions of the Real Exchange Rate 82 3. Effects of Disturbances 86 4. Goods, Factor, and Asset Markets 91 5. Country Experiences 96 6. Exchange Rate Rules 101 7. Undervaluation 104 Notes 107 6. Devaluation and Inflation 108 Stanley Fischer, Massachusetts Institute of Technology 1. Exchange Rate Arrangements 108 2. Exchange Rate Indicators 114 3. How Does Devaluation Work? 117 4. Assets Market Problems 121 5. Devaluation and High Inflation 122 Notes 127 7. Multiple Exchange Rates, Capital Controls, and Commercial Policy 128 Susan M. Collins, Harvard University 1. Overview of Policy Alternatives 129 2. Multiple Exchange Rates and Trade Taxes 132 3. Quantitative Restrictions 140 4. Capital Controls 144 5. Dual Exchange Rates 148 6. Conclusions 154 Appendix 1. Jamaica in the 1970s 155 Appendix 2. Capital Flight 157 Appendix 3. Black Markets for Foreign Exchange 158 Selected Bibliography 160 Notes 163 8. Exchange Reserves as Shock Absorbers 165 John Williamson, Institute for International Economics 1. Reserve Composition 167 2. Purpose of Holding Reserves 172 3. The Norm for the Reserve Stock 175 4. Speed of Adjustment 178 5. Conclusions 182 Glossary 184 Selected Bibliography 185 Notes 185  Contents vii 9. External Borrowing and Debt Management 187 Albert Fishlow, University of California, Berkeley I. History of Capital Movements 188 2. The First Oil Price Shock 196 3. The Second Oil Price Shock 202 4. Coping with the Debt Crisis 205 5. Prospects for New Capital Flows 210 6. Policy Lessons 217 Appendix. Debt Dynamics 220 Selected Bibliography 221 Notes 222 10. Opening Up: Liberalization with Stabilization 223 Michael Bruno, Hebrew University 1. Sequence of Liberalization Measures 224 2. Liberalization of Trade 230 3. Liberalization of Financial Markets 237 4. Stabilization from High Inflation 239 Selected Bibliography 246 Notes 247 11. Policymaking and Economic Policy in Small Developing Countries 249 Arnold C. Harberger, University of California, Los Angeles, and University of Chicago 1. Some Simple Demographic Facts 250 2. Demography's Hidden Curse 251 3. Professionalism 254 4. The Lessons of Experience 256 Notes 263 Part II. Country Studies 12. Argentina 267 Domingo F Cavallo, Fundacion Mediterranea 1. Foreign Terms of Trade and Commercial Policies 271 2. Short-Term Cycles and the External Sector 275 3. The Real Exchange Rate for Exports 277 4. Speculation and Hyperinflation 282 Notes 283 13. Brazil 285 Mario Henrique Simonsen, Fundacoao Getilio Vargas 1. From Coffee Valorization to External Debt 285 2. Indexation Issues 288 3. Balance of Payments Policies 295 4. Conclusions 304 Selected Bibliography 306  viii Contents 14. Indonesia 307 Malcolm Gillis and David Dapice, Duke University 1. Historical Overview 307 2. A Tale of Three Devaluations 317 3. Trade Policy 324 4. Conclusions 328 Appendix. The Real Exchange Rate and the Index of International Competitiveness 329 Notes 333 15. Korea 336 Yung-Chul Park, Korea University 1. Exports as an Engine of Growth 337 2. Disadvantages of the Strategy 341 3. Conclusions 345 Selected Bibliography 346 Notes 347 16. Mexico 348 Eliana A. Cardoso, Fletcher School, Tufts University, and Santiago Levy, Boston University 1. From 1935 to 1982 348 2. Industrialization and Trade Policy 355 3. Exchange Rates from the Mid-1920s to 1982 358 4. Budget Deficits and Inflation 362 5. Recent Policies 364 Selected Bibliography 368 Notes 369 Appendixes A. Sources of Current Data 373 Rudiger Dornbusch B. National Accounting Identities 375 F Leslie C. H. Helmers 1. Gross Domestic Product and Gross National Product 375 2. The Balance of Payments Equation 380 3. Examples of the Basic Balance of Payments Equation 384 Notes 390 C. Real-Exchange-Rate Indexes 392 F Leslie C. H. Helmers D. Effective Protection 398 F Leslie C. H. Helmers Index 405  Preface THE ECONOMIC DEVELOPMENT INSTITUTE (EDI) of the World Bank has several objectives. One is to teach officials in the developing world the principles and practices of project analysis. Another is to familiarize government officials with current issues of economic policy. To meet these objectives, the EDI organizes courses and seminars in cooperation with its partner institutions in the developing world. In addition, the EDI makes its materials available for independent study. In principle, EDI materials are written in nontechnical language so that they will be understandable to interested persons outside the economics profes- sion. This volume was produced at the request of EDI. The editors drew up an outline of the aspects of economic policy concerning an open econ- omy and assembled a team of economists to write about the issues. The book does not offer specific policy prescriptions for every conceivable situation a country might face, but the writers give valuable advice based on their experience. Although intended for use in courses and seminars at EDI and its partner institutes, the study should also be of interest to government officials and policymakers in the developing world who want to have an overview of the various policy issues. Profes- sional economists may also find it useful. The editors would like to express their thanks to the fourteen other authors who contributed to this volume. They all welcomed with enthu- siasm the idea of producing in nontechnical language a guide to the issues facing an open economy, and all met the deadlines for their own contributions. Numerous persons gave support or advice or constructive critical comments. Unfortunately, the list of indebtedness has become too long to be produced here. An exception may be made, however, for Sonia Hoehlein, Carman Peri, and Marshall Schreier, who processed the man- uscript in record time. ix  x Preface One of the editors, F. Leslie C. H. Helmers, died on March 3, 1988, just before this book went to press. An economist's economist, he brought economic analysis to bear on topical problems of public man- agement in such a way that the noneconomist could understand and apply the concepts developed, and he helped developing countries build their own capacity for economic management. His contributions to this volume and to the work of the Economic Development Institute and of the World Bank are gratefully acknowledged.  1 Introduction X F Leslie C. H. Helmers DIFFERENT ECONOMISTS writing about a subject inevitably have, in addition to different styles, nuances, and judgments, different percep- tions of the economic background of the reader. In addition, they some- times use different terminologies. To give the interested noneconomist reader a basic understanding, this book begins with the present over- view, and a discussion in chapter 2 of how the real exchange rate is used as a policy instrument. Four appendixes deal with data sources and basic economic concepts. Readers familiar with real-exchange-rate analysis and basic economic concepts need not read chapter 2 and the appen- dixes. No summary can do justice to a writer's views. The following para- graphs therefore present only a few salient aspects of the various chap- ters in this volume. Part I deals with policy issues and policy tools. In chapter 3, which sets the tone, Dornbusch derives the basic balance of payments identity, which provides the unifying theme for the following chapters. The identity equation shows that a current account surplus has three guises. It is at one and the same time (a) the excess of the nation's income over its expenditures, (b) the excess of the nation's exports of goods and ser- vices over its imports, and (c) the net increment to the nation's foreign asset holdings. If national income exceeds domestic expenditures, then the surplus manifests itself as an excess of current foreign exchange inflows over current foreign exchange outflows, or (in other words) as a surplus in the current account of the balance of payments. Conversely, if domestic expenditures exceed national income, then current foreign exchange outflows exceed current foreign exchange inflows, and the current account of the balance of payments will show a deficit. Surpluses or defi- cits in the current account of the balance of payments of course mean that the country's foreign assets position has improved or deteriorated, respectively. It is also self-evident that, if there is a surplus, the country I  2 F. Leslie C. H. Helmers must build up its foreign exchange reserves, lend abroad, or invest abroad. The converse also holds: if there is a deficit, foreign exchange reserves are being drawn down, or foreigners must finance the deficit. The equality of the three separate ways of measuring the current account surplus or deficit is an ex post identity that holds at every and any level of domestic activity. Labor may be unemployed or fully employed; exports and imports may be at low levels or at high levels, depending on the openness of the economy. In all cases, the basic iden- tity will be valid. The interesting aspects appear if one analyzes how one moves from one equilibrium to another. Basically, this book considers the different policy measures that can be taken to move from one situa- tion to another and the different effects. In principle, three types of policy measures, corresponding to the three guises of the current account can be distinguished. Expenditure- changing policies, such as fiscal and monetary policies, directly affect the level of economic activity. They act on the national income and domestic expenditure part of the basic equation. Expenditure- switching policies, such as trade and exchange rate policies, change the pattern of economic activity. These policies lead to changes in the com- position of production, spending, and foreign exchange flows. Finally, financial policies toward the rest of the world concern capital flows, debt management, and the net foreign assets position of a country. In addition, there are the so-called structural policies, the objective of which is to enhance the efficiency of domestic production processes. These policies are not, however, discussed in this volume because this study deals principally with balance of payments issues. The order of the chapters in this book follows closely the above- mentioned typology of policy instruments. After chapter 3, which pro- vides the basic framework, Krugman in chapter 4 gives an overview of expenditure-changing and expenditure-switching policies. In chapters 5, 6, and 7, Dornbusch, Fischer, and Collins explain issues of real exchange rate, multiple exchange rates, and trade policies. In chapters 8 and 9, external financial policies are discussed by Williamson and Fishlow. In chapter 10, Bruno discusses the order in which the different policy instruments should be used. Finally, in chapter 11, Harberger provides persuasive documentation that small economies do as well as large economies and that professional economists' policy prescriptions lead to better economic progress. The five country studies in Part II review the types of policies the countries have actually followed in the recent past. 1. Policy Issues and the Main Policy Tools After Dornbusch sets the stage in chapter 3, Krugman starts in chapter 4 with a detailed account of the type of external adverse disturbances a  Introduction 3 country can encounter and then reviews the difference between expenditure-reducing and expenditure-switching policies. The objec- tive of both types of policies is to transform a deficit in the current account of the balance of payments into a surplus. The former policies have the disadvantage that cutting domestic expenditures may cause the economy to enter into a recession and that consequently unemployment may increase; moreover, private investment may be reduced to such an extent that future growth may be jeopardized. In contrast, expendi- ture-switching policies, which increase the domestic prices of the inter- nationally traded goods, will curtail imports and stimulate exports so that the economy can continue to grow. The problem here, however, is that the price increases of imports as well as of export goods sold in the home market may lead to a demand for wage increases. This demand may lead to an inflationary spiral, which may make import goods again attractive and production of export goods less profitable, so that the effects of the policy will be negated. It becomes immediately clear that the objective of expenditure switching is not only to increase the nomi- nal prices of the traded goods but also to increase their real prices. The real exchange rate therefore emerges as an important policy instru- ment. Krugman sees the need to combine nominal increases in the prices of traded goods with expenditure-cutting policies to curtail domestic demand or to reform the labor market to permit increases in real prices. He recognizes that political reality may lead policymakers to stray from policies of strict economic efficiency. In chapter 5, Dornbusch discusses the question of overvaluation. He introduces a number of real-exchange-rate indexes. Basically, each of these indexes tries to measure in real terms the international competi- tiveness of the domestic producers of traded goods. A very powerful concept is to define the real exchange rate as the ratio of domestic wages, say, pesos per hour, to the nominal exchange rate in pesos per dollar. In this case the real exchange rate is defined as simply the domes- tic wage in dollars. If wages in dollars are high, then domestic producers of tradables will have difficulty competing with imports and producing export goods because neither type of activity will be very profitable in competition with the world market. A high real exchange rate, as defined here, means that the exchange rate is overvalued: the high value of domestic currency, reflected in the high wage rate in dollars, leads to high imports and low exports. This definition is very powerful because it quickly conveys an understanding of how difficult it some- times is to move from an overvalued currency (with real wages that are above the level at which labor supply equals labor demand) to an equilib- rium level with significantly lower wages. In the remaining part of chapter 5, Dornbusch emphasizes again the importance of the real exchange rate as a policy instrument. He also dis- cusses in detail how a domestic currency can become overvalued and  4 F. Leslie C. H. Helmers reviews all the possible adverse impacts on the economy. In addition, a number of detailed country experiences are reviewed. His conclusion may be put in this way: no automatic mechanism will ensure that exchange rates will not become misaligned. The real exchange rate must therefore be considered an important policy guideline. Fischer's chapter 6 complements Dornbusch's review in chapter 5. After a brief discussion of different exchange rate arrangements, Fischer provides helpful guidelines to determine whether the domestic currency is overvalued. Also, Fischer points out that a devaluation must be accompanied by restrictive macroeconomic policies to ensure that domestic price and wage increases do not offset the effects of a nominal devaluation. According to Fischer, "every successful stabilization pro- gram has been preceded by an unsuccessful attempt in which the gov- ernment sought to stabilize purely by fixing the nominal exchange rate, without taking accompanying macroeconomic measures." In chapter 7, Collins discusses multiple exchange rates and quantita- tive restrictions. These policies have been introduced for a variety of reasons. The objective behind special nominal exchange rates or special restrictions, for example, may be to favor certain food or energy imports and to keep prices in these sectors low. It may be to stimulate the domestic production of some types of exports or some types of import substitutes or to protect an infant industry. Countries may also intro- duce a series of special rates and restrictions in hopes of curtailing total imports and capital outflows and improving their balance of payments. Collins points out, however, that other effects of these policies can make them very harmful. In particular, they alter relative prices, and they affect the government budget. They create distortions in the incentive system and can be very difficult to enforce, thereby encouraging illegal activity. Collins concludes that these special measures are not appropri- ate remedies for large and persistent balance of payments deficits. They will not, for example, enable a country with a substantially overvalued exchange rate to postpone devaluation indefinitely but may in fact exac- erbate the problem. Collins identifies some situations in which the spe- cial policies can be useful. Capital controls, for example, can provide an effective buffer against disruptive speculative capital flows. In chapter 8, Williamson deals with foreign exchange reserve poli- cies. When a country experiences, say, an export boom, how much of the extra foreign exchange receipts should it add to its foreign exchange reserves and how much should it spend on imports? The answer depends on how far the country may wish to diverge from the long-run equilibrium path-that is, the path along which the country produces its normal, maximum output level (internal balance) while the balance of payments is also in equilibrium (external balance). Williamson leads us step by step through the different considerations,  Introduction 5 such as the size of the external shock, the costs of reserve depletion, the opportunity cost of holding reserves, the speed of adjustment, the type of exchange rate regime, the structure of the balance of payments, the opportunity for foreign borrowing, and so on. In the final analysis, the speed of adjustment is a very important factor because countries that cannot adjust quickly will need to hold relatively large reserves. The strategy proposed by Williamson is that, except in special circum- stances, countries should target reserves at 30-40 percent of a year's imports. In addition, a continuous review should seek to determine whether divergences from internal and external balance are occurring and whether within a time frame of five years the reserve targets can again be reached in cases of shortfall. In chapter 9, Fishlow provides a historical overview of international capital movements and makes the important point that, from a histori- cal perspective, the present debt of the developing countries is not high. Its maturity, however-six to ten years-is much shorter than that of the pre-World War I debt. Furthermore, the debt is expressed in U.S. dollars, and much of it has floating interest rates. For these reasons and others, several developing countries at present have problems paying off their debts. In general, the developing countries coped well with the first oil-price shock of 1973, but they had problems with the second oil-price shock of 1980. The reasons were that since 1980 countries worldwide had fol- lowed restrictive monetary and fiscal policies, which led to a decline in the developing countries' export earnings at the same time as interest rates rose worldwide. The developing countries continued to borrow, but the new debt was used to a large extent to service the old debt at higher interest rates. Often when there was some inflow of foreign exchange, capital flight emerged and caused much of it to disappear. Serious debt servicing problems started to appear in 1982, especially for Mexico. The response consisted of a rescheduling of debt combined with restraints on domestic demand, mainly on investments. It became clear, however, that continuation of expenditure-reducing policies would lead to intolerably low consumption levels. Fishlow there- fore welcomes the 1985 plan of U.S. Treasury SecretaryJames Baker, which again stresses growth. Specifically, the Baker plan seeks for the developing countries an enhancement of the productivity of domestic assets through liberalization and through more capital lending from external official and private bank sources. Fishlow endorses the overall thrust of the plan but believes that more resources than planned should be made available to the fifteen debtor countries in greatest trouble and that a greater diversity in internal liberalization strategies should be welcomed because this will ensure a greater internal commitment to the implementation of liberalization. He sees the need for an active import-  6 F. Leslie C. H. Helmers substitution strategy in developing countries to curtail imports. (Appen- dix D offers a different point of view.) In the final part of his analysis, Fishlow also makes the point that financial openness cannot be pursued as a substitute for effective adjustment policies. Capital inflows may temporarily resolve balance of payments deficits, but a country must effectuate real adjustments in order to solve the problem in the longer run. Numerous studies have shown that growth will be higher in countries that follow open-economy policies than it will in closed economies. The explanation for this phenomenon is basically very simple: when the economy is opened up, domestic production processes will become competitive with those of the rest of the world, thus ensuring enhanced efficiency. Suppose we are at a relatively closed stage, characterized by high tariffs, quantitative restrictions, foreign exchange controls, and so on. How should we go about liberalizing the economy? Bruno addresses this question in chapter 10. Bruno makes the important point that adjustment in the financial markets may be very fast, whereas the response of exports and import-substitute producers to changes in the real exchange rate tends to be sluggish. Rapid liberalization may thus lead to high unemployment costs, and in such cases Bruno favors a grad- ual approach to the liberalization of commodity flows while maintaining controls on capital flows. The latter are necessary because massive short-term capital inflows may lead to an unwanted appreciation of the domestic currency, which will result in increased imports and reduced exports. In Bruno's view, the wrong order of liberalization of markets has in many cases caused crises, and he argues strongly that the liberal- ization of goods markets should precede the liberalization of capital markets. Bruno also reviews the stabilization attempts in high-inflation countries. In such cases, he concludes that the approach should consist of policies that provide not gradual disinflation but very fast disinflation, because prolonged contractionary monetary and fiscal pol- icies will entail substantial unemployment. Finally, in chapter 11, Harberger argues persuasively that small developing countries (as a cutoff rate he takes countries with a 1983 population of less than 20 million) should economize on the use of gov- ernments because they have relatively few trained officials. Policies should therefore be simple and robust. At the same time, however, the many ties among and within the small leadership elite require the policymaker for the sake of survival to take into account the many spe- cial interests of the educated elite. Can we expect to find special interest pressures so large that most small developing countries will have infe- rior economic growth? Far from it! Harberger's review shows that one small group of small countries has done badly but also that a much larger group has done relatively well. Furthermore, according to his  Introduction 7 review, the small countries with policy weaknesses show symptoms simi- lar to those revealed by larger countries with policy weaknesses. 2. Country Studies Part II of the volume consists of five country studies, which review briefly the types of policies these countries have followed in recent years. Although Argentina had a spectacular growth during the first three decades of this century, its performance deteriorated substantially dur- ing the 1940s and 1950s. Cavallo contends that the reason was mainly the effect of trade distortions and domestic currency overvaluation. Very elucidating too is Cavallo's analysis of the period 1956-84, which shows that economic policies rather than external shocks led, through large fluctuations in the real exchange rate, to several stagflation crises. Simonsen reviews Brazil's economic policies. Brazil could be consid- ered to have been an open economy until 1929, but with the collapse of coffee prices in that year, Brazil started a thirty-five-year period of inward-looking policies. Emphasis on investments and diversification led to high growth rates, but the economy became more and more closed. Imports as a percentage of gross national product (GNP) fell from about 24 percent in 1929 to less than 6 percent in 1964, when a major change in policy emphasized real-exchange-rate adjustments. The results were spectacular through 1973, the year of the first oil-price shock, and remained very good because of external borrowing through 1980, the year of the second oil-price shock. At about that time, it became apparent that Brazil's external debt was becoming a problem. Subsequent adjustments resulted in a severe decline in investment. The crucial issue in Brazil will be to maintain growth by restoring the savings ratio to its old levels. From 1965 to 1986, Indonesia has had a spectacular economic per- formance matched by only a small number of other countries. To a large extent, Gillis and Dapice ascribe this to the very sensible real- exchange-rate policies followed by the Indonesian government. Although Indonesia's exchange rate strategy may be characterized as outward looking, trade strategy has turned inward looking since 1973, when import quotas and bans were imposed on automobiles, motorcy- cles, some textiles, and newsprint. The protection of domestic industry by means of quantitative restrictions accelerated after 1980: by 1984, some 22 percent of imports had some form of restriction. One possible explanation for Indonesia's success, offered by Gillis and Dapice, is that the deft management of exchange rate policy and the economic cushion provided by petroleum earnings enabled Indonesia to withstand the protectionistic excesses.  8 F. Leslie C. H. Helmers Park looks critically at the development process in the Republic of Korea. Like many other authors, Park argues that Korea's spectacular economic progress has been due to its outward-looking strategy, in par- ticular its export-led industrialization. Unlike other authors, however, he argues that this progress took place in a regime that was not laissez- faire but highly centralized and interventionist, a regime in which the government gave high priority to export promotion. In addition, he believes that Korea's highly educated and disciplined work force, together with massive foreign assistance, paved the way for Korea's out- standing growth. As negative results of these policies he sees too much concentration of economic power in a few hands and an excessive sus- ceptibility to external shocks. He believes that the pursuit of economic growth has led to too much borrowing from abroad. In retrospect, Park feels that if Korea had relied more on market mechanisms than on inter- ventionist policies, it would have prevented some misinvestments in heavy industries. Cardoso and Levy review Mexico's economic policies. During the "Mexican miracle" period from 1956 to 1970, gross domestic product (GDP) grew at 6.7 percent a year. The government budget had small def- icits or surpluses, and the average inflation rate was only 3.8 percent a year. Investment increased from 14 percent to 23 percent of GDP. Although absolute poverty decreased, some authors criticized the poli- cies during this period for not having improved relative income distri- bution. The "shared development" policies during 1971-76 emphasized the public sector as the engine of growth and import substitution by means of protection. The public deficit rose from 2 percent of GDP in 1971 to 10 percent in 1976. Public debt increased from $7 billion in 1971 to $21 billion in 1976.1 Substantial deficits in the current account of the balance of payments made an adjustment program necessary, which was, however, abandoned when oil production and exports came on stream. The oil euphoria caused many problems. Between 1977 and 1981, the domestic currency was allowed to become more and more overvalued. Exports increased spectacularly because of oil exports, but imports increased even more so. Current account deficits soared, and the public debt tripled. A substantial capital flight ensued. In 1982, the budget deficit reached a peak of 17 percent of GDP; inflation reached 60 percent. As Cardoso and Levy write, "overvaluation and budgets defi- cits proved to be a deadly combination." In 1982, Mexico could no longer service its debt, and a massive adjustment program was undertaken. Between 1982 and 1986, public investment was reduced by some 60 percent in real terms; similarly, real wages were reduced by more than 30 percent. The adjustment process  Introduction 9 is still going on. The lesson to be learned is that "imbalances allowed to accumulate for too long are extremely painful to correct." 3. Not Policy Miracles This volume covers a number of important issues faced by open econo- mies and discusses how to deal with them. One disadvantage of this approach is, perhaps, that too much attention has been paid to problems and not enough to the very positive aspects of an open economy. For it is widely acknowledged that by opening up its economy, welcoming new technology, and competing with the rest of the world a country will become a genuine partner in worldwide progress. Another issue that has not been treated in this volume, as mentioned above, is the attempts of many developing countries at present to enhance the productivity of their domestic capital. Unviable govern- ment plants are being closed, subsidies are being abolished, and in many cases efforts are being made to have the market mechanism replace gov- ernment interventions. Although this volume does not treat, or deals relatively lightly with, such matters and the ongoing debt crisis, this fact does not in any way diminish their importance. One last point: From this volume, as from many studies of economic policy, the reader may garner the impression that it is easy to determine the right economic policies to follow at any moment and that it is a rather straightforward matter to implement them. Reality is far differ- ent. Indeed, no country in the world has always followed the right eco- nomic policies. It is hoped, however, that this study has shown that it does not take policy miracles to produce good results. If major policies are largely in the right direction, then economic performance is likely to be quite successful. Note 1. "Billion" refers to 1,000 million throughout. The symbol "$" refers to U.S. dollars unless otherwise specified.  2 The Real Exchange Rate ME Leslie C. H. Helmers ONE MAJOR LINE that is central to many (and present in all) of the dis- cussions and reviews in this volume is the importance of the real exchange rate. This chapter provides an overview of how the concept of the reall exchange rate may be used. The nominal-exchange-rate con- cept remains important because we need it for the analysis of debt issues, of the process of short-run market-clearing under flexible rates, and of many other problems. For the analysis of trade and current account bal- ances, however, the nominal concept must be replaced by a real- exchange-rate concept. There are two main reasons for working with a real exchange rate. First, there is a need to work in real terms to put the analysis of trade and current account movements on the same basis as the analysis of real sup- ply, real demand, and real price of a single comnmodity. This is not diffi- cult to understand. If export revenues increase in nominal terms just as much as the costs of producing the exports, for instance, then nothing has changed in real terms, though substantial changes may have occurred in nominal terms. Second, there is a need to introduce some discipline into the analysis of the current account in a world that has many different exchange rate systems. We have the fixed-exchange-rate system, which does not con- template devaluation of the domestic currency, but under which we in fact observe intermittent devaluations in many countries. We also have the clean-floating exchange-rate system, whereby the domestic cur- rency is allowed to find its own level without any government interven- tion. In addition, there is a pure crawling-peg system, which contem- plates frequent changes (say, once or twice a month) in the nominal rate in order to adjust for the inflation differential between the country con- cerned and the outside world. In other words, under this system the tar- get is to fix the real exchange rate. In practice, none of the three systems is pure. Multiple exchange rates are common in the fixed system, government interventions often take I0  The Real Exchange Rate 11 place in the floating system, and divergences from an absolute real- exchange-rate target are frequent under crawling-peg systems. Fur- thermore, the domestic currency may be tied to the U.S. dollar or to a basket of currencies of the main trading partners. In addition, there may be quantitative restrictions on the amounts of foreign exchange that the residents of a country are allowed to use for certain purposes. Real-exchange-rate analysis has the virtue of providing a common framework within which one can analyze current account movements under many different systems. 1. The Real Exchange Rate as the Price of Foreign Exchange The real exchange rate can be defined in several different ways, as indi- cated in the previous chapter. One key distinction is whether the exchange rate is viewed as the price (measured in units of local money) of foreign currency or as the value (measured in foreign currency) of the local monetary unit. Some countries quote their official exchange rates in one way, some in the other. So, too, do authors. In the present volume, Bruno, Cavallo, Simonsen, and Gillis and Dapice base their def- inition of the real exchange rate on the first approach. Thus they mea- sure the number of units of domestic goods (in real terms) per unit (that is, per real dollar's worth) of foreign output. This type of index may also be viewed as the price of a real dollar measured in real domestic cur- rency units. Looked at in this way, the real exchange rate is nothing more than the relative price variable in a simple supply-and-demand analysis in which the quantity of real dollars demanded or supplied is expressed as a function of its real price.' There are many other distinctions one might draw among the differ- ent views of and approaches to the real exchange rate. Most particularly, such distinctions would deal with the various alternative indexes that can be used to deflate the local currency component and the foreign currency component of the real exchange rate. For example, important insights flow from the use of the wage rate as the deflator of the local currency component. This use automatically makes the real exchange rate connote a real-wage-rate index (or its reciprocal), and it makes clear how policies influencing the real exchange rate often carry important political overtones and how they intertwine with wage policies (in those countries where that term is meaningful). Both Dornbusch and Fischer emphasize wages as a relevant deflator in a real-exchange-rate measure, at least for some important purposes. The second way of measuring real exchange rates-the value of the domestic currency, measured in foreign money-is used by Dornbusch, Krugman, Fischer, and Cardoso and Levy in the present volume. This way of expressing the real exchange rate is simply the mirror image of the first, so no conceptual issues are involved.  12 F. Leslie C. H. Helmers In discussing the second approach, however, the above-named authors use a framework centered on the market for nontradable goods. To complement their approach, this chapter presents the frame- work of the foreign exchange market for the analysis of real exchange effects so that the reader will be familiar with both types of analysis. As will become evident, real-exchange-rate analysis in the context of the foreign exchange market is similar to standard microeconomic analysis of commodity markets. 2. Analysis of the Real Exchange Rate Our treatment of real-exchange-rate effects is built on a series of assumptions that have traditionally been made in order to render the analysis more manageable. First, we assume that the world prices of imports and exports in dollars are given, that the country's imports and exports do not change these prices, and that the state of the world econ- omy does not change. In addition, we take the tariff structure of the country as given. To avoid confusion, we will use the following terminol- ogy. For the domestic currency unit we take the rupiah, for the foreign currency unit the dollar. When more rupiah are paid for a dollar, we will say that the exchange rate increases or rises and that the rupiah depre- ciates or devalues. Vice versa, when fewer rupiah are paid for a dollar, the exchange rate falls or declines, and the rupiah appreciates or revalues. Consider now figure 2-1. On the horizontal axis are measured the quantity of demand and supply of real dollars (for example, dollars deflated by the U.S. wholesale price index). On the vertical axis is mea- sured the real exchange rate-that is, real rupiah per real dollar. The formula for the real exchange rate E is: E E lPd where En is the nominal exchange rate, Pd is the domestic price deflator (for example, the domestic consumer price index), andPw is the deflator for the U.S. dollar (for example, as above, the U.S. wholesale price index).2 Appendix C presents further details about this concept. Figure 2-1 presents functions representing the demand for (DD) and the supply of (SS) foreign exchange. These functions should not be taken as simple supply and demand curves-in fact, they reflect a far subtler concept. The best way to think about these curves is to conceive of them as reflecting alternative equilibrium situations in the foreign exchange market. The intersection point of the curves reflects a situa- tion in which there is no net movement of capital (or of the central bank's monetary reserves of foreign currency) in either direction.  The Real Exchange Rate 13 Figure 2-1. Demand for and Supply of Real Dollars: Interaction between Capital Flows and the Real Exchange Rate 5-~~~~~~~~~i -D S Rp7 -_ ___ - Rp5 $100 million of capital inflow of loan ofloan ofloan l\ S I I D l I I I l I 500 525 565 600 Millions of real dollars (nominal dollars IPw) 500O 525 565 600 $25 million $40 million $35 mlloan proceeds proceeds proceeds spent on spent on exportables spent on importables nontradables If Rp7 is the equilibrium real exchange rate for no net movement of capital, a real exchange rate such as Rp5 per dollar would reflect an equilibrium with a capital inflow of, say, $100 million (or with a reserve loss of like amount). Either the capital inflow or the reserve loss will reflect a situation in which total national expenditure exceeds total out- put.  14 F. Leslie C. H. Helmers We can now contemplate a situation in which, starting from an equi- librium real exchange rate of Rp5, the situation is modified so that the new equilibrium real rate is Rp7 to the dollar. In the simplest descrip- tion of such a scenario, foreigners, having up to now been content to lend the country in question, say, $100 million a year, now decide to make new loans only to the extent that old ones are amortized. That is, the country's line of credit from abroad, which in the past had been expanding at the rate of $100 million a year, now stops increasing. The first consequence of this drying up of foreign credits will be that those entities that were previously borrowing from abroad will have less money (real purchasing power) to spend. They will accordingly cut back their purchases, presumably by a like amount. Part of this cutback in spending will reduce the demand for exportables, part will reduce that for importables, and still another part will typically fall on the nontradables sector. The reduction in home demand for exportables (here assumed to equal $25 million) will tend to increase actual exports and add to the available supply of foreign exchange. The reduction in home demand for importables (here taken to be $35 million) will quite clearly reduce the demand for imports and hence for foreign exchange. But these two effects will suffice to eliminate the initial trade deficit only in the case where the full proceeds of the loan were spent on the two classes of tradable goods. Most likely, the borrowers would spend some part of the loan on nontradables. In the present example, such spending amounts initially to $40 million. When part of the loan was initially spent on nontradables and the sup- ply of new loan money dries up, there will also be a reduction (by the domestic currency counterpart of $40 million) in the demand for nontradables. The result will be a pair of disequilibria, fully "compati- ble" with the assumption that total domestic spending equals total income produced plus the net inflow of funds from abroad. On the one hand, there will be, at the old real exchange rate of Rp5 to the dollar, a gap ($40 million in figure 2-1) between the demand for and supply of foreign exchange. On the other hand, corresponding to the excess demand for tradable goods will be an excess supply of nontradables, itself the outcome of the reduction in demand for that category of goods. To equilibriate both markets, a relative price adjustment is required. This is the movement of the real exchange rate from Rp5 to Rp7 to the dollar. The rise in the real exchange rate will reduce the supply of and stimulate the demand for nontradable goods, thus eliminating the excess supply in that market. At the same time it will stimulate an expan- sion in the production of tradables (by making the production of both exportables and importables more profitable), while curtailing their demand through a rise in their relative price.  The Real Exchange Rate 15 The full adjustment to the cessation of the net capital inflow will have taken place when the real exchange rate has moved up to Rp7. Note that in the above scenario no explicit policy is required to cut back expenditures. That problem is directly solved by putting no new loan money into the borrowers' hands. All that is needed in this case, therefore, is an expenditure-switching policy (typically a devaluation of the currency) to produce the required rise in the real exchange rate. Other scenarios are more complicated and may entail the use of both policies. The easiest case to treat is identical to the above, except that the government is assumed to be the entity that initially borrowed. Now, when the loan funds are no longer forthcoming, some explicit decision will typically be required to bring about a cutback in expenditures by the amount of the fall in net foreign borrowing. In this case, a combina- tion of expenditure-reducing and expenditure-switching policies would be required to carry the economy to its new equilibrium at a real exchange rate of Rp7 to the dollar. A third scenario is one in which the central bank jumps into the breach, as it were, and uses new credits to support the same outlays as those previously financed by foreign loans. The easiest variant of this scenario occurs with a fixed exchange rate. In such a case, the same trade deficit that was previously financed by foreign loans is now financed by a loss of international reserves of the banking system. Once the step is taken to replace foreign loans by domestic credit expansion, a combination of expenditure-reducing (in this case stopping credit expansion) along with expenditure-switching policies (a real devalua- tion) is now required in order to reach the new equilibrium. A fourth scenario is like the third but does not entail an initial capital inflow from abroad. That is, from the beginning an excess of domestic spending over domestic production is financed in a direct way by the creation of new credit in the banking system but in a more fundamental way by the loss of foreign exchange reserves. (Note that there is no trade deficit nor any ex post excess of spending over production in this case, unless reserves are drawn down. Otherwise, the credit expansion spree will produce only inflation, which of course would have to be accompa- nied by a devaluation or a flexible exchange rate or new trade restric- tions in order to avoid a reserve drain.) In a sense, the fourth scenario is the most common among developing countries. In a country pursuing a fixed-exchange-rate policy, interna- tional reserves provide a sort of cushion, which allows a degree of inde- pendence in monetary and fiscal policy. If such a country engages in a spurt of deficit financing and if the spurt is reversed quickly enough, the final result may be merely a once-and-for-all loss of reserves. What is critical in this case is that the process should not go on so long that wages and other domestic costs rise to a level that is difficult if not impossible to reduce.  16 F. Leslie C. H. Helmers The scenario in this simple case would be that a spurt of credit expan- sion (either to the private or to the public sector) would cause the demands for exportables, importables, and nontradables all to shift to the right. This shift would cause (a) a deficit in the current account owing to increased imports and reduced exports and (b) a rise in the rel- ative price of nontradable goods, that is, a fall in the real exchange rate. To rectify this situation in a sufficiently timely way, the only policy needed is to eliminate the spurt of credit expansion. This will shift the demand curves for importables, exportables, and nontradables back to their initial positions, and the country will be none the worse for the experience, apart from the loss of reserves that occurred in the interim. In the above simple case no expenditure-switching policy is required. Because the entire "problem" came from expenditure generation, all that is needed to cope with it is expenditure curtailment Readers should not consider this case as some sort of esoteric rarity. Quite to the contrary, it is part of the normal existence of any economy which suc- cessfully pursues a fixed-exchange-rate policy for any extended period. The fifth scenario occurs when the fourth scenario gets out of hand. In this case, the credit expansion lasts so long that it has significant effects on nominal wages and other costs. The supply curves of all three categories of goods therefore shift upward and the nominal supply prices increase for given quantities. In this case the simple reversal of credit expansion will not do the trick because supply shifts have inter- vened. The easiest way to deal with this case is expenditure curtailment, together with devaluation, with the former policy working to curtail the current excess of expenditure over income and the latter one working to effectuate the needed amount of expenditure switching. 3. Expenditure Reducing and Expenditure Switching The objective of the expenditure-reducing policy is to leave GDP unchanged while reducing domestic expenditures on consumption goods C and on investment goods I so that exports X minus imports M can rise. One way of achieving this objective is for the government to reduce its expenditures on consumption goods and investment goods. Another way is to force the private sector to reduce its expenditures. In theory, wages could be reduced in order to cut private consumption, but in practice, political reasons make this almost always unfeasible. Tax increases are also possible but are difficult to implement, particularly in the short run. Another, more feasible policy is to restrict bank loans to the private sector so that private investment is curtailed. In principle,  The Real Exchange Rate 17 then, in some ways expenditures on C and I (and therefore on M, because M forms part of C and I) can be reduced. Unfortunately, domestic expenditure-reducing policies have many negative side effects. Reduced expenditures on C and I, and therewith on M, will lead to unemployment and excess capacity in the C and I industries. The export industry X will not be able to absorb the freed resources immediately because the adjustment process takes time. Fur- thermore, if prices and wages are inflexible, then the export industry may not expand at all, because its profitability does not increase. Thus, although indeed there will be some improvements in the balance of pay- ments because of reduced imports, GDP will very likely decline. In other words, expenditure-reducing policies may well lead to a recession. Let us consider now how expenditure-switching policies work. To do so we must understand how the different types of domestic industry will be affected. In principle, we can distinguish four types of industries. First, there are the import-competing industries. They satisfy part of the domestic demand for importables, whereas the remaining part is provided by actual imports. Thus, when the rupiah price of imports increases, imports will decrease, whereas the import-competing indus- tries will expand production (until the marginal resource cost equals the new higher import price in rupiah). Second, there are the exportables industries, whose production satisfies domestic demand as well as export demand. When the export price in rupiah increases, they will increase production (until the marginal resource cost equals the new higher export price in rupiah). The higher price for the products of these industries reduces domestic demand, so that production will be diverted to exports. In addition, exports will increase because of the production expansion. Third, there are industries producing goods and services only for the domestic market, but these products may become internationally traded if the exchange rate changes. Imports may take place if the rupiah price of imports falls substantially, for instance, and exports may take place if the rupiah price of exports increases substan- tially. Finally, some industries produce only pure home goods and ser- vices, which are not tradable at all, such as transport, construction, electricity, and many banking and insurance services. Both the third and the fourth type of industries may be considered producers of home goods or noninternationally traded goods. Consider now what happens if the real exchange rate rises. Imports as well as export goods sold in the domestic market become more expen- sive. The reduced demand for these goods will be switched to import substitutes and home goods because both are relatively cheap. In addi- tion, exports will increase because they become more profitable. In sum, there is a reduction in the demand by residents for foreign-  18 F. Leslie C. H. Helmers produced goods and an increase in demand by residents, as well as increased foreign sales of domestically produced goods. Another way of expressing the switching effect is to say that the increase in the price of traded goods (imports and exports) switches domestic demand from the traded goods to home goods. Figure 2-2 presents the effects of the expenditure-switching policies. The increase in the real exchange rate leads to movements along the supply and demand curves for real dollars. The result of these move- ments is that the balance of payments deficit, which is equal to D at a real exchange rate of Rp5 to a dollar, is being converted to a surplus S at a real exchange rate of Rp8 to a dollar. In this case, in contrast to the expenditure-reducing policy, we see that the expenditure switch caused by the increase in the real exchange rate does not lead to unemploy- ment. In fact, we will see a resumption of growth because the export industries will expand, as will the import-substitute and home-goods industries. The slopes of the supply and demand curves determine to what extent the real exchange rate needs to be raised to turn a deficit into a surplus. These slopes represent the responsiveness of trade to expendi- ture switching, and this will, of course, differ from country to country. In other words, the structural differences between countries determine the magnitude of required changes in the real exchange rate for balance of payments improvements. Figure 2-2. The Effect of Expenditure-Switching Policies S~~~S 0~~~~~~~~~~~~~~~~ Rp8