- 'upa ,~ ~ - a' i= t$6> W~ b -1!Q 1~~~~~~~~4 1- W 0 R L D B A N K COMPARATIVE MACROECONOMIC STUDIES Thailand's Macroeconomic Miracle Stable Adjustment and Sustained Growth I For Suthida and Divina Thailand's Macroeconomic Miracle Stable Adjustment and Sustained Growth PETER G. WARR B HANUPO NG NI D HIPRAB HA THE WORLD BANK, WASHINGTON, D.C. OXFORD UNIVERSITY PRESS, KUALA LUMPUR C) 1996 The International Bank for Reconstruction and Development / The World Bank 1818 H Street, N.W., Washington, D.C. 20433 Distributed exclusively in Asia and non-exclusively in the rest of the world by Oxford University Press 4Jalan UJl/5. Seksyen U] Hicom Glenmarie Industnial Park 40000 Shah Alan Selangor Darul Ehsan Malaysia All rights reserved Manufactured in the United States of America First printing October 1996 The World Bank Comparative Macroeconomic Studies series emerges from a research project that reviewed the macroeconomic experiences of eighteen developing countries over a period roughly from 1965 to 1990. So that the studies might be published with relatively little delay, the books have been edited outside the World Bank's Office of the Publisher by the Macroeconomic Research Department. The findings, interpretations, and conclusions expressed in this publication are those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to the members of its Board of Executive Directors or the countries they represent. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Dr., Danvers, Massachusetts 01923, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects, authors, and countries and regions. The latest edition is available free of charge from Distribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'[ena 75116 Paris, France. Cover desigu by Saot Ferro Peter G. Warr is the John Crawford Professor of Agricultural Economics at the Australian National University, Canberra, Australia. Bhanupong Nidhiprabha is Associate Professor in the Faculty of Economics at Thammasat University, Bangkok, and Research Fellow at the Thailand Development Research Institute, Bangkok, Thailand. World Bank ISBN 0-8213-2654-6 Oxford University Press ISBN 983-56-0000-7 Library of Congress Cataloging-in-Publication Data Warr, Peter G. Thailand's macroeconomic miracle: stable adjustment and sustained growth / Peter G. Warr, Bhanupong Nidhiprabha. p. cm. Includes bibliographical references. ISBN 0-8213-2654-6 1. Structural adjustment (Economic policy)-Thailand. 2. Thailand-Economic policy. 3. Thailand-Economic conditions. 1. Nidhiprabha, Bhanupong, 1954- . 11. Title. HC445.W363 1996 338.9593-dc2O 95-13672 CIP Contents Foreword vii Acknowledgments ix Thai Names x Acronyms xi Data Notes xii Chapter 1. Introduction I Part One. The Macroeconomic Policy Environment Chapter 2. The Political and Historical Setting: The Paradox of Policy Stability 7 Thailand's Turbulent Political History 8 Sources of Economic Conservatism /9 Conclusions 29 Chapter 3. The Economic Setting: Structure and Performance of the Thai Economy 30 Social Character 30 Market Structure 31 Economic Performance 42 Appendix: Thailand's Macroeconomic Data Base 60 Chapter 4. The Policy Setting: Role of the Public Sector 67 The Institutional Framework 68 The Development Plans 70 The Fiscal System 75 Sectoral and Market Interventions 77 Public Provision of Infrastructure 87 Public Enterprises 89 Regulation of External Public Sector Debt 93 1* vi Thailand's Macroeconomic Miracle: Stable Adjustment anti Sustained Growth Part Two. Macroeconomic Adjustment to External Shocks Chapter 5. The Shocks 99 Balance of Payments Shocks 99 Effects of the Shocks on National Income 106 Effects of the Shocks on Economic Growth 112 Conclusions 116 Chapter 6. Thailand's Macroeconomic Response 117 Analyzing the Adjustment Response 117 Thailand's Aggregate Response 121 Conclusions 135 Chapter 7. The Role of Fiscal Policy 138 Fiscal Adjustment: 1970-90 138 Discretionary or Automatic Fiscal Stabilizers? 143 Sources of Stabilization 154 Conclusions 165 Chapter 8. The Role of Monetary Policy 168 Autonomy of Monetary Policy 169 Control of the Monetary Base 184 Instruments of Monetary Policy 190 The Performance of Monetary Policy 197 Conclusions 203 Chapter 9. The Role of Exchange Rate Policy 205 The Record of Exchange Rate Policy 205 The Policy Environment of Devaluation 210 Effects of Devaluation: Wages, Balance of Payments, Inflation 212 Post-1984 Exchange Rate Setting 215 Exchange Rate Policy and Relative Prices 216 Conclusions 227 Chapter 10. Conclusions: Thailand's Miracle? 228 Pattern of Response to Shocks 228 Were Fiscal and Monetary Policy Stabilizing? 230 Why the Macroeconomic Conservatism? 232 How Well Has Thailand Adjusted? 235 Notes 237 Bibliography 240 Index 251 Foreword This volume is the product of an intensive World Bank study of recent macroeco- nomic policy that reviewed the recent experience of eighteen countries as they at- tempted to maintain economic stability in the face of international price, interest rate, and demand shocks or domestic crises in the form of investment booms and related budgetary problems. The project paid particular attention to the 1974-79 period (which included the first and second oil price shocks), the 1980-82 period of worldwide recession and external debt problems for many developing coun- tries, and the 1983-90 period of adjustment to economic difficulties and the re- sumption of growth. The objective of the project was to glean instructive lessons by analyzing the stabilization and adjustment policies pursued by these countries and assessing the outcomes. The authors of each country study were asked to deal with a common set of questions concerning the nature of the shocks or crises: their origin and de- gree of seriousness; the fiscal, monetary, exchange rate, and trade policies adopted in hopes of preventing permanent harm to the economy; and the results of the pol- icies. No single computable macroeconomic model w as used in the project, but the framework of the open-economy macroeconomic model was followed to ensure consistency in generalizing about policy results. This intensive study of many ep- isodes generated ideas and suggested relationships showing the cause and effect behind the policies, the nature of the shocks and cri\es, and the governmental re- sponse to them. The overall findings of the project are presented in a synthesis volume by 1. M. D. Little, Richard N. Cooper, W. Max Corden, and Sarath Rajapatirana. Boom, Crisis, and Adjustment: The Macroeconomic Experience of Developing Countries. Thailand provides a case study of relatively successful adjustment. Despite significant external economic shocks and some internal political instability, Thai- land was less adversely affected by the turbulence of the 1970s and early 1980s than virtually any other oil-importing developing country. The successful adjust- ment took the form of consistently high rates of economic growth, low inflation, and only moderate growth of external debt. The present study provides details of how this was accomplished. The role of economic policy is a central theme. The results of the study show the importance of cautious macroeconomic pol- icies, combined with reliance on market mechanisms as the principal means of re- source allocation. While the Thai achievement of sustained economic growth vii viii Thailand 's Macroeconomic Miracle: Stable Adjustment and Sustained Growth combined with macroeconomic stability is impressive, the policies which pro- duced this outcome appear to be capable of being emulated by other developing countries, should they choose to do so. Publication of this study was made possible by a generous grant from SIDA, the Swedish International Development Authority. Sarath Rajapatirana Director, "Macroeconomic Policies, Crisis, and Grovvth in the Long Run" Research Project. Economic Adviser, Operations Policy Group Acknowledgments The authors wish to acknowledge the contributions of the many individuals who have assisted in this study. Work on this manuscript began in 1986, when Peter Warr was a visiting professor in the Faculty of Economics at Thammasat Univer- sity, Bangkok. At that time, Bandid Nijathaworn, then of Thammasat University, was a member of our team. In 1987 Bandid Nijathawom left Thammasat to join the International Monetary Fund and subsequently the Bank of Thailand. We gratefully acknowledge his important contribution to the initial stages of our re- search, without implicating him in the many changes of direction and emphasis that occurred thereafter. The World Bank project leaders-Richard Cooper, Max Corden, Ian Little, and Sarath Rajapatirana-were helpful and constructive at all stages of the re- search. Detailed comments and suggestions from Max Corden and Sarath Rajap- atirana were especially helpful. Sarath Rajapatirana was a patient and constructive project coordinator. Many colleagues also provided helpful comments on earlier drafts and offered suggestions for the research. In particular, numerous valuable comments, suggestions, and corrections of factual errors were received from Am- mar Siamwalla, president of the Thailand Development Research Institute. Two anonymous readers also provided many helpful comments and saved us from er- rors of fact and judgment. Other colleagues who assisted us with their comments included Vittorio Corbo, Peter Drysdale. Malcolm Falkus, George Fane, Ross Garnaut, Alan Meltzer, Dennis de Tray, James Reidel, and David Vines. We had the benefit of research assistance from many able individuals. These included Zita Albaeea, Sunee Budsaiyavith, Soonthorn Chaiyindeepum, Ahmed Hafi, Elsa Lapiz, Wimolmas Lulitanonda. Siriwan Pichitwonglert, Maneerat Pin- yopusarerk, Kobkul Pitarachart, Agus Setiabudi. and Cao Yong. Monique Lumb provided much-valued secretarial assistance and Suthida Warr worked heroically to assist in assembling the final manuscript. The assistance of all the above individuals, and any others we have neglected to mention, is greatly appreciated. We were not able to adopt all the suggestions we received, including many with which we were in general agreement. As al- ways, final responsibility for what remains resides with the authors. i.A Thai Names The Western scholarly convention of referring to individuals and citing authors by their family names is awkward in the case of Thailand because the Thai custom is to identify individuals by their first names. These are the names by which Thais are most readily recognized, both within Thailand and abroad. Thai family names tend to be long and unfamiliar. Indeed, the use of family names at all is a relatively recent innovation. In this book we have followed the Thai custom by referring to Thai individuals and authors primarily by their first names. Thus, for example, the 1984 study by Chaiyawat Wibulsawasdi is cited as Chaiyawat (1984) rather than Wibulsawasdi (1984). When referring to Thai individuals in the text we have in- cluded family names as well as first names whenever it seemed that confusion might arise from stating first names only. In the bibliography, both the first names and family names of Thai authors are given in full, and the authors are listed al- phabetically by their first names. The other names are listed in the usual order. x Acronyms ASEAN Association of Southeast Asian Nations BAAC Bank of Agriculture and Agricultural Cooperatives BOI Board of Investment CAO Changwat (provincial) administrative organization CGD Comptroller-General Department DSR debt service ratio ERP effective rate of protection GDP gross domestic product GHB Government Housing Bank GNP gross national product GSB Government Saving Bank IFCT Industrial Finance Corporation of Thailand IMF International Monetary Fund LC letter of credit LIBOR London interbank offer rate LPG liquefied petroleum gas LTF long-term flows NAD National Accounts Division (of the NESDB) NCF net claims on financial institutions NCG net claims on government NDC net domestic credit NDPC National Debt Policy Committee NEDB National Economic Development Board NEPR net export performance ratio NESDB National Economic and Social Development Board NFA net foreign assets NIE newly industrializing economy NOL net other liabilities NSO National Statistical Office OECD Organisation for Economic Co-operation and Development OLS ordinary least squares OPEC Organization of Petroleum Exporting Countries SDR special drawing right SES Socio-Economic Survey xi xii Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth SIFO Small Industries Finance Office SNA standard accounting framework STF short-term flows TDRI Thailand Development Research Institute TFP total factor productivity USAID United States Agency for International Development VAT value added tax Data Notes * Billion is 1,000 million throughout. * Dates indicated with a slash (1967/68) are fiscal years. * Dollars are current U.S. dollars unless otherwise specified. * The symbol - in tables means not available. * n.a. means not applicable. Chapter One Introduction The performance of the Thai economy in recent decades has been nothing short of stunning. In 1950. following an entire century of economic stagnation, Thailand was one of the world's poorest countries. Since then. its economy has experienced rapid growth. declining poverty, and macroeconomic stability. What is particularly impressive about this achievement is that it occurrcd in a volatile economic and political environment. Like most developing countries. Thailand has had its share ol' internal and ex- ternal economic and political shocks. Internally, it has suffered political turmoil throughout much of the postwar period. Authoritari.an military governments alter- nated with democratic or semidemocratic regimiies for briet' periods, al[l in an atmo- sphere of military coups, attempted coups. threatened coups, and general political unrest. Paradoxically, Thailand maintained economic stability despite this politi- cal chaos. Thailand has also been affected by strong e\terial shocks. They have included: a perceived military threat from Vietnam in the I 960s; some side effects ot' the Viet- nam war, among them a flood of American aid anLd military spending during the war, followed by its abrupt cessation; the boom in internationatl primary commodity pric- es of 1972-73A the two massive petroleum price increases of 1973-74 and 1979-80X the high interest rates of the early I 980s. the world rccessioni of the first half of the 1980s; and a foreign investment boom in the late 19h0s. Moreover, Thailand has ex- perienced a very large long-term decline in its terms of trade-ifrom an index ot' I (( in 1970. to about 60 in 1990. Nevertheless, whereas many developing countries, in- cluding some of Thailand's Southeast Asian neighbors. were badly destabilized by these and even lesser shocks, Thailand showed surprisinig resilience. From 1965 to 1990 the real gross national product (GNP) per capita in Thai- land grew at an annual rate of 4.2 percent. compared with an average of 2.5 percent for low- and middle-income countries. En en more remarkable W.as the stability of' Thailand's growth. In a pattern almost unique among oil-importing countries, Thailand's real output per head of population has niot experienccd a single year of 2 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth negative growth since 1958 (figure 1.1). And between 1986 and 1990, the Thai economy was the fastest growing in the world. Although income distribution ap- pears to have become more unequal over the past two decades (figure 1.2), the in- cidence of absolute poverty has declined significantly. The rate of inflation has been low, averaging only 3.8 percent from 1965 to 1990 compared with an average rate of 32 percent for all low- and middle-income developing countries, and its ex- change rate has been stable. All this was achieved with only moderate growth of the external debt and with stable international reserves. Considering the degree to which other economies were destabilized by inter- nal and external shocks seemingly no more severe than those listed above, the Thai experience of stable growth seems almost miraculous. The thesis of this book, however, is quite the opposite. There was no miracle. Although its achievement was impressive, Thailand did not accomplish anything that other countfies. simi- larly endowed with physical and human resources and experiencing similar shocks, could not have matched. The Thai achievement of stable growth could have been equaled or even bettered by others, had they been disposed to adopt the required policies. This study attempts to show how Thailand did it. Such a study is of interest not only because it can shed light on Thai economic history, but also because it may hold important lessons for other developing coun- tries. There are presumably an unlimited number ot ways of adjusting poorly to external shocks, but far fewer possible ways of adjusting well. Although case stud- ies of mismanaged adjustment in this or that country are clearly of value, particu- larly to specialists studying those countries, the choices made usually turn out to Figure 1.1. Per Capita GDP at Constant 1972 Prices, 1951-91 Baht 12,000 t0,000 8,000 6,000 4,000 2,000 >UHHHVUTfHl QIL 1951 53 55 57 59 61 63 65 67 69 71 73 75 77 79 81 83 85 87 89 91 Source: National Economic and Social Development Board, Bangkok. Introduction 3 Figure 1.2. Income Shares by Population Quintile, 1975-76 and 1988 Income share (percent) 60 T 50 40 30 + 20 10 0 1 1 I [l III IV V 0 1975-76 * 1988 Note: Quintile I is the poorest, quintile V the richest. Sources Suganya and Somchai (1988); Thailand, National Statistical Office (1988). have a country-specific explanation. A detailed study of successful adjustment. where it has occurred, may have more useful lessons to offer. That is, the lessons of success may be more transferable than the lessons of failure. Our central contention in this book is that Thailand was able to achieve eco- nomic stability despite its apparent political turmoil because it followed cautious economic policies. These policies were greatly influenced by the attitudes of the country's policymakers, most notably by the conservatism of its bureaucrats. These attitudes helped Thailand avoid major mistakes. If valid, this conclusion has broad implications for the economic history of the world since 1970. As noted earlier, many developing countries were badly desta- bilized by the volatile events of the external environment surrounding them. Why not Thailand? If the answer is that Thailand followed good policies, then it pre- sumably follows that others could in principle have responded similarly but that, for some combination of reasons-no doubt different in every case-they did not do so. But their destabilization was not an inevitable consequence of the shocks they experienced: it was at least in part a consequence of their own policy respons- es to those shocks. It could have been otherwise. 4 Thailand 's Macroeconomic Miracle. Stable Adjustment and Sustained Growth Part One of the book describes the economic and political setting in which Thailand's macroeconomic achievement took place. Part Two examines the shocks themselves and Thailand's adjustment to them. Our study concentrates on the period from 1970 to 1990. An earlier starting point might have been chosen except for data considerations. The data series in the revised national accounts recently published by the Thai government begins with 1970. This revised data series, released in 1987, differs so much from previously available data that it is difficult to compare Thai economic data immediately be- fore and after 1970. Data for the years before 1970 must be considered somewhat unreliable, and if it is possible to avoid using them it is advisable to do so. We end with 1990 because it was the latest year for which full statistical coverage was available at the time our study came to a close. Part One The Macroeconomic Policy Environment Chapter Two The Political and Historical Setting: The Paradox of Policy Stability Throughout most of its modem history, Thailand has tolerated heavy military in- volvement in its political affairs. Thailand's political history since World War ll has been punctuated by a succession of military coups and attempted coups, some- times followed by relatively democratic periods, sometimes not. In the two de- cades since 1970 the role of the military in civilian matters has become increasingly contentious because the country's growing urban, educated middle class has begun demanding democratic reform and a reduced role for the military in the country's economic as well as political affairs.I It is hardly surprising that to outsiders Thai political life often seems unstable. The institutions governing the succession of political power in Thailand have in- deed experienced many crises, but these events should not be taken to mean that Thai political life is no more than an unruly succession of coups. The struggle for the top jobs is one matter, but policy formation is quite another, and there the un- derlying reality has been far more stable than the appearance. This stability stems in large part from fact that the Thais are united by a popular monarchy and by the religion of the majority, Buddhism. The present monarch is His Majesty King Bhumipol Adulyadej, who has reigned since 1946. Ethnic Chinese constitute a large minority group heavily concentrated in the capital, Bangkok. They are treat- ed as Thais, however, and the nation has clearly benefited from their entrepreneur- ial abilities. Unlike some of its neighboring countries, Thailand has experienced little racial conflict. Despite many military coups or attempted coups, Thai policymakers have continued to follow similar basic political and economic philosophies. They have differed on distributional matters, but they have all been committed to economic growth and shared the belief that market forces combined with prudent public sec- tor infrastructure investment should be the principal means for achieving that end. Thailand's bureaucracies have played a particularly important role in maintaining 7 8 Thailand s Macroeconomic Miracle: Stable Adjeustment and Sustained Growth the continuity of economic policy along these lines. But where did these conser- vative attitudes come from? This chapter presents the political and historical context in which Thailand's economic adjustment has occurred over the past two decades. The chapter opens with a sketch of Thai political deve]opments since the founding of Bangkok two centuries ago. It then addresses the question of why the conservative attitudes of Thailand's economic policymakers developed as they did. Thailand's lurbulent Political History For the purposes of this discussion, the historical events of interest begin with the establishment of Bangkok as Thailand's capital in 1782, following the destruction of the former capital of Ayutthaya by Burmese forces in 1767. The Founding of Bangkok The birth of the new capital coincided with the founding of the Chakri dynasty, which has reigned since that time. The present king is the ninth of the dynasty (thus known as Rama IX). These historical events help explain why Bangkok con- tinues to play a prominent role in Thai public life and in the Thai imagination, a role that extends well beyond its being merely the national capital. Partly because of this history, Thailand's economic activity is heavily concentrated in this one city. Its position in this respect is unmatched in any other country of Thailand's size. Although military considerations were the primary reason for the relocation of the capital from Ayutthaya to Bangkok, the move had important economic con- sequences. Whereas Ayutthaya is situated 80 kilometers upstream from the mouth of the Chao Phya River, Bangkok is right at the mouth and adjacent to the Gulf of Thailand. Being on the river, Ayutthaya was able to maintain significant maritime trade connections, but it would never have been able to engage in the commerce and international communication that the new capital fostered from its more fa- vorable location. Avoidance of Colonization Thailand avoided colonization by European powers in the nineteenth century by successfully playing off the British in Burma and India to the west and in Malaya to the south against the French in Indochina to the east and in Laos to the north. To preserve the independence of the central plains, however, particularly that of Bangkok, the government made major territorial concessions to both powers. The British took over what is now Penang in Malaysia, and the French gained much of The Political and Historical Setting: The Paradox of Policv Stability 9 present Cambodia and Laos. Under the Bowring Treaty signed by King Mongkut (Rama IV) and Great Britain in 1855. Thailand agreed to embark on some free trade in its economic policies. But unlike similar treaties the Western powers signed at about the same time with China and Japan, the treaty with Thailand was apparently entered into without much overt coercion from the British (Ingram 1971). In Thai historical accounts, King Chulalongkorn, the fifth king of the Chakri dynasty (Rama V), who reigned from 1868 to 191(0, is given credit for avoiding colonization, for establishing the Thai civil service, and for abolishing slavery. Al- though recent historical accounts claim that Thailand was able to avoid coloniza- tion more because of a convenient compromise between the British and French (Wyatt 1984), King Chulalongkorn had clearly recognized that internal financial stability was vital to keeping the European colonizers at bay. Any sign of Thai fi- nancial instability, it was feared, might have been used as a pretext for the Euro- pean gunboats to attack Bangkok. The king engaged a British financial adviser, from the Bank of England, to consult on economic matters. Predictably, his advice was conservative. The British financial adviser continued to be an influential force in Thailand's economic affairs for many years, and the position was not abolished until the late 1950s. More important, the Bank of England became the conserva- tive model for the establishment and operation of the Bank of Thailand following World War 11. Free trade policies promoted the expansion of agricultural exports but did not produce rapid economic growth. In the century following the Bowring Treaty. Thai economic growth barely kept pace with population growth (Sompop 1989), and there was virtually no structural change. Agricultural exports were the main source of both foreign exchange and government revenue. The agricultural growth was not driven by improved productivity but by the expansion of the cultivated land area (Ammar, Suthad, and Direk 1993). Land remained abundant until the 1960s. Constittttional Monarchv, 1932 The period of absolute monarchy ended in 1932 with a coup d'etat against King Prajadhipok (Rama VII). The king agreed to support a constitutional monarchy. Although the coup seems to have lacked public support, it was successfully por- trayed by the new military leaders as a "revolution."A succession of military-dom- inated governments followed. In 1939 a government edict changed the country's English-language name from Siam to Thailand. "Siam" is apparently a non-Thai word (possibly Khmer or Chinese) that came into use only after its formal appearance in the original nine- teenth-century treaties with Britain and France, "Thailand" is a literal translation of the Thai-language name for the country. The change of name symbolized the abrogation of the old treaties and declared that the Thai state warranted a legal sta- tus on a par with that of the imperial powers of the world. 10 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth The Thai government allied itself with Japan during World War II, once inva- sion by Japanese forces became imminent. Thailand subsequently shared in the Japanese hyperinflation toward the end of World War 11, and this inflation contin- ued for several years after the war. This was the country's sole experience of sus- tained rapid inflation. The inflation came on the heels of the founding in 1942 of the central bank, the Bank of Thailand, which has rigorously pursued anti-infla- tionary monetary policies ever since-not unlike those of the Bank of England, on which it was modeled. Today the Bank of Thailand plays a central role in Thai- land's macroeconomic policymaking environment. Establishment of the Bank of Thailand The idea of creating a central bank in Thailand originated during the reign of King Chulalongkorn (Rama V).2 Following his return from a tour of Europe, the king began a program to modernize the country. In 1904, Prince Mahisra Raj Haruetai, the minister of finance, set up the Book Club Association, the first banking office owned by a Thai national. The first bank to open an office in Bangkok had been the Hong Kong and Shanghai Banking Corporation, which began operation there in 1888. In 1914 the Ministry of Finance employed Sir Bernard Hunter, a British banker, to prepare a plan for upgrading the Book Club Association into a central bank. Although the study was undertaken, the plan was not implemented. Follow- ing the 1932 coup, the Book Club Association was transformed into a commercial bank named the Siam Commercial Bank. In 1939 the plan to create a central bank was reactivated, and the government en- acted legislation establishing the National Banking Bureau. Functioning somewhat like a central bank, the bureau was to receive deposits from government institutions and other banks, oversee exchange control, and manage government borrowing ac- tivities. It opened in June 1940 with Praya Sonauraraj as its first director. During the Japanese occupation of World War I1, Japan proposed three major monetary changes: the value of the Thai baht was to be fixed in relation to the Jap- anese yen at the exchange rate of 1: 1; transactions between Japan and Thailand were to be conducted in Japanese yen; and a central monetary authority was to be established in Thailand with advisers and key personnel from Japan. Thailand adopted the first two proposals, although the new monetary alignment implied a 36 percent depreciation of the baht against the yen. Before the war, the Thai baht had been fixed to the gold standard at the exchange rate of I baht per 0.66567 grams of pure gold, 11 baht per pound sterling, and 100 baht per 155.7 Japanese yen. Thailand was obliged to accept the second proposal under the terms of the al- liance, which prohibited trade with countries other than Japan and its occupied ter- ritories. Thailand rejected the third proposal, however, as it would have given Japan complete control of the Thai currency. To prevent Japan from exercising di- rect control over the country's financial affairs, the Thai government hastily enact- ed the Bank of Thailand Act in April 1942, establishing the bank as an independent institution. Eight months later, in December, the Parliament passed a bill outlining The Political and Historical Setting: The Paradox of Policy Stability Jl the bank's functions and method of operation. The Bank of Thailand was opened on December 10, 1942, with Prince Wiwat as its first governor.3 Constitutionally, the Bank of Thailand is governed by a Board, which origi- nally comprised six members: the minister of finance, as chairman of the board; the governor and deputy governor; and another three members appointed by the king (this number of appointed members was expanded to five in 1946 and later to the current eleven members). World War If Inflation During World War ]1, government expenditure increased, stimulating aggregate demand and exacerbating inflation. But this domestic source of inflation was small compared with the inflationary forces from outside-notably the effects of the Japanese occupation. The high inflation during the war was caused mainly by an expansion of the money supply, beginning with the Japanese occupation in 1941 and the 36 percent depreciation of the baht imposed by Japan. Since Thailand was forced to trade only with Japan and its occupied territories, the trade balance sub- sequently moved in Thailand's favor as Thai exports became somewhat cheaper in yen prices. Meanwhile, the value of imports did not rise as much because Japan could not supply Thailand with goods needed for the war effort. Thus, Thailand built up foreign reserves in yen, and the Bank of Thailand had to sell baht to ex- porters, thereby increasing the domestic money supply. The monetary expansion was also assisted by an increase in direct Japanese military expenditure in Thailand following the establishment of the new 1:1 ex- change rate. Between 1942 and 1945 Japanese military expenditure increased from about 2 million baht per month to 100 million baht per month (Bank of Thai- land 1962). Because fiscal policy could not be used to restrain government expenditure during wartime, the remaining instrument available to control inflation was mon- etary policy, specifically, credit control, a salary deposit scheme, the use of checks to pay for certain transactions, and forced saving. CREDIT CONTROL. When the money supply rose sharply in June 1943, the govern- ment imposed credit controls. First, it raised the currency reserve ratio, with the result that the commercial banks had to raise their deposits with the Bank of Thai- land. The precise ratio was to be announced from time to time by the Ministry of Finance, but whereas the ratio had previously varied from 3 to 5 percent, it was now to be varied within the range of 9 to 45 percent. Second, the government required financial institutions to hold government bonds at varying ratios: commercial banks were to hold at least 40 percent of their total deposits received, and insurance companies were to hold 45 percent of their total investible funds, or 75 percent of their total premiums received; the holdings for other credit institutions were to be determined from time to time by the Min- istry of Finance. 12 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth After barely three months, the decree was amended to limit the currency re- serve ratio to a maximum of 25 percent; and to specify that the amount of govern- ment bonds to be held by commercial banks would be determined by the Ministry of Finance from time to time, except that the amount of bonds plus currency re- serves deposited at the Bank of Thailand were not to exceed 50 percent of total deposits received. In addition, the government bonds held by insurance companies were to be 30 to 40 percent of total investible funds, or 50 to 70 percent of the total premiums received; again, the precise figure was to be announced by the Ministry of Finance from time to time. These credit control measures were in effect until November 19, 1946. THE SALARY DEPOSIT SCHEME. In 1943 the government required high-ranking of- ficials to have saving accounts with commercial banks, and their salaries were to be paid into those accounts. This was an attempt to reduce the amount of cash in circulation. THE USE OF CHECKS. In April 1944 the government decreed that starting on May 1, 1944, any transactions of goods and services of more than 1,000 baht were to be paid for by check. Otherwise, there was to be a surcharge of 10 percent on the tax otherwise levied on that transaction. The purpose of this measure was to reduce the amount of money in public circulation, encourage the use of checks rather than bank notes, which became scarce during the war, and prevent the use of money to hoard goods for speculative purposes. The decree was in effect for only a few months before Parliament reconvened and rejected the bill. It is therefore difficult to assess the effectiveness of this policy in reducing the amount of cash in circu- lation. FORCED SAVING. In February 1945 the government issued a decree titled Saving Bonds in Emergency, which demonetized the 1,00() baht bank note that had been issued during the war. Holders of these notes were to return them to the Bank of Thailand in exchange for saving bonds of equivalent value. The bond was redeem- able after one year at an interest rate of I percent. This forced saving proved to be the most effective instrument for reducing the money supply. Following the de- cree, 371.5 million baht were withdrawn from circulation, which amounted to about one-third of the total currency then in public hands. INTEREST RATE POLICY. The government also raised interest rates to restrict credit. On February 23, 1945, the government raised the prime rate from 4.5 to 8 percent per year. This measure proved ineffective, however, and the money supply contin- ued to rise. The Political and Historical Setting: The Poirad(kr of P(olicy Stability 13 Postwar Military Governments, 1946-63 Following World War 11, Thailand was ruled by a succession of conservative mil- itary-controlled governments. Phibul Songkhram was prime minister from 1948 to 1957, when he was forced into exile as the result of a coup sparked by charges of election rigging. The military strongman Field Marshall Sarit Thanarat then in- stalled two prime ministers, first diplomat Phote Sarasin and then General Thanom Kittikachorn, but in 1958 he assumed the premiership himself. Sarit's government began with the arrest of scores of suspected communists and the dissolution of Parliament. An event of considerable significance for subsequent Thai economic history was the arrival of a World Bank advisory mission in 1957 (Suehiro 1989). The re- port of this mission (World Bank 1959) influenced Thai economic policy in two important ways. First, it recommended a fundamental shift in the nature of public sector involvement in the economy: away from direct production, via the extensive and highly inefficient public enterprise sector that existed at that time, toward in- vestment in the public infrastructure required for economic development, particu- larly roads, ports, electricity supply, and telecommunications. Second, the report recommended that the government change its method of promoting private sector investment. The World Bank report found a receptive audience in the government of the time, particularly in Sarit himself, and significant changes in economic policy fol- lowed. There was a new surge of public investment in infrastructure and the newly formed National Economic and Social Development Board (NESDB) began formu- lating regular five-year development plans to guide these investments. Although most of the public enterprises then in existence remained in place, they became less significant as private sector economic activity in manufacturing grew. Also, the Board of Investment was created in 1958 to render assistance to private investors. The Thanom Military Government. 1963-7? With Sarit's death in December 1963, power shifted to a new military alliance headed by Generals Thanom Kittikachorn and Prapas Charusathien. Thanom be- came the prime minister and Prapas assumed the powerful posts of the commander in chief of the army and the minister of interior. The new government was less rig- id than its predecessor but continued to ban political parties and political gather- ings. The administration was preoccupied with the security problems associated with the growing communist threat. A new constitution was promulgated in June 1968 and an election was held in February 1969. Thanom continued as premier. The new government's main po- litical weakness was its inability to control the large number of independent mem- bers of Parliament. In November 1971 a plan by the opposition to defeat the government in a vote of no confidence was uncovered and prompted Thanom to stage a coup against his own government. The constitution and the Political Parties 14 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Act were abrogated, and the country was brought under the management of the National Executive Council headed by Thanom. In December an interim constitu- tion was announced giving sweeping power to the government. Thanom continued as prime minister. In the following months, the government became increasingly unpopular, partly because of its unwillingness to return to democratic rule. The military was also growing wary of Thanom and Prapas, who had built up lucrative networks in business and dominated the succession in the army. In October 1973 several stu- dents and former members of Parliament were arrested while distributing leaflets calling for democracy. They were charged with inciting public unrest. The arrests sparked a student protest at Thammasat University in Bangkok, which led to a bloody confrontation between students and soldiers. Many students were killed and the king personally intervened, asking Thanom and Prapas to leave the coun- try. Under additional pressure from other members of the military, the two went into exile. Civilian Government, 1973-76 The king appointed Sanya Dharmasakdi, rector of Thammasat University, as the new prime minister. The Sanya government prepared the way for a period of de- mocracy and civilian government. With Thanom and Prapas no longer on the scene, the military had lost a good deal of its political strength, leaving the country with the opportunity to experiment with civilian-based politics. A new constitution was promulgated in October 1974. Unlike previous constitutions, this one incor- porated many of the basic elements of a British-style parliamentary democracy. The Political Parties Act was liberalized, although the Communist party remained illegal. There were forty-three political parties registered for the general election held in January 1975. The election returned to the House of Representatives 269 members belong- ing to twenty-two political parties. The Democrats headed the list with seventy- two elected members and managed to form a minority government headed by Seni Pramoj, who became Thailand's first elected prime minister. The new government was defeated in a vote of no confidence in March 1975, and the premiership went to Kukrit Pramoj, leader of the Social Action party, who had engineered the Dem- ocrats' defeat. Some interesting policy initiatives were introduced under Kukrit's govern- ment. Thailand reestablished diplomatic relations with China, closed American military bases in the country, and introduced several welfare programs and mea- sures that were to be emulated by subsequent governments. These measures in- cluded the rice price support program, the Tambon (district) development funds, and the policy for free medical care for the poor. Kukrit still found the army hos- tile, however, a problem that later democratic governments were to share. The mil- itary became increasingly impatient with the public demand for social change and the growing influence of left-leaning ideologies among students and labor. Kukrit The Political and Historical Setting: Tie Paradox ofPolicy Stabilitv 15 also had to contend with in-fighting among members of government vying for cab- inet portfolios. Under pressure from the military, Kukrit reshuffled cabinet mem- bers and finally dissolved Parliament in January 1976, when a motion to oust him in Parliament through a vote of no confidence was about to be launched. The election of April 1976 was unusually violent. A number of farmers, stu- dents, and labor leaders who supported liberal political parties were killed. The Democrats again headed the list with 1 14 elected members. Kukrit's Social Action party won 56 seats but he himself failed to be reelected. The Democrats formed a four-party coalition government under the premiership of Seni Pramoj. A split developed in the new coalition government when former premier Thanom Kittikachorn returned from exile as a Buddhist monk. Students mobilized in protest against the return of the former dictator, which many of them saw as a provocation to disorder. The situation became uncontrollable when civilian mobs and police stormed Thammasat University to silence the student protesters. On the morning of October 6. 1976. many students were killed and many more arrested. The Seni government was unable to keep order. That same evening, the military staged a coup, dissolved Parliament, and abandoned the constitution. This coup ended three years of civilian government marked by perpetual political turmoil and student unrest and put the premiership and the control of government back in the hands of the military. The brief years of civilian government had coincided with significant eco- nomic changes. New interest groups had emerged and new policies initiated and implemented. An important new development was the involvement of the business class in politics. A number of bankers and well-known business people obtained ministerial portfolios and were able to influence the direction of economic poli- cies. This was a radical departure from the previous periods, when control over policy was shared only by the military and the bureaucracy. Military Rule, 1976-79 The new military strongman was Admiral Sangad Chaloryu, who launched the coup under the banner of the National Administrative Reform Council. The new government, backed by the military, was headed by an ultraconservative lawyer, Thanin Kraivichien. Thanin's strong anticommunist stand and nationalistic atti- tudes led him to adopt extreme policies, which alienated his government from the bureaucracy, the press, foreign allies, and factions in the army, particularly the ar- my's young field commanders. The administration's emphasis had shifted back to- ward national security, law, and order and ignored the country's basic social problems. The military became increasingly embarrassed by Thanin's conserva- tive vision. In October 1977 the military staged a coup to dismiss the man they had put into power. Although Thanin made no major changes in economic policy during his one year of tenure, he did take the important step of empowering the Ministry of Fi- nance to borrow directly from 'abroad. Foreign borrowing by the Thai public sector 16 Thailand X Macroeconomic Miracle: Stable Adjustment and Sustained Growth dates back to the first railway loans in the period of King Rama V, but Thanin re- laxed the administrative procedures involved. It is not clear whether the decision was made to please the military or to facilitate expansion of public investment, but it led to a surge in both public expenditure and the country's foreign debt. The 1977 coup was again led by Admiral Sangad, who headed the National Revolutionary Council, which took power. Nevertheless, the premiership was of- fered to an army general, Kriangsak Chamanand, whose soft diplomacy during the Thanin period had made him popular among many sections of the Thai communi- ty. The premiership, which was granted under an interim constitution, was to be limited to one year, during which Kriangsak was to prepare for national elections and civilian rule. The military had been divided over the choice ot the premiership succeeding Thanin. Kriangsak's leadership was known to have the backing of the young field commanders in the army and was favored by Washington and Peking because it was thought he could ease tensions in the region. His immediate task was to con- solidate power within the military and to ease the tension that had built up during the Thanin period. To defuse the situation, Kriangsak personally sponsored an am- nesty bill pardoning all those arrested in the October 6 incident of the previous year and lifting censorship of the press. He concurrently made himself defense minister and reshuffled the military to consolidate his power base. In addition, he appointed General Prem Tinsulanonda commander in chief of the army. A new constitution was promulgated in November 1978 establishing two chambers in Parliament: an elected House of Representatives and an appointed Senate. The constitution gave limited power to the elected house and preserved military influence in the Parliament through the appointed Senate. The leader of the Senate, who was also leader of the Parliament. was to nominate candidates for the premiership for a vote in a joint sitting. This tacitly brought the prime min- istership under the influence of the military and the bureaucracy. which dominat- ed the Senate. The constitution was thus seen by many as a perpetuation of military rule. Quasi-Civilian Government, 1979-91 In April 1979 a general election was held under the new constitution. No single party obtained a majority, and the Senate selected Kriangsak to continue in the premiership, leading a coalition government made up largely of pro-Kriangsak in- dependents. The composition of the Kriangsak cabinet was a focal point for attack by the opposition because it reduced the role of the government's elected mem- bers. The regime ran into further problems in February 1980, when it announced a string of petroleum product price increases, which met strong public protest. The parliamentary opposition united over the issue and submitted a parliamentary mo- tion calling for a vote of no confidence. The motion was openly supported by a number of members of Parliament. Unable to control the situation, Kriangsak re- signed one day before the scheduled motion was to be debated. The Political and Historical Setting The Paradox of Policv Stability /7 The prime ministership was offered to General Prem Tinsulanonda, a respect- ed army general. who was favored by the military and the main political parties. Prem became prime minister in March 1980. He was not an elected member of Parliament but was able to take the premiership under a constitutional provision that was to produce havoc 12 years later. The distinctive feature of the Prem government was that power was shared by the political parties, the military, and the bureaucracy. Through the appointed Sen- ate, the military and the bureaucracy controlled the premiership and screened im- portant policies. The prime minister ensured that the bureaucracy maintained its influence in government by controlling three important cabinet portfolios: de- fense, finance, and interior. The other portfolios were shared among the coalition parties. Although the Prem administration experienced several political crises- which led to cabinet reshuffles, attempted coups, and unscheduled elections-the basic arrangement of power sharing described above remained undisturbed for eight years. Threats of a vote of no confidence and ot embarrassing revelations re- garding Prem himself led to his resignation in 1988. Following the election of July 1988 Chatichai Choonhavan became prime minister, the first elected member of Parliament to attain this post since Seni Pramoj in 1976. The Chatichai government presided over the period of the most rapid economic growth in Thai history, but the government itself was widely re- garded as corrupt, even by Thailand's generous standards. Civilians were appoint- ed to key economic posts that the military leadership regarded as their own, and the atmosphere became increasingly tense. The 199/ Militanr Coup In February 1991 the military leadership staged a sudden coup against Chatichai, installing a National Peace Keeping Council to run the country and abolishing the 1978 constitution. The coup leaders, Generals Suchinda Kraprayoon and Sunthorn Kongsompong, cited the corruption of the Chatichai government as the reason for the coup. A more likely reason, according to close observers, was that the Chat- ichai government had excluded the top military leaders from the benefits of gov- ernment corruption. The coup leaders installed a widely respected businessman, Anand Panyara- chun, as prime minister. Elections were held early in 1992, and parties supporting the military won control of the Parliament. General Suchinda became prime min- ister even though he was not himself an elected member of Parliament and had publicly promised not to claim the office. Pro-democracy demonstrators were in- furiated. They were led by the former governor of Bangkok, Major General Cham- long Srimuang. who staged a public hunger strike in protest at Suchinda's actions. The composition of the protesters reflected the rapid socioeconomic changes that had begun to take place in Thailand. Unlike the youthful student demonstrators of 1973 and 1976. the demonstrators in 1992 included a high proportion of affluent, middle-aged, middle-class professionals-many reportedly carrying mobile tele- 18 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth phones and driving to the demonstrations after work in Volvos and BMWs-but a significant number were indeed veterans of the student demonstrations of the 1970s. The chaos led in mid-May to a massacre of demonstrators by soldiers appar- ently acting under the orders of Suchinda himself. Suchinda was forced from of- fice in disgrace. Further conflict was avoided when the king intervened and reappointed Anand Panyarachun prime minister, in preparation for fresh elections. These elections were held in September 1992 and produced a democratically elected civilian government. Political Instability? The preceding summary of Thailand's turbulent political history would hardly lead one to expect stability of either economic policy or performance to result. But, as later chapters demonstrate, that is exactly what happened. How can it be explained? The facts are indeed puzzling, but four points can be mentioned in par- tial explanation of this apparent paradox. First, Thailand's political history has not been as volatile as it may seem. Al- though power has often changed hands through violence or the threat of violence, the various contending forces have had a great deal in common. In general, polit- ical leaders have had to be acceptable to the military, or at least to some powerful factions within it. Most potential political leaders either are or have recently been senior military men. The degree of military involvement in political affairs in Thailand has surprised many foreign observers, as has the public's tolerance of this involvement as a normal state of affairs. Strong military connections are in- dispensable for aspiring political leaders. This means that politicians espousing views that conflict strongly with the conservative outlook shared by most Thai military men can expect trouble. Second, partly because of the military factor, Thailand's political parties are not separated by wide ideological differences. Despite the vigorous competition between them, the major parties all agree that it is vital to preserve Thai traditions and institutions, and especially to remain loyal to the present monarch, King Bhu- mipol. In these respects, the fact that Thailand was not colonized by European powers is clearly very important. Thai traditions and institutions were not eroded or discredited by the militarily superior Europeans in the way that they were in most former colonies. The Thai intelligentsia have little difficulty in identifying with their country's past. Third, over the past thirty years, most political leaders in Thailand have per- ceived a potential military threat from communist Vietnam. This has had a unify- ing effect and encouraged caution in domestic and foreign policy. Thailand also experienced a domestic communist insurgency of its own, which it successfully suppressed. Fourth, Thailand's conservative bureaucracy exhibits a surprising degree of independence from political control. Thailand has adopted essentially the British The Political and Historical Setting: The Paradox of Policy Stabilirt 19 system of a permanent and professional civil service. As a result, the Thai bureau- cracy perceives its long-term loyalty as being owed primarily to the king, repre- senting the nation, rather than to the current government. This system is in contrast to that in Thailand's near neighbor, the Philippines, for example, which is modeled on the civil service system of the United States. There political appointees domi- nate the upper ranks of the bureaucracy, only to be replaced en masse when the administration changes. Sources of Economic Conservatism The sustained conservatism of Thailand's economic policy can be traced to a num- ber of factors. Five will be discussed: the desire to maintain a balance of trade, the impact of macroeconomic instability on the livelihood of policymakers, the policy attitudes of senior bureaucrats, the example of China in the 1940s, and the influ- ence of the World Bank's recommendations in 1959. Maintaining a Trade Balance Thai monetary policy prior to 1970 has its roots in the nineteenth century, when all of Thailand's neighboring countries were colonized by Western powers (see In- gram 1971: 170-74). Under the influence of British financial advisers, the coun- try's leaders in Bangkok believed that a sound financial policy would leave the least possible excuse for foreign intervention on the grounds of financial irrespon- sibility. Thus they favored conservative monetary policies that gave high priority to achieving a balance of trade and to maintaining the value of the baht and its ex- change rate with the pound sterling. With the fixed exchange rate system (at II baht per pound sterling), the Thai money supply was dependent on foreign re- serves and hence on the country's trade position. The legal mechanism used to control the money supply was the Currency Act of 1928, drafted by Sir Edward Cook, then financial adviser to Siam. Under the terms of the act, baht notes could not be issued unless the equivalent value in gold and pounds sterling was placed into the currency reserve. Thus, the value of the baht was backed 100 percent by the equivalent value of strong foreign liquid as- sets. Indeed, from 1928 to 1932 it was backed by more than 100 percent. The composition of the currency reserves was later amended to include ster- ling securities, U.S. dollars, and dollar securities; however, the amount of gold, foreign exchange, and foreign securities with a maturity of less than one year was at no time to equal less than 60 percent of the total amount of baht issued. In effect, this put a ceiling on the printing of money such that it could not be more than 1.67 times the total foreign liquid assets held in the currency reserves. This principle was strictly adhered to, and in fact the ratio of currency reserves to baht in circu- lation never fell below 70 percent until World War 11. 20 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth During the Japanese occupation of World War 11. gold and foreign securities deposited in England and the United States were frozen by the Allies, and the val- ue of the baht was not fully backed. In 1941 the government passed the Currency Emergency Act to allow Japanese yen and government bonds to be included in the currency reserves. Thus the rule governing the maintenance of foreign liquid as- sets came to be upheld by the Japanese yen instead. This provision allowed the money supply to increase rapidly in response to Japan's growing military expen- diture in Thailand during the war. After the war, the large part of the currency reserves that were held in Japa- nese yen were virtually worthless. Between 1946 and 1947 the ratio of total for- eign liquid assets to the total amount of baht issued fell below 60 percent for the first time, and the government had to replenish the foreign assets in the currency reserves. It did so by restricting imports and introducing a multiple exchange rate regime (Corden 1967). At the same time, it curtailed the amount of baht in circu- lation to match the amount of currency reserves. By 1958 the reserve position was again stable, and the government passed a new currency act, drafted by Prince Wiwat, then one of the government's financial advisers. It restored the monetary principle contained in the 1941 Currency Emer- gency Act, whereby foreign liquid assets must be maintained at a ratio of not less than 60 percent of the total amount of baht issued. This principle has been adhered to since that time (see figure 2.1). Figure 2.1. Ratio of Foreign Liquid Assets to Total Baht Issued, 1928-89 Percent 120 100 80 60 v 40 202p 1928 1938 1948 1958 1968 1978 1988 Source: Bank of1Thailand. The Political and Historical Setting: Thie Pairadox of Policv Stabilin' 21 The Impact of Macroeconomic Instability o(t the Livelihood of Policymakers During the period of absolute monarchy, which lasted until 1932, the majority of senior civil servants in Thailand were members of the royal family or the nobility. Until 1875. they earned their living from collecting rents or from the use of land granted to them by the crown. The area of land granted was based on their posi- tion: this system was called sak-dina. In return, they had to pay taxes on part of their earnings. Under Thailand's subsistence economy, this system provided its civil servants with an adequate and flexible income. The tax burden was propor- tional to their earnings. Exactly how much civil servants could earn per month is not known accurately because no records were kept, but it can be said that they were well-off. According to an old Thai saying, "it is better to have as patron a sin- gle civil servant than ten merchants." The situation changed with the government reform initiated during the reign of King Chulalongkorn (Rama V). In 1875 tax collection and government budget- ing came under central control. Thereafter, civil servants were paid fixed salaries and the sakdina was abolished. The salary adjustments were intermittent, howev- er, occurring only when grievances arose. and they were not formally indexed to the cost of living. Indeed, systematic statistical measurement of the cost of living was not introduced until after World War II. During the rule of King Vajiravud (RamaVI), from 1910 to 1925, the govern- ment's financial position weakened, owing to the expenditures required for the king's modernization program, as well as his own extravagance. In 1911 the king ordered that his second coronation proceed in grand style, with foreign dignitaries in attendance, and the affair cost about 4.5 million baht. or nearly 8 percent of the national budget of just over 60 million baht (Terwiel 1983: 295). In February 1912 this lavish spending provoked an unsuccessful plot to overthrow him. Siam experienced its first financial crisis in the period 1919-22, when the in- ternational price of silver rose sharply. As a result, the value of the country's silver coin climbed beyond its face value, and this encouraged the melting down of coins for export. W. J. F. Williamson, the British financial adviser to the kingdom, rec- ommended that the purity of silver coins be reduced. but this suggestion was re- jected (Ingram 1971: 155). Instead, the government opted in November 1919 to revalue the baht from 13 to 9.54 baht per pound sterling. Another important event at this time was the fivefold increase in the price of rice, owing to the unprecedent- ed overseas demand for Thai rice after World War 1. Between 1914 and 1918 the average price of number 1 White Garden rice had ranged from 5.5 to 7.8 baht per picul (60 kilograms) and then in mid- 1918 it soared to 34 baht per picul (Ingram 1971: 156). As a result, the real incomes of civil serxvants fell drastically in relation to those in commerce, By the time King Prajadhipok (Rama VII) ascended the throne in 1925, the government was in serious financial trouble. The country was then hit by the Great Depression, and government revenue plummeted from 107 million baht in 1929- 30 to 79 million in 1931-32 (Statistical Year Book of Siam 1933-35: 26). To pre- 22 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth vent a financial crisis, the government began cutting its expenditure and raising additional taxes: In February 1932 the financial situation became such that the govern- ment, in panic, once more raised the tariff on all goods except those still restricted by treaty. As a result, the prices of many articles rose quickly-by amounts of 5 percent to as much as 20 percent in a month. Port dues, land and house taxes, and the excise duty on match- es and cement were all increased. At the same time, hundreds of civil and military officials were retired, and 5 percent was deducted from the salaries of those remaining. (Thompson 1941: 549) In April 1932 a tax was imposed on all salaries above 600 Ticals (baht) per year. This fell hardest on the small-saliried intelligentsia and brought home to every one the seriousness of the government's finan- cial plight. (Thompson 1941: 554). Once again the real incomes of the civil servants were seriously affected, al- though this time the cause was not inflation but a reduction in their nominal incomes. These events culminated in the revolution of June 24, 1932, which brought the ab- solute monarchy to an end. Although the decline in the real incomes of the civil ser- vants was not the direct cause of the revolution, their discontent made the upheaval more acceptable after it came. They clearly blamed the hardships they had experi- enced on the financial mismanagement of the government. During and shortly after World War II, Thailand experienced its only period of sustained high inflation, the causes of which were outlined above. Accurate records from 1933 onward on the salaries of civil servants, the price of rice, and the cost-of-living index attest to the drastic effects the inflation had on the welfare of civil servants. Table 2.1 shows salaries for the lowest-ranking and the highest- ranking civil servants, in columns (I) and (2), respectively. For present purposes, the highest-ranking are of most interest. Rice prices are available from 1933 on- ward. A cost-of-living index is also available from 1948 onward. These two price series are shown in columns (3) and (4), respectively. They are the same until 1948, because the price of rice is used for both series, but diverge after that date. The last four columns show the real salaries of each of the two categories of civil servants, deflated by each of the two price series. These calculations are summa- rized in figures 2.2 to 2.4. The story told by these records is very clear. The rapid inflation of World War II and its aftermath were a disaster for civil servants, especially for those at the top. Indeed, to this day the salaries of civil servants, in relation to the price of rice, have not returned to their prewar levels. Moreover, they have never recovered their former position in relation to the wages of lower-ranked civil servants. Not only did civil servants experience genuine economic hardship as a result of the rapid inflation of the 1940s, but corruption also became entrenched. The public standing of civil servants was thereby permanently damaged by these events. Table 2.1 A Comparison of Nominal and Real Monthly Salaries of Civil Servants, 1933-90 (bohlt per miontht) Real monthlv salaries Real montih/ salaries NoiVinal monthlvxsalaries Rice price Cost of living jor lowest rank/ tf7r highest rank Date of Lowest Highest inidex indlex enforcemetn (1J (2) (3) (4) (I)-( (.?)*100 (I l(4)*100 (2)-(3)*100 (2)((4)*100 1933 20 - OO) 100.0 1().0 20.00 20.00 1,000.(00 1,000.00 12/28/1936 20 1.000 95.4 95,4 20.96 20.96 1,048.22 1,048.22 10/07/1939 20 1,000 106.4 106.4 18.80 18.80 939.85 939.85 10/24/1942 20 1O0() 218.3 218.3 9.16 9.16 458.09 45,809 01/01/1948 30 1,200 2,275.2 2,275.2 1.32 1.32 52.74 52.74 05/18/1951 30 1,400 2,422.0 2,502.7 1.24 i.20 57.80 55.94 05/16/1954 30 1.400 2,807.3 3,071.5 1.07 0.98 49.87 45.58 11/01/1959 450 8.)00 3,009.2 3,640.3 14.95 12.36 265.85 219.76 10/01/1967 540 8.600 5,064.2 4.438.9 10.66 12.17 169.82 193.74 06/01/1973 600 8.6()( 5.504.6 5.528,7 10.9( I0.85 156.23 155.55 01/(0/1974 750 9,855 7,486.2 6,816.5 10.02 11.0( 131.64 144.58 1()/09/1975 750 1(190(1 7,871.6 7,094.1 9.53 (0.57 13847 153.65 10/01/1978 9()0 12.69() 9,529.9 8.7X(.() 9.44 10.25 133.16 144.53 (1/(1/1980 1.080 15.225 11,339.4 11,608.1 9.52 9.30 134.27 131.16 01/01/1982 1,255 17,745 14,110.1 13.871.9 8.89 9.05 125.76 127.92 01/01/1989 2.100 25.000 17,284.4 17,170.9 12.15 12.23 144.64 145.60 (04/01/1990 2.350 30,600 18.183.5 18.308.5 12.92 12.84 168.28 167.14 Note: The cost of lMi in index before 1948 was calculated on the basis oi the rice price index. Source: Thailand. Ministrv of Finance (1990): and National Statistical Officc. Statistical Yetar-hook of Thailad, Narious issues. 24 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 2.2. Nominal Salaries of the Highest and Lowest Rank of Civil Servants, 1909-90 a. Highest Baht per month 35,000 - 30,600 30,000 25,000 20,000 15,225 15,000 10,000 8,000 5,000 I1,000 0 ml r-n r-i~ ~ 7- ir- 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 Baht per month b. Lowest 2,500 - 2,350 2,000 1,500 1,255 1,000 500 450 Apr Dec Oct Oct Jan Mar Mar Nov Oct Jun Jan Sep Sep Jan Jan Jan Apr 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 90 Source: Table 2.1. The Political and Historical Setting: The Paradox of Policv Stability 25 Figure 2.3. Real Salaries of the Highest and Lowest Rank of Civil Servants Adjusted by the Rice Price Index, 1909-90 a. Highest Baht per month 1,200.0 1,048.2 1,000.0 800.0 600.0 458.1 400.0 265.9 200.0 K 578125.8 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 b. Lowest Baht per month 25.0 - 21.0 20.0 15.0 ~~~~~~~~~15.0 Apr Dec Oct Oct Jan Mar Mar Nov Oct Jun Jan Sep Sep Jan Jan Jan Apr 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 90 Source: lable 2.1. 26 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 2.4. Real Salaries of the Highest and Lowest Rank of Civil Servants Adjusted by the Cost-of-Living Index, 1909-90 a. Iiighest Baht per month 1,200 T 1,048.2 1,000 800 600 458.1 400 200 -127.9 0 I F -dSR I F- - 02 - - I- 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 b. Lowest Baht per month 25 r 21.0 20 15~~~~~~~~~~1. 12.4 1 ' CUL"r '2.2 10.3 1 0 5 - 0 LI ~~~~~~~1.2 Apr Dec Oct Oct Jan Mar Mar Nov Oct Jun Jan Sep Sep Jan Jan Jan Apr 1909 36 39 42 48 51 54 59 67 73 74 75 78 80 82 89 90 The Political and Historical Setting: The Paradox of Policy Stability 27 The Policy Attitudes of Senior Bureaucrats The preference for conservative economic policies among senior bureaucrats is particularly evident in Thailand's monetary affairs. Among those who espoused conservatism in monetary matters were two of the most influential governors of the Bank of Thailand: Prince Wiwat, who, as already mentioned, was the first gov- ernor of the bank and a long-time financial adviser to the government; and Dr. Puey Ungphakorn, the longest-serving governor of the central bank, who held of- fice from 1959 to 1971. In an August 1944 report to the finance minister on postwar monetary recon- struction, Prince Wiwat stated that it was urgent to restore the stability of the cur- rency before World War II ended.4 Inflation had to be slowed down, he argued, in order to restore the internal stability of the financial system. An informed student of history, he pointed to the hyperinflation in Germany after World War I as an ex- ample of the catastrophic consequences of failing to make such an adjustment im- mediately after a war. This warning, not made public in Thailand at the time, reflected Prince Wiwat's study for degrees in history in England and France be- tween 1913 and 1920. Dr. Puey Ungphakorn's conservative views are best revealed by his rule of thumb for managing the monetary policy of the Bank of Thailand. He described his rule in a lecture at Thammasat University in 1969: the rate of monetary growth, he stated, should be limited to 2 to 3 percent above the GNP growth rate.5 In his view, economic stability was more desirable than rapid growth. This conservative monetary approach appears to have derived from Dr. Puey's egalitarian outlook and his belief that high inflation would lead to an inequitable distribution of wealth. The Example of China in the 1940s Until the Communist victory in China in 1949, connections between the Chinese business and professional elite in Thailand and China itself were very close. Many of the Chinese elite in Thailand were sent to China for their education. China fell to the Communists, however, following the ravages of World War 11 and the inabil- ity of the Kuomintang government to control the ensuing inflation. The Commu- nists promised to stop the hyperinflation and subsequently did so, albeit brutally. What effect, if any, did these events have on postwar policymaking in Thailand? Members of the ruling elite in Thailand were acutely aware of events in Chi- na, in part because their own country was also experiencing rapid inflation. Thus it would seem logical to suppose that they, along with Thailand's Chinese busi- ness owners, feared a similar Communist insurrection within Thailand. But the lack of clear evidence of such a connection suggests that any such link was prob- ably minor. The domestic factors described above were apparently much more important. 28 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth The Influence of the World Bank's 1959 Recommendations At the request of the Thai government, the World Bank sent a mission to Thailand in 1957-58 to help the authorities formulate a public development program. Many of the mission's recommendations (see World Bank 1959) were subsequently adopted. As discussed earlier, one suggestion was that the government should shift the emphasis in its economic activities from direct involvement in industrial pro- duction through public enterprises to the provision of public infrastructure. Addi- tional recommendations relating to sound fiscal management and effective control of the banking system are especially important for the present discussion. First, the World Bank mission recommended a strict procedure of budgetary appropriation and disbursement. Under earlier practices, politicians often failed to adhere to budgetary appropriations for particular purposes. Expenditures in excess of appropriations were usually debited against a suspense account until they re- ceived legal authorization in the future budget. The Budget Bureau was estab- lished in 1959 independently of the Ministry of Finance to control the budgeting of all government departments. The procedure for annual budget appropriations was governed by the Annual Appropriation Act, and extra expenditure was not al- lowed without the approval of Parliament. Budgetary accounting and fiscal report- ing functions were consolidated and were required to provide the government with up-to-date information on its fiscal position. Second, in the area of banking, the mission recommended that authority over the supervision and control of the commercial banks be transferred from the Min- istry of Finance to the Bank of Thailand and that the control be tightened. Previ- ously, the commercial banks had been rife with shady practices and with dubious political connections between the banks' executives and politicians. This led to the Commercial Banking Act of 1962, which empowered the Bank of Thailand to monitor and supervise the commercial banks. The new act also aimed at prevent- ing the commercial banks from becoming involved in political dealings that could lead to irresponsible banking practices. For example, it banned commercial banks from lending money to their directors. It also required commercial banks to main- tain cash reserve ratios between 5 and 50 percent of their total deposits, the actual rate to be announced by the Bank of Thailand. The ratio was eventually set at 6 percent of total deposits. Third, the mission recommended that the currency reserve be used for devel- opment purposes and that the Currency Act of 1928 be revised to this end. That act had been passed to ensure sound management of the currency reserve and thus re- quired at least 60 percent of the total baht issued and circulated to be backed up by an equivalent value of foreign liquid assets in terms of gold, foreign currency, or foreign securities with maturity of less than one year. The mission argued that although this principle had proved successful in earlier decades, it was was now too conservative. It recommended that the government be authorized to use the currency reserve for development of the country. In 1958 the government passed a new currency act relaxing the restriction on the maturity of foreign securities to The Political and Historical Setting The Paradox of PolicY Stability 29 a period of not less than three years. But the 60 percent rule governing money is- sued and circulated remained in place. It must be emphasized that the World Bank's recommendations would not have been greeted with such approval had they not suited the purposes of the Sarit government. Nevertheless, the package of recommendations clearly had an effect, which was still apparent more than three decades later. Conclusions This chapter has summarized Thai political history since the founding of Bangkok with a view to explaining the origins of the conservative outlook underlying Thai- land's economic policies. Thai views on economic policy have their roots in the traditional attitudes of the country's ruling elites. The Thai aristocracy of the nine- teenth and early twentieth centuries held a virtual monopoly on government af- fairs. Avoiding domination by the European colonial powers and maintaining the existing social order domestically were given overriding priority. Financial stabil- ity at home was seen as a necessary means to those ends. The government's do- mestic economic role was severely constrained by these attitudes, which continued well into the twentieth century. At least until the late 1950s this conservatism had largely achieved its objec- tives, but it had also retarded Thailand's economic growth. As explained in chapter 3, output per capita over the entire preceding century had remained at a virtual standstill. The neglect of public infrastructure was without doubt a leading cause of the country's poor economic performance. The change of attitude in the late 1950s was to have dramatic economic consequences. During the Japanese occupation of World War II and shortly thereafter, Thai- land had its sole experience of hyperinflation. The effects were devastating for many groups, above all for the country's civil servants. This experience, which oc- curred almost at the same time as the founding of Thailand's central bank, had long-lasting effects on the attitudes of the country's economic policymakers, es- pecially in regard to monetary policy. It is not surprising that their overriding pol- icy concern in subsequent years was to avoid inflation. Chapter Three The Economic Setting: Structure and Performance of the Thai Economy This chapter provides a profile of the contemporary Thai economy and its recent economic performance. It focuses on three main topics: Thailand's social charac- ter, market structure, and economic performance. Social Character The discussion opens with the characteristics of the Thai population and some sig- nificant social indicators. Population As of mid-1990, the population of Thailand was 56 million. Between 1980 and 1990 the population grew at an average annual rate of 1.8 percent, a reduction from 2.7 percent over the decade before. Population density was 107 persons per square kilometer of total area and 275 persons per square kilometer of cultivable land. In 1990 the urban population was 23 percent of the total, compared with 13 percent in 1965, and 69 percent of the population worked in agriculture, compared with 6 percent in industry and 25 percent in services. The corresponding data for 1965 were 82 percent in agriculture, 5 percent in industry, and 13 percent in ser- vices. The degree of urbanization in Thailand is unusually low among countries in its income group, and the importance of agriculture in total employment is high. Even more unusual is the degree to which the urban population is concentrated in a single city, Bangkok. In 1990, 60 percent of the population were of working age (fifteen to sixty- four years), compared with 50 percent in 1965. Life expectancy at birth was sixty- 30 The Ec onomic Setting: Structure and Performntn e of the Thai Economy 31 three years for males and sixty-seven for females, which represented an increase of about nine years each from 1965. The crude birth rate and crude death rates in 1990 were 22 and 7 per thousand, respectively, compared with 43 and 12 per thou- sand. respectively, in 1965. The majority of the Thai population is Buddhist, but Islam is an important force in the southern provinces. There is a large population of Chinese origin, which, as mentioned in chapter 2, is concentrated in Bangkok. Ethnic and religious conflicts have occurred but are of minor importance by international standards. Social Indicators Table 3.1 provides social data on Thailand in comparison with other selected Asian countries. Thailand's adult literacy rate in 1990 was 93 percent. In 1988-89 the percentage of persons in the relevant age groups who were enrolled in formal education were 86 percent in primary school. 28 percent in secondary school, and 16 percent in higher education. The low participation rate at the secondary school level is especially significant. In the Philippines. for example, the comparable sta- tistic reported for 1988-89 was more than twice as high, at 73 percent. In 1965 the participation rates in Thailand were 78 percent in primary school, 14 percent in secondary school, and 2 percent in higher education. Thus it is clear that educa- tional investment has been concentrated at the higher level, at the expense of sec- ondary school education. Thailand's infant mortality rate-not reported in table 3.1-was 27 per thou- sand live births in 1990, compared with 90 per thousand in 1965. The rate in the Philippines was 41 and 73 per thousand, respectively. In 1988 the percentage share of household income by quintile, beginning with the lowest level, was 4.0, 8.1, 12.5, 20.5, and 54.9. The percentage share of the highest 10 percent was 34.8. Comparable data for the Philippines (relating to 1985) were 5.2, 8.9, 13.2, 20.2, and 52.5, with the highest 10 percent having a 37.0 percent share. These data sug- gest a marginally more unequal distribution in Thailand than in the Philippines. The Gini coefficients reported in table 3.1 also suggest this conclusion (the higher the Gini coefficient, the more unequal the distribution). Thai data for 1981 suggest a somewhat less unequal distribution than the above Philippine data or the 1988 Thai data. Income inequality in Thailand appears to be widening. The data in table 3.1 also show that poverty in Thailand, as in most Asian countries, is particularly concentrated in rural areas. Market Structure To understand the market structure of the Thai economy, one must look at the mac- roeconomic environment as well as the structure of the country's product, labor, and financial markets. Table 3.1 Basic Social Indicators for Thailand and Some other East Asian Countries, 1965-90 (percent unless otherwise indicated) Average Average annual Life annual GNP growth of real Average expectancy growth of per capita GNP per annual rate of at birth population (U.S. dollars) capita inflation 1990 Employmenta (1986-89) Country 1980-90 1990 1965-90 1965-90 (years) Agriculture Industry Services China 1.4 370 5.8 2.1 70 74 14 13 India 2.1 350 1.9 7.7 59 63 11 27 Indonesia 1.8 570 4.5 24.7 62 54 8 38 Korea, Republic of 1.1 5400 7.1 13.1 71 18 27 56 Malaysia 2.6 2,320 4.0 3.6 70 42 19 39 Philippines 2.4 730 1.3 12.8 64 42 10 49 Thailand 1.8 1,420 4.4 5.1 69 70 6 24 Poverty: Percentage Percentage population Expenditu re Secondari urban Income distributiond belowi, povertv lined on school Urban populationz in Lowest 20 educationb enrollmient' population' largest citv percentshare Gini Country (1989) (1988-89) (1990) (1980) of total coefficient Total Rural China 2.4 44 33 6 - - - India 3.2 43 27 6 8.1 0.42 48 51 Indonesia 0.9 47 31 23 8.8 0.31 39 44 Korea, Republic of 3.6 87 72 41 - 0.36 16 11 Malaysia 5.6 87 72 41 4.6 0.48 27 38 Philippines 2.9 73 43 30 5.5 0.45 58 64 Thailand 3.2 28 23 69 4.0 0.47 30 34 a. Percentage ot total. b. Percentage of GNP. c. Percentage of age group. d. Data relate to 1983 (India), 1987 (Indonesia and Malaysia) 1985 (Phiiippines). and 1988 (Thailand). Source: World Bank. World Developmnern Report 1992. 34 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Macroeconomic Environment Since the end of World War 11 Thailand has followed a fixed exchange rate policy, with the baht pegged to the U.S. dollar. The trading system is quite open, but the capital account is less so. Movements of foreign exchange and Thai currency out of Thailand are regulated, although some of these regulations have recently been relaxed. Large conversions of Thai currency into foreign exchange for the purpose of capital export require Bank of Thailand approval. Until the early 1990s interest rates on both borrowing and lending transac- tions were regulated by ceilings set by the Bank of Thailand. Interest rates have remained at these ceiling levels throughout most ot Thailand's recent economic history. Wages are officially subject to minimum wage controls, but these controls are effective only within the public sector and among large business enterprises. Else- where-in agriculture, small industrial enterprises, and in much of the service sec- tor-the minimum wages cannot be enforced. In any case, the minimum wages seem to follow market forces, remaining somewhat above the wages seen in the small private manufacturing enterprises. As explained in chapters I and 2, Thailand has followed rather conservative macroeconomic policies for most of the past century and for all of the postwar pe- riod. Its central policy objectives have been to maintain a stable exchange rate backed by secure international reserves and, above all, to avoid domestic inflation. The postwar policy environment has also been influenced by the relative indepen- dence of the central bank, the Bank of Thailand. The Thai macroeconomic experience since 1970 cannot be understood with- out considering the implications of its fixed exchange rate, partly controlled capi- tal account, and domestic interest rate controls. These variables determine the extent to which the domestic money supply can be controlled by the Bank of Thai- land. Monetary policies appear to have been countercyclical and stabilizing in the past (see chapter 8), which suggests that capital flows are not fully free. But capital market reforms of the early 1990s are expected to enhance capital mobility and hence reduce the Bank of Thailand's ability to exercise this stabilizing role. Product Markets Like most low- and middle-income developing countries, Thailand has favored product market policies that implicitly tax agriculture and subsidize industry. In the past, the government levied export taxes on several agricultural export com- modities, but these have been slowly phased out. Rubber is now the only commod- ity subject to an export tax. Tariffs and quantitative import restrictions are used to protect part of the manufacturing sector. The production of these commodities is highly competitive, except in the one or two cases described below. The Economic Setting: Structure and Performance of the Thai Economy 35 AGRICULTURAL TAXATION. Rice is by far the most important agricultural commod- ity and a leading earner of export revenue. Until recently, rice exports were taxed by a combination of instruments: the rice premium, which was a specific export tax; an export duty, which was an ad valorem export tax; and a reserve require- ment. which was equivalent to an ad valorem export tax. The effect was to keep both consumer prices and prices received by farmers well below international prices. The combined effect of these policies was equivalent to a 31 percent export tax in 1970. using the f.o.b. export price as a base, and a 67 percent export tax in 1973-74. years of very high international prices, which then declined to 13 per- cent by 1984 (Chirmsak 1984). The rice premium was suspended in early 1986 in response to the low international prices for rice and has not been reinstated. In addition to these policies, the government assigns export quotas to individ- ual export agents. The effect is to introduce a noncompetitive element into the rice export market. Each year the government announces target prices for paddy, but these are generally understood to be cosmetic. Some farmers are able to sell at support prices above current market prices with funds derived from the Farmer Aid Fund, which are the funds originating earlier from the proceeds of the rice pre- mium. The majority of farmers derive no benefit from this provision. The notable exception to this general story of taxing smallholder agriculture for the benefit of urban consumers is sugar. Domestic sugar prices are held above inter- national prices. In the mid-1980s, domestic prices were at least three times as high as international prices. The distribution of holding sizes in the sugar industry is not typ- ical of Thai agriculture, in that there are a small number of large farms and a much greater number of small holdings. This distribution is more typical of Latin America than of Asia. The small farmers are dependent on the larger ones for their markets be- cause the large farmers tend to hold the supplier contracts with the large sugar mills. Together, these large farmers and the mill owners fonr a powerful political lobby. MANUFACTURING PROTECTION. Although parts of the manufacturing sector are high- ly protected and inefficient, in general the sector is competitive and less highly regu- lated than its counterparts in some of Thailand's Southeast Asian neighbors. Thailand does not practice free trade, but its protection levels are moderate and quite stable. There have been several empirical studies of protection policies in Thailand, most of which have concentrated on estimating effective rates of protection (ERPS)-that is, the proportion by which the overall structure of protection raises the value added received by an industry per unit of output in relation to what it would be under free trade. Unfortunately, these studies have used different types of data, different product definitions, and different methodologies. Some have used official tariff rates, whereas others have used tariff rates estimated from customs duty collections or from price comparisons. It is therefore difficult to compare the results of these studies over time. Nevertheless, the studies do show a similar pat- tern over the past three decades (their results are summarized in tables 3.2 and 3.3): namely, that the protective system has been biased against the agro-based industries 36 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 3.2 Effective Protection of Industry Groups by Levels of Fabrication and End Uses, 1964-84 (percent) Industry group 1964 1969 1974 1984 Processed food 47.47 -32.6 -19.41 7.93 Beverages and tobacco 215.45 241.3 2280.55 26.50 Construction materials - 47.4 46.91 17.38 Intermediate goods, 1 82.02 2.8 15.91 17.63 Intermediate goods, II 60.09 79.1 48.53 241.84 Consumer nondurable goods 70.95 32.5 90.63 23.84 Consumer durable goods 63.87 69.1 200.62 19.29 Machinery 37.48 30.6 30.02 32.40 Transport equipment 118.00 34.9 353.88 45.70 Source: Estimated rates of protection (ERPs) for 1964 and 1969 were obtained from Juanjai, Supote. and Sorrayuth (1986). The ERPs for 1974 and 1984 were obtainied from Narongchai (1977) and Pai- toon, Rachain, and Nattapong (1989), respectively. The weight, used to aggregate products for 1974 and 1984 were value added at market prices of 1975 and 1985. rcspectively, obtained from Thai input- output tables. and toward the manufacture of both import-competing and non-import-competing goods. This is the typical pattern of protection found in developing countries. Trairong (1970) studied Thailand's ERPs in twenty-three manufacturing in- dustries in 1964 using input coefficients from Belgium and the Netherlands, which he derived from Balassa's standardized input-output table. From official tariffs and the 1962 Investment Promotion Act, Trairong concluded that the system of protection was biased toward consumption goods, followed by intermediate goods, and biased against capital goods (for an English-language summary, see Juanjai, Supote, and Sorrayuth 1986). Narongchai (1973, 1977) used input-output coefficients obtained from indus- trial surveys to estimate the ERPs of fifty-eight industries in 1969 and eighty indus- tries in 1971 and 1974. When tariffs were used as the main instrument of trade policy and industries were classified by trade orientation, his results showed that import-competing and non-import-competing industries received the highest pro- tection over the study period. Pairote's (1975) study was also based on input-output coefficients from indus- trial surveys, which he used to estimate the ERPS of fifty-eight industries in 1964 and eighty-two industries in 1971 and 1974. The results confirmed that the protec- tion system over the period of study was inward-looking, favoring firms selling on domestic markets. Again, import-competing and non-import-competing industries received the greatest protection. The structure of protection was biased against ex- The Economic Setting: Structure and Perfrirmance of the Thai Economy 37 port industries. When industries were classified by end uses and levels of fabrica- tion, the incentive effects were strongly in favor of consumer goods, especially beverages and tobacco and transport equipment. followed by capital goods (Juan- jai and others 1986). And in a study by Paitoon, Rachain. and Nattapong (I 989), ERPs in Thailand's manufacturing sector for 1981. 1984, and 1987 were estimated from input coeffi- cients from the 180-sector input-output tables of 1982 and 1985. The protective instruments covered in the study were mainly tariffs, import surcharges, export taxes, tax rebates and refunds, and royalties. When industries were classified by trade orientation, the results showed that the effective protection was biased against export industries. The non-import-competing industries received the high- est protection, followed by the import-competing industries. As table 3.3 shows. over the period 1969-87 the export industries received the lowest effective protec- tion, followed by the import-competing industries, while the non-importing indus- tries received the highest effective protection. Labor Markets Thailand's labor markets can be divided into four principal sectors: the civil service, the public enterprises, large private firms, and small private firms. Average wages are Table 3.3 Effective Rates of Protection of the Manufacturing Sector Classified by Trade-Oriented Group, 1969-87 (perc ent) Sector 1969 1974 1984 1987 Export group -43 -35 2 4 Import-competing group Excluding tires and tubes 54 63 21 39 Including tires and tubes (648) Non-import-competing group Excluding cigarettes and sodas 187 77 53 55 Including cigarettes and sodas (812) Note: Products are classified on the basis of Narongchai's studies (1973, 1977), into three groups: export oriented, import competing and nonimport competing. depending upon their trade orientation. A product is classified as export if its export level is greater than 10 percent of its domestic production and if its net export is positive. It is import competing if its import is greater than 10 percent of its total consumption and if its net import is positive. The rest are classified as non-import competing. The 1975 value added at market price of each industry are used as weights to estimate the aggregate ERP of each product group of all years. Source: The figures for 1969 are calculated from Narongchai (1973), those for 1974 are calculated from Narongchai ()1977) and those for 1984 and 1989 are calculated from Paitoon, Rachain, and Nat- tapong t 1989). 38 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth higher in the first two sectors, but so are the educational requirements for the jobs they offer. The data on wages in Thailand are very unreliable. Under the Labor Relations Act of 1975, trade unions are not permitted in the civil service but are legal elsewhere in the economy. In practice, trade unions are strong only in the public enterprises, which have a long history of labor organization. Many of these firms enjoy monopo- lies in their industries and have only a single large plant, located in Bangkok. These features make it relatively easy for unions to organize the work force. Large private firms have a history of opposing the formation and operation of unions. In this, the firms have generally received government support. A variety of tactics, including physical intimidation, have been used against workers attempting to organize (Mabry 1984; Hewison 1989). The unions cannot expect much help from the government. Small private firms, which make up almost all of agriculture, most of manu- facturing, and almost all of the service sector, obviously employ the bulk of the work force. The employees are not organized because the cost of organizing work- ers in scattered small firms would be very high. Labor markets in this dominant part of economy are in general competitive, and wages appear to be flexible, in re- sponse to variations in labor supply and demand (Bertrand and Squire 1980). The evidence presented by Bertrand and Squire (1980) does not support the "dual economy" hypothesis in regard to Thailand's labor markets. Minimum wage legislation exists but is effective only in the public sector and in some, but not all, of the large private firms. As noted above, the legislated minimum wages tend to be marginally above the wages paid by small private firms. The discrepancy is greatest among the youngest, least skilled employees. Labor in Thailand is quite mobile, but for the most part recent migrations have moved in the direction of Bangkok, where the new jobs have been heavily concen- trated. Nevertheless, seasonal migration among agricultural regions is also important (Chalongphob 1993). Although official statistics on unemployment rates are unreli- able, open unemployment seems to be a rare phenomenon in Thailand, except among the most highly educated. Chalongphob (1993) provides a valuable discussion of the reasons for high rates of open unemployment among university graduates. Financial Markets In recent decades the Thai financial system has deepened considerably, with the ratio of financial assets to gross domestic product (GDP) increasing from 40 per- cent in 1960 to 120 percent in 1990 (Robinson, Yangho, and Ranjit 1991). The number of financial instruments available to both users and providers of funds re- mains limited, however. As in many other developing countries, Thailand's finan- cial markets include both substantial organized and unorganized sectors. The organized sector can be broadly defined to include all legally registered institu- tions. The unorganized sector encompasses financial transactions that do not go through organized financial institutions; the most prevalent forms of such trans- actions are borrowing and lending among individuals and in the rotating credit societies. The Ec onomic Setting: Structure atnd Performance of the Thai FEconomy 39 Thailand's organized financial markets are made up of eight main financial in- stitutions: commercial banks; finance, securities, and credit companies; special- ized banks; development finance corporations; the stock exchange; insurance companies; saving cooperatives; and a variety of mortgage institutions. The com- mercial banks make up the largest component in terms of total assets, credit ex- tended, and savings mobilized. In 1990 they accounted for 71 percent of total financial assets in the country. The second largest is the finance companies, which began operating in 1969. Thailand has three specialized banks-the Government Saving Bank (GSB), the Bank of Agriculture and Agricultural Cooperatives (BAAC). and the Government Housing Bank (GHB); and two development linance corporations-the Industrial Finance Corporation of Thailand (IFCT) and the Small Industries Finance Office (SIFO). These specialized institutions are either owned or partly owned by the government. COMMERCIAL BANKS. The financial market is dominated by the activities of com- mercial banks, which absorb roughly three-fourths of all deposits placed with fi- nancial institutions. They are therefore the central actors in Thailand's financial system (Naris 1993). The current structure consists of sixteen local (Thai-owned) banks and fourteen foreign banks. The role of foreign banks is very limited. In 1990 they accounted for only 5 percent of bank assets. Foreign banks operate at a competitive disadvantage in relation to local banks. They must pay a withholding tax on dividends transferred abroad; they cannot be quoted on the stock exchange, since they are subsidiaries of overseas parent companies and thus cannot apply for the 30 percent concessional corporate tax rate; and. most important, they are pro- hibited from opening new branches. The 1962 Commercial Banking Act restricted entry to the banking business. Licensing permits are required, and only one permit has been granted since 1965. Of the sixteen local commercial banks, one is a state enterprise (Krung Thai Bank), one has the Crown Property Bureau as a major stockholder (Siam Commercial Bank). another is partly owned by the government (Sayam Bank), and one is owned by military organizations (Thai Military Bank). Over the past two decades, Thai commercial banks have enjoyed remarkable growth. From 1972 to 1986, the total assets of the banking system increased more than twelvefold, registering an average annual growth rate of 19.5 percent. This growth was fueled largely by an expansion in deposits, which has made the struc- ture of bank assets deposit based rather than equity based. Some experts have pointed to this aspect of the current structure as a weakness. The most popular in- strument in the Thai banking system is the time deposit, which commands an av- erage share of 73 percent of banks' total deposits. Bank lending, by contrast, is dominated by overdrafts, which account for an average of 50 percent of bank cred- it extended. The rest is made up of loans and bill discounting, which have average shares of 20 and 30 percent, respectively. A significant feature of the commercial banking industry in Thailand is the high degree of concentration in ownership. Ownership is dominated by sixteen families of Chinese origin (Naris 1993). Thai monetary authorities consider this 40 Thailand's Macroeconomic Miracle: Stable Adjustment atnd Sustained Growth concentration to be a problem, and attempts have been made to diversify bank ownership. The stock exchange has been the main venue for transferring owner- ship. Special legislation has limited the number of shares a person may own. This legislation has proved ineffective, however, as banks have not been able to meet the deadlines for ownership diversification and the deadlines have had to be ex- tended repeatedly. The Thai commercial banking industry has a cartel-like structure with its six- teen banks organized loosely under the Thai Bankers Association, through which they collectively set the standard rates for service charges and loan rates. Because of this oligopolistic structure, it takes time for all the banks to agree on the same adjustment, particularly in the downward direction, with the result that interest rates (on loans and deposits) respond rather slowly to market conditions. As a col- lective body, however, Thai bankers possess substantial power in dictating the cost and the allocation of domestic credit and in influencing the effectiveness of mon- etary policies. Within the banking system, firms vary greatly in size and market share. This feature is important to keep in mind in any attempt to understand the bank credit market in Thailand. At present, the market is led by four large banks whose market shares in deposits and credit totaled 70 percent in 1990. These four banks are the Bangkok Bank, the Siam Commercial Bank, the Thai Farmers Bank, and the Krung Thai Bank. They dominate the interbank loan market since they are the main suppliers of liquidity for smaller banks and the foreign banks. In addition, they are the leading players in foreign transactions and thus can exert a degree of control on the supply of foreign exchange. Important decisions regarding interest rates and other price-setting decisions in the money market are influenced by these four banks, through the Thai Bankers Association. REGULATIONS. A number of government regulations affect the financial market, especially the banking industry. The main measures of this kind have been the reg- ulations imposing interest rate ceilings for loans and deposits, the controls on new entry, agricultural credit policy, and compulsory bond holding for branch expan- sion. The Bank of Thailand adjusts the ceiling rates to keep the domestic rates in line with foreign rates, to alleviate liquidity problems, or to implement monetary policy. Another important stipulation is that the ratio of the banks' capital funds to risky assets must be kept above a compulsory minimum. This measure is designed to prevent excessive expansion of bank credit and thereby to ensure the soundness of the banking system. In some years, this measure has made it difficult for banks to reduce excess liquidity through loan expansion. A novel feature of monetary policy that is administered through the banking system is the use of rediscount facilities. The basic objective here is to assist high- priority sectors by providing low-cost funds. The rediscount facility was intro- duced in 1960 to finance rice exports and since then has been extended throughout industry, agriculture, and construction. The Economic Setting: Structure and Perrwrmance ?f the Thai Economv 41 Recent financial market reforms have relaxed the regulatory environment sig- nificantly. In June 1989 the Bank of Thailand lifted interest rate ceilings on time deposits with a maturity of more than one year. In March 1990 this relaxation was extended to time deposits of one year or less. At present, the only remaining ceil- ings are on savings deposits and on lending rates. In May 1990 the capacity of commercial banks to engage in foreign exchange transactions without seeking prior approval from the Bank of Thailand was ex- tended, and capital account transactions in general were subsequently liberalized. From April 1991. foreign exchange accounts could be opened with commercial banks in Thailand for up to $500,000 for individuals and up to $5 million for cor- porations. Thai citizens were permitted to transfer up to $5 million abroad for di- rect foreign investment purposes. Furthermore, Bank of Thailand approval is no longer required for the repatriation of investment funds, dividends, and loan re- payments. BANK FAILURE AND RESCUE OPERATIONS. Thailand's financial system has some history of bank defaults. Before 1966, there were two such defaults, in 1959 and 1960. Under the rescue operation, deposits were mobilized from other commercial banks to address the immediate liquidity problem of the troubled bank. The gov- ernment, through the Bank of Thailand, then organized a new management team to take over the operation of the bank. This procedure has remained the central style of rescue operation until today. Between 1979 and 1986 Thailand experienced two serious crises relating to finance companies and three cases of bank failure. The finance companies had problems with outright fraud, losses due to stock market speculation, loans with- out proper collateral, and excessive lending to one's own business or an affiliated one. The financial crises in 1979 and 1984 shook public confidence in local finan- cial institutions and prompted the Bank of Thailand to launch a major rescue op- eration in 1984. The rescue, which introduced a management-pool style operation, brought ailing finance companies under close government supervision. In 1984, nine finance companies had their licenses withdrawn, seven were brought under the control of the Ministry of Finance, and nineteen were brought into the man- agement pool. Public confidence in local financial institutions was further shaken by a string of bank difficulties in 1984, 1986, and 1987. In each of these cases, the Ministry of Finance and the Bank of Thailand rescued the failing banks by injecting low- interest loans and reorganizing the management team. Several liquidity funds, which received contributions from other healthy commercial banks and finance companies, were set up in the process as a means of injecting the necessary finance into troubled financial institutions. Toward the end of 1985, a new commercial bank decree was approved by the cabinet that empowered the monetary authorities to deal more decisively with ailing financial institutions. The decree led to the es- tablishment of the Fund for Rehabilitation and Development of Financial Institu- tions. This fund, which receives interest-free contributions from all financial 42 Thailands Macroeconomic Miracle: Stable Adjustment andl Sustained Growth institutions and is managed by the Bank of Thailand, is designed to be the basic form of concessional assistance to ailing financial institutions. Private foreign borrowing is relatively free. Although a withholding tax on foreign borrowing exists, it appears to have little effect on the inflow of interna- tional credit. Since the mid-1970s, local commercial banks and large companies have used foreign borrowing as a means of adjusting their liquidity positions. This feature makes local liquidity highly responsive to changes in foreign interest rates and the exchange rate. It seems that when the foreign interest rate was high or when there was speculation about a baht devaluation, capital inflow tended to slow down. Capital outflows, while officially requiring Bank of Thailand approval, oc- cur through quasi-legal channels such as transfer pricing. Domestic interest rates, constrained by the ceilings set by the Bank of Thailand, do not rise corresponding- ly. As a result, liquidity in the domestic money market is tight. The reverse is true when foreign interest rates are low or the baht is strong. Thailand's financial system is therefore prone to excess liquidity when the world interest rate declines, and this excess liquidity tends to be prolonged. The problem is that local commercial banks have a rather limited portfolio choice because the coun- try's capital markets are not well developed and capital outflow, in the form of invest- ing in foreign assets, has in the past been tightly regulated. Most banks therefore hold substantial amounts of government bonds and investment in short-term money mar- kets such as treasury bills and bonds in repurchase markets. Another factor prolong- ing excess liquidity is the rigidity in interest rate adjustment, as noted earlier. THE STOCK MARKET. Since the establishment of the Securities Exchange of Thai- land in 1975, the stock market has grown steadily. The market recovered from a crash in 1979, and the ratio of market capitalization to GDP increased from less than 4 percent in 1980 to 29 percent in 1990 (Robinson and others 1991: 22). Confi- dence in the market was increased by reforms instituted in 1984 to prohibit all in- sider-trading activities and to improve the supervisory and regulatory framework. Economic Performance Thailand's economic performance can be assessed from indicators such as nation- al income, inflation, balance of payments, foreign debt, and income distribution. National Income and Growth A detailed statistical summary of Thailand's macroeconomic performance from 1970 to 1991 is provided in table 3.4. In 1990, GNP per capita (not shown in the table) was U.S.$ 1,420 and in the preceding twenty-five years GNP per capita grew at an average annual rate of 4.4 percent, in real terms. From 1965 to 1980, GDP (total, not per capita) grew at an average annual rate of 7.2 percent and then slowed Table 3.4 Macroeconomic Summary, 1970-91 (percent growth rate, unless otherwise indicated) Terms of trade Current (export/import account Total debt Total debt Exchange unit value) Inflation balanceJGDP Real money GNP service/exports rate (baht/ Year Real GNP Exports Imports (percent) (percent) (percent) supply (Ml) (percent) (percent) U.S. dollar) 1970 7.4 0.3 4.0 100.0 0.8 -3.8 9.7 16.6 17.1 20.8 1971 4.6 10.7 1.2 101.0 0.4 -2.5 11.0 17.2 18.9 20.8 1972 5.4 26.6 13.8 111.0 4.8 -0.6 17.7 16.8 17.4 20.8 1973 9.1 33.2 36.1 155.0 15.6 -0.5 17.9 14.3 15.3 20.4 1974 4.1 44.9 49.2 130.0 24.3 -0.7 13.0 13.2 14.8 20.0 1975 5.0 -7.8 3.8 116.0 5.3 -4.1 11.0 15.5 15.1 20.0 1976 8.9 24.9 12.2 107.0 4.2 -2.7 12.4 13.1 12.8 20.0 1977 9.8 14.9 30.3 101.0 7.1 -5.7 9.0 14.8 16.7 20.0 1978 9.3 21.7 15.9 102.0 8.4 -1.5 17.1 18.5 17.4 20.3 44 1979 5.9 29.4 38.4 105.0 9.9 -7.7 17.0 20.2 19.1 20.4 1980 6.2 27.0 23.2 100.0 19.7 -6.2 13.8 25.7 14.5 20.5 1981 5.2 14.1 14.3 87.0 12.7 -7.1 6.5 31.0 14.4 21.8 1982 4.8 6.0 -9.6 79.0 5.2 -2.7 12.0 34.2 16.0 23.0 1983 7.1 -4.6 20.1 85.0 3.8 -7.3 10.3 35.0 19.1 23.0 1984 6.3 14.1 3.8 83.0 0.9 -5.1 5.4 36.4 21.5 23.6 1985 3.0 10.5 4.6 80.0 2.4 -4.1 8.4 46.9 25.3 27.1 1986 4.6 20.7 -3.0 89.0 1.9 0.6 18.2 44.6 25.4 26.3 1987 9.7 28.8 39.0 89.0 2.5 -0.7 24.9 35.9 17.1 25.7 1988 13.3 33.9 46.1 86.0 3.8 -2.7 8.0 30.1 13.7 25.3 1989 12.4 27.7 29.8 83.0 5.4 -3.6 11.7 28.2 12.4 25.8 1990 10.0 14.4 28.5 81.0 6.0 -4.9 8.5 31.3 9.8 25.6 1991 8.2 22.5 12.6 77.8 5.7 -8.6 13.8 39.0 13.1 25.5 Source: Bank of Thailand, Quarterlv Bulletin, various issues; World Bank, World Development Report, various issues; International Monetary Fund, International Financial Statistics, various issues, and World Outlook, May 1993 (for 1991 GNP only); and Asian Development Bank, Asian Development Outlook, 1992 (for 1991 current account only). 44 Thailands Mac roetonomic Miracle: Staible Adjustmenit andl Sustained Growth somewhat to 5.6 percent in the period to 1987. Growth accelerated dramatically in the following four years, exceeding 10 percent in each year. Agriculture accounted for 12 percent of GDP in 1990, industry for 39 percent, and services for 48 percent. The corresponding distribution for 1965 was agricul- ture 32 percent, industry 23 percent, and services 45 percent. In the Philippines during the same year, the contribution of agriculture to GDP was substantially higher (22 percent), even though Philippine agriculture absorbed a smaller propor- tion of the total work force. From 1980 to 1990 aggregate GDP grew at an average annual rate of 7.6 per- cent, compared with 7.3 percent from 1965 to 1980. From 1980 to 1990 the annual growth rates of sectoral contributions to GDP were agriculture 4. I percent, industry 9.0 (of which manufacturing accounts for 8.9 percent), and services 7.8 percent. From 1965 to 1980 the corresponding growth rates were agriculture 4.6 percent, industry 9.5 (of which manufacturing made up 11.2 percent). and services 7.4 per- cent. The slower growth of agricultural output than of either industry or services is especially important. Total factor productivity growth in Thai agriculture has been slower than elsewhere in the economy (Ammar, Suthad, and Direk 1993), but this is only part of the story. Agriculture has also been releasing vast resources to the rest of the economy through the process of structural change, described below. Thai national income data before 1970 are considered to be questionable and the NESDB has released a revised national income series for the period since 1970. Thailand's annual rate of GNP growth from 1971 to 1991 calculated from this new series is shown in figure 3.1. The annual fluctuations around the average rate of growth of real GDP of 7.1 percent principally involved external phenomena. Do- mestic political events have had remarkably little apparent effect on aggregate economic perfornance. Thailand experienced an export commodity boom from 1972 to 1974. This boom affected Thailand mainly through the price of rice. Ex- ports surged and economic growth rose sharply in 1973. This event was quickly followed by the first international oil price increases of 1973-74 induced by the Organization of Petroleum Exporting Countries (these increases are known as OPEC 1). Since Thailand is a substantial petroleum importer, the rise in oil prices slowed Thai growth in 1974 and 1975. Recovery occurred from 1976 to 1978, but the second round of oil price increases of 1979-80 (OPEC 11) slowed growth again. Other oil-importing countries, including the Philippines, were devastated by these petroleum price increases, but Thailand experienced only a growth recession, from about 9 percent to about 6 percent. Having borrowed internationally to finance the increased cost of petroleum imports during the mid- to late 1970s, Thailand suf- fered again from the high international interest rates of the early 1980s. By 1985 serious macroeconomic problems were evident, as already mentioned. The boom that began in 1987 surprised Thai observers as much as outsiders. The boom was apparently driven by three principal forces: the depreciation of the U.S. dollar in relation to other currencies and the fact that the baht was pegged to it, which made Thai exports more competitive internationally; foreign investment, especially from two of the present newly industrizalizing economies (NIES), Taiwan The Economic Setting: Structure and Performance of the Thai Fconomy 45 Figure 3.1. Real GNP Growth Rate, 1971-91 Percent 12 T 6 4 2t 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 Source: Table 3.4 and authors' calculations. and Hong Kong, which wished to avoid rising labor costs in their own countries. and continuing low international petroleum prices in relation to those of Thailand's export commodities. In explaining Thailand's remarkable growth boom of the late 1980s, it would be easy to give too much weight to Thai-specific causes. Similar foreign invest- ment-led booms were occurring in Malaysia and Indonesia, and even more signif- icantly, in the southeastern corner of China, adjacent to Hong Kong and Taiwan. Country-specific factors did play a role, of course. The foreign investment did not flow to the Philippines. Its unreliable infrastructure, labor unrest, and high level of political uncertainty were enough to discourage it. In focusing on Thailand's recent record of rapid growth, it is also easy to lose sight of the fact that Thailand's economic performance over the long term is very different from this. Long-term changes in Thai real income can be estimated from raw data provided in Sompop (1989). These estimates, expressed as real GDP per capita over the period 1870 to 1950. are shown in table 3.5. These data imply an average growth rate over this period of a mere 0.2 percent per year. Growth since World War 11 can be estimated more reliably. (Figure 1 .1 shows this information for 1951-91, calculated from national income data provided by the NESDB.) In contrast with the century before, the average annual rate of growth of measured real GDP per capita over this period was 4.3 percent. 46 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 3.5 Estimated Gross Domestic Product, 1870-1950 (million of baht per year at 1950 prices) GDPper capita Population (bahtper Year Agriculture Manufacture Services GDP (millions) year) 1870 2,417 678 2,524 5,619 5.775 973 1890 2,959 828 3,035 6,822 6.670 1,023 1900 3,222 883 3,274 7,379 7.320 1,008 1913 4,459 1,245 4,281 9,985 8.689 1,149 1929 5,735 1,603 5,749 13,087 12.058 1,085 1938 7,490 2,091 7,337 16,918 14.980 1,129 1950 10,196 2,794 9,559 22,549 19.817 1,138 Source: Calculated from data in Sompop (1989). Although Thailand has experienced rapid economic growth over the past for- ty years, it is still a poor country today. For at least the preceding eighty years, and one presumes much longer, economic stagnation rather than sustained growth had been the norm. Figure 3.2 illustrates this point for the period 1870-1990. The fig- ure combines Sompop's data on real GDP per capita for the period 1870-1950 (from table 3.5, re-based to 1972 prices) with the NESDB data for the period since Figure 3.2. Per Capita GDP at Constant 1972 Prices, 1870-1990 Baht 12000 10,000 8,000 6,000 4,000 2,0000K - 1870 1890 1900 1913 1929 1938 1950 1960 1970 1980 1990 Source: Sompop (1 989) and National Economic and Social Development Board, Bangkok. The Economic Setting: Structure and Pertormnance of the Thai E(onomv 47 1951. the latter series summarized at intervals of ten years. Economic growth, rap- id or otherwise, is quite a recent experience in Thai history. Structural Change Figures 3.3, 3.4, and 3.5 show the long-term patterns of structural change in Thai- land, expressed as the share (at constant prices) of agriculture and manufacturing in GDP and also the share of the agricultural labor force in total employment. In 1985 manufacturing overtook agriculture as a share of Thailand's GDP. The decline in agriculture's share of GDP coincided with an expansion of industry's GDP share. The share of services in GDP barely changed (figure 3.3), but still ex- ceeded industry's share. Agriculture's share of the total labor force also declined but its fall lagged well behind agriculture's declining share of national income (figure 3.5). The slow decline in agriculture's share of employment was matched by a rise in the services share, but not in that of industry (table 3.5). Table 3.6 shows that the share of services in Thai national output is unusually high. Exports, Imports, and the Terms of Trade Thailand's external terms of trade (the ratio of the average international prices of its exports to those of its imports) are shown in figure 3.6. From 1965 to 1990 Figure 3.3. Sectoral GDP Shares, 1965 and 1990 Percent (current prices) 100 90 80 1t990 70 60 50 40 307 20 10 Agriculture Industn Services Source. World Bank, 1992. 48 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 3.4. Sectoral Labor Force Shares, 1965 and 1986-89 Percent 100 90 80 70 _ 1 1965 60 * 1986-89 50 40 30 20- 10 1- Agriculture Industrm Services Source. UnitedNations Development Programme, 1992. Thailand's terms of trade apparently declined from an index of 100 to an index of 65. The index surged upward following the 1973 commodity price boom and then fell significantly with each of the two OPEC petroleum price shocks (in 1973-74 and 1979-80). The long-term decline in the terms of trade tapered off in the 1980s, and the growth rate of Thailand's merchandise exports accelerated through this period. The composition of these exports moved away from primary commodities and to- ward manufactured goods (table 3.7). By 1985 the total exports of manufactured goods exceeded those of primary goods, and by 1990 exports of textiles and cloth- ing alone well exceeded those of rice, Thailand's traditional export commodity. Until the late 1970s, most of Thailand's export industries were agro-based and included rice milling, frozen seafood, and canned fruit, which use agricultural products as raw materials. From 1980 onward, these exports became more diver- sified, with the addition of canned fish and crustaceans, garments, rubber sheets and rubber products, wood products, jewelry, and footwear. The new export indus- tries-most notably, garments and footwear-tend to be labor-intensive. We now present a new measure of the export performance of an industry that is designed to reveal the degree to which that industry is competing successfully in international markets. Our index of trade performance is based on earlier work of Balassa and others on "revealed"comparative advantage but is intended to im- prove upon these earlier measures. We call it the net export performance ratio (NEPR): The Economic Setting: Structure and Performance qf the Thai Economy 49 (XTLMT )/Xw (3.1) NT= __ ' _ X T/X * where NT denotesThailand's NEPR for industryj, X"fand MTdenote Thailand's gross exports and gross imports of commodity j, respectively; XJW denotes world exports of commodity j; X* denotes Thailand's total exports of all goods: and X *w denotes total world exports of all goods. Thus the index measures the de- gree to which Thailand's net exports of commodity j, as a share of world exports of that commodity, exceed or fall short of Thailand's share of world exports in gen- eral. Table 3.8 shows the value of the NEPR for sixteen industries from 1970 to 1989. The annual trade data are summarized by computing the NEPR ratios over five-year periods to minimize the effect of random year-to-year variations in trade data. Ac- cording to these analyses. Thailand's agricultural exports continue to perform well. Textiles are also becoming important in this respect. and furniture and nonmetallic Figure 3.5. Agricultural Share of GDP Compared with Agricultural Share of Labor Force and Manufacturing Share of GDP, 1960-90 Millions of U.S. dollars 100 - 80 70 \ -~ 60 50 7 40 30o ___ _- ----- ,- 20 .... . _ . - . _ _ 10 0 TV ITfl VVTVfI I I 1960 63 66 69 72 75 78 81 84 87 90 - - - --- Manufacturing's share in GDP - - - - -Agriculture's share in GDP Agriculture's share in labor force Source. National Economic and Social Developing Board, Bangkok. 50 Thailand 's Mac roeconomic Miracle: Stable Adjustment and Sustained Growth Table 3.6 Structure of Production: Distribution of Gross Domestic Product, 1965-90 (percent) Agriculture Industrv Manufacturing Senrces Country 1965 1990 1965 1990 1965 1990 1965 1990 China 44 27 39 42 31 38 17 31 India 44 31 22 29 16 19 34 40 Indonesia 56 22 13 40 8 20 31 38 Korea, Republic of 38 9 25 45 18 31 37 46 Philippines 26 22 28 35 20 25 46 43 Malaysia 28 19 25 42 9 32 47 39 Thailand 32 12 23 39 14 26 45 48 Note: Manufacturing is a component of industry. Except for rounding errors, the shares of agricul- ture, industry, and services should sum to 100. Soun re: World Bank. World Developmenit Report 1992. manufactured products show a steadily improving export performance. The relative performance of less highly processed wood products has declined. At least part of Thailand's export growth can be attributed to changes in policy, but the degree to which this is so is the subject of some controversy. It is true that Figure 3.6. Overall Terms of Trade Index, 1961-90 (1970 = 100) 140.00 120.00 100.00 Bk o Ths ofatrade 80.00- 60.00 _ 40.00 20.00i 0.00 , I I tI I t + + - I I I I I - 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 Source: BankorThailand. Table 3.7 Growth and Structure of Merchandise Exports, 1965-90 (percentage of total) Machinery, Annual growtth rate Fuels, minerals, Otherprimarv transport, and Other Textiles and of exports and metals comnmodities equipment manufacturing clothing Country 1965-80 1980-90 1965 1990 1965 1990 1965 1990 1965 1990 1965 1990 China - 11.0 15 10 20 16 9 17 56 56 29 27 India 3.0 6.5 10 8 41 19 1 7 47 66 36 23 Indonesia 9.6 2.8 43 48 53 16 3 1 1 34 0 11 Korea, Republic of 27.2 12.8 15 2 25 5 3 37 56 57 27 22 Malaysia 4.6 10.3 34 19 60 37 2 27 4 17 0 5 Philippines 4.6 2.5 11 12 84 26 0 10 6 52 1 7 Thailand 8.6 13.2 11 2 86 43 0 20 3 44 0 16 Source: World Bank, World Development Report 1992. 52 Thailand s Macroeconomic Miracle: Stable Adjustment andl Sustained Growth Table 3.8 Net Export Performance Ratios, by Industry, 1970-89 Export 1970-74 1975-79 1980-84 1985-89 Agriculture 5.54 6.16 6.43 4.43 Tobacco -0.74 -0.57 -0.46 -0.35 Textiles 0.20 1.24 1.77 2.16 Wood products 1.29 0.93 0.30 0.42 Furniture -0.16 0.15 0.57 1.05 Paper -0.96 -0.98 -0.86 -0.50 Printing -0.99 -0.61 -0.53 -0.38 Rubber -0.65 -0.49 -0.33 0.04 Chemicals -2.20 -2.07 -1.74 -1.44 Petroleum -2.27 -1.69 -1.56 -1.43 Nonmetallic manufacturing -0.74 -0.49 -0.22 0.09 Metal products -0.68 -0.75 -0.67 -1.27 Machinery -2.18 -1.95 -2.06 -1.75 Electrical machinery -1.23 -0.95 -0.71 -0.40 Transport -1.28 -1.25 -0.88 -0.74 Other manufacturing -0.05 -0.10 0.27 0.76 Source: Computed from United Nations trade data provided by the International Economic Data Bank, Australian National University. export taxes were reduced from 1976 onward and that some export industries, in- cluding textiles and sugar, received assistance from the Bank of Thailand through loans under concessional terms channeled through the commercial banks. Generous terms were also offered for the repayment of concessional export credits. Neverthe- less, these policy changes could hardly have generated the growth of manufactured exports that actually occurred. The rhetoric of Thailand's development plans has led some observers to as- sume mistakenly that export promotion policies produced the export expansion. Beginning with the Third Plan (1972-76), export promotion was indeed stressed over import substitution in the plan documents, but this change of language reflect- ed more an effort to keep up with intellectual fashion than a policy commitment. The protection of inefficient manufacturing sectors actually increased during this period, as it did through the remainder of the 1970s. The relationship between the government's various instruments of industry policy and the performance of the in- dustries receiving assistance is explored in greater depth in chapter 4. The Economic Setting: Structure and Performance of the Thai Economy 53 Between 1974 and 1981 the volume of merchandise exports grew at 11.3 per- cent, whereas real GDP and the volume of merchandise imports grew at 7.1 and 6.6 percent, respectively. Between 1970 and 1982, agricultural exports as a share of total merchandise exports declined from 71 percent to 64 percent. Nonagricultural primary exports declined from 15 to 7 percent and manufactured exports grew from 14 to 29 percent. Over the same period, total merchandise exports as a pro- portion of GDP grew from 10 to 19 percent. Roughly two-fifths of this growth was accounted for by primary exports and three-fifths by manufactured exports. From 1986 to 1990 manufactured exports continued to flow rapidly. Led by the growth of manufactured exports, total merchandise exports grew at an average rate of 25 percent from 1986 to 1990. In the decade to 1990 the share of manufac- tured exports in total exports more than doubled from 30 to 60 percent. As a share of GDP, the total exports of goods and nonfactor services also doubled over the same decade, from 17 to 35 percent. Exports of textile products exceeded the value of rice exports for the first time in 1985, and only two years later were more than twice as important as rice exports. Three features of the destinations of Thailand's exports are notable. First, a high proportion goes to developing countries (38 percent compared with an aver- age of 27 percent for lower middle-income countries). Second, of the developing countries, a good number are the fast-growing East Asian countries. Third, of the exports going to industrial countries, an unusually high proportion goes to Japan (26 percent), which also has enjoyed a high long-term growth rate. Data on the composition of Thailand's imports show a declining relative im- portance of consumer goods, matched by a steadily rising share of intermediate goods and, to a lesser extent, capital goods (Warr 1993). These changes reflected Thailand's rapidly developing but still early industrialization, which greatly in- creased the demand for imported intermediate goods. Imported petroleum prod- ucts consist primarily of crude oil. Their share in total imports rose steeply with the petroleum price increases of 1973-74 and 1979-80 and declined again with the price declines of the 1980s. Some import replacement also also took place, as domestically produced natural gas increasingly replaced imported crude oil through the 1980s (Praipol 1993). Inflatioti Thai economic policy has been shaped by a strong aversion to inflation. This is es- pecially true of the monetary policies implemented by the Bank of Thailand since World War II. To what extent has this goal been achieved? Figure 3.7 shows the an- nual rate of inflation for the period since 1938. The data for 1938-48 are based on rice prices, and those for the period since 1948 are from cost-of-living surveys. After a rapid increase during and shortly after World War II, inflation has re- mained below 5 percent. except for two brief surges associated with the petroleum price increases of the 1970s. Even then, inflation was quickly brought under con- trol by stringent monetary contractions. It is well known in Thai financial circles 54 Thailand's Macroeconomic Miracle: Stable Adjustment aind Sustained Growth Figure 3.7. Inflation Rate, 1938-90 Percent 45 40 35 30 25 20 1 5 10 5 0- 1940 1950 1960 1970 1980 1990 Source: Bank of Thailand. See also data discussion in chapters 2 and 3. that the Bank of Thailand will contract monetary policy whenever inflation rises above 6 percent and will persist with this policy until the rate falls below 6 percent. Thailand's monetary policy therefore has considerable credibility. Inflationary ex- pectations do not become a serious obstacle to restraining inflation. Thailand's record of monetary management is clearly impressive. External Debt Thailand avoided the severe debt crises experienced by many developing countries in the 1 980s, although by Thailand's conservative standards its levels of debt were still a problem. In 1990 total debt service as a proportion of exports of goods and services was moderately high by East Asian standards (table 3.9). The stock of debt in relation to GDP and the annual volume of exports was large but manageable, be- cause of the high volume of exports. Nevertheless, the level of debt was such that Thailand could ill afford a significant decline in the value of its exports. Thailand's adjustments to the oil price shocks of the 1 970s and the high inter- est rates of the early 1980s were financed to a large extent by foreign borrowing. Such a policy is sustainable as long as the borrowed funds are invested wisely, so that the loans can be repaid. Thailand has managed to follow this dictum, but the Philippines, for example, failed in this respect, while Indonesia, which is more heavily indebted than either of these two countries, appears to have used the bor- rowed funds in productive ways. The Econonmic Setting: Structure and Perlarmrnance *)Jthe Theai Econommy 55 Table 3.9 Total External Debt and Total External Debt Ratios, 1980 and 1990 Total external Total debt Total external debt as Total external service as debt percentage of debt as percentage o?f (nmillions of export ofgoods percentage of exports ofgoods U.S. dollars) and services GVP anld services Coutitry 1980 1990 1980 1990 1980 1990 1980 1990 China 7,972 52.555 22.1 77.4 1.5 14.4 4.6 10.3 India 20,560 70,115 136.0 282.4 11.9 25.() 9.3 2X8 Indonesia 20,888 67,908 94.2 229.4 28.0 66.4 13.9 30.9 Korea, Rep. of 29,749 34,014 130.6 44.( 48.7 14.4 19.7 10.7 Malaysia 5,195 19,502 44.6 55.9 28.0 48.0 6.3 11.7 Philippines 17,386 30,456 212.5 229.2 49.5 69.3 26.5 21.2 Thailand 8.257 25,868 96.3 82.0 25.9 32.6 18.7 17.2 Source: World Bank, World D)evelopment Report. various hsUcs Balanc e of PaYments As in all countries operating with fixed exchange rates, Thai economic policy dis- cussion is dominated by a concern about the balance of payments. This preoccu- pation is almost certainly excessive, but it does act as a restraint on policies that would imply unsustainable external deficits. Thailand's international reserves are satisfactory, in relation to its level of imports and GDP' (summarized in table 3.10). In 1990 these reserves were equivalent to 4.4 months of import coverage, com- pared with 1.5 months for the Philippines. 3.2 months for Indonesia, and 1.9 months for India. For many African countries, reserves are equivalent to less than one month's import coverage. Foreign Investment In the late 1980s Thailand became a major recipient of direct foreign investment. The magnitude of this foreign investment boom is shown by figure 3.8, and data on the net inflows and the regions they came from can be seen in table 3.11. By 1991 the composition of this foreign investment had shifted away from Thailand's traditional sources-Japan, the United States, and Europe-and toward the coun- tries of Northeast Asia. In 1990 Northeast Asia accounted for 60 percent of the $2 billion total net direct foreign investment in Thailand. the Association of South- 56 Thailand x Mlacroe( onornic Miracle: Stable A dju smentr autdi Sustained (;ro wth Table 3.10 Current Account Balance and Reserves, 1970 and 1990 (inillion7s of'UNS. dollars) Current account Ajter official Befr) refi ii al Gross international tran sfers transfers reserves Country 1970 1990 1970 1990 1970 1990 China -81 12.000 -81 11,935 - 34,476 India -385 -9,304 -591 -9,828 1.023 5,637 Indonesia -310 -2,369 -376 -2.430 160 8,657 Korea, Rep. of -623 -2,172 -706 -2,181 610I 14,916 Malaysia 8 -1,672 2 -1,733 677 10,659 Philippines -48 -2,695 -138 - 3,052 255 2,036 Thailand -250 -7,053 -296 -7,235 911 14,258 Source: World Bank, Wlorld Development Report 1992. east Asian Nations (ASEAN) 13 percent, Europe 10) percent, and the United States 12 percent. Japan alone represented only 30 percent, compared with 44 percent in 1986. Rising in importance were Hong Kong (23 percent). Taiwan, China (6 per- cent), and Singapore (13 percent). These three sources together accounted for 40 percent of the total, well exceeding investment from Japan. Figure 3.8. Net Incoming Private Foreign Direct Investment, 1971-91 Millions of lU.S. dollars 2,500 2,000 1.500 1,000 500 1971 1 973 1975 1977 1979 1981 1 983 1985 1987 1989 1991 Source: Bank of Thailand. i/7e EF yoini( Setting: Structire (11(l Perfirman, c eof (the Tlhai EcanomY 57 Table 3.11 Direct Foreign Investment into Thailand, Net Flows, 1980-85 to 1991 f millio}ns (?/ t U.s dlolhvs) Source 1980-85" 1986 1987 1988 J989 1990 1991 ASEAN 14 14 21 65 109 252 253 Europe 45 30 67 111 196 198 196 North America 84 51 71 128 2(9 232 237 Northeast Asia 103 159 188 831 1,16() 1,683 1.179 Other 22 9 5 4 97 77 130 Total 268 263 352 1,139 1,771 2,442 1,995 a. Average per Xcar. Source: Bank of Thailand(. Faced with tising labor costs that were making labor-intensive manufacturing less profitable, firms in established NiEs began looking for new places to invest. Thai- land was seen as an attractive host. By 1989 this source of investment had helped boost Thailand's total foreign investment (net direct investmiient plus portfolio invest- ment) to $3.2 billion (the direct investment was $ 1 .8 billion and the portfolio invest- ment $1.4 billion). This total accounted for 15 percenit of Thailand's gross fixed investment. Although foreign investmetnt has becomne important for Thailand, domes- tic investment remains f:ar more important (Bank ol'Thailand 1992). Inctomrne Distribution tntid Poi'ertY Although the distribution of income and the incidence of poverty are significant factors in Thailand, little analytical work has been done on this subject. Most of the studies in this area have been descriptive works based almost entirely on a sin- gle set of flawed and incomplete data compiled during periodic socioeconomic surveys (SES) conducted by the National Statistical Office (Nso). Because the sam- pling procedures and definitions used by the NSO have changed repeatedly-and there is in any case no standard definition of poverty threshold levels of income- it has been difficult to make comparisons across time. Furthermore, the economic determinants of Thailand's income distribution have yet to be investigated in de- tail. The question remains, what broad conclusions, if any. can be drawn about the changing distribution of income in Thailand" The SES data on the distribution of income over the period from the mid- 1970s to the late 1980s were presented in figure 1.2. We concentrate on this interval be- cause the NSO statistical definitions were reasonably constant during that time. The proportion of total income received by the poorest one-fifth of the population de- 58 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 3.12 Poverty Incidence and Economic Growth, 1975-76 to 1988 Poverty 1975/76 1981 1986 1988 Lines (baht per capita per year constant prices) Urban 2,961 5,151 5,834 6,203 Rural 1,981 3,454 3,823 4,076 Incidence CommunitY (percent) Municipal areas 12.5 7.5 5.9 6.1 Sanitary areas 14.8 13.5 18.6 12.2 Villages 36.2 27.3 35.8 26.3 Region Bangkok and vicinities 7.8 3.9 3.5 3.5 Central 13.0 13.6 15.6 12.9 North 33.2 21.5 25.5 19.9 Northeast 44.9 35.9 48.2 34.6 South 30.7 20.4 27.2 19.4 Whole kingdom 30.0 23.0 29.5 21.2 Average growth rate of real GNP over preceding period (percent per year) 5.9 7.5 5.2 11.5 Source: The poverty incidence data for 1975/76. 1981, 1986, are from Suganya and Somchai (1988) and those for 1988 from Medhi, Pranee. and Suphat (1991). In both of these studies poverty incidence was calculated by applying the rural poverty lines to sanitary areas. The GDP growth rates are calcu- lated from table 3.4. clined over this period, whereas the proportion received by the richest quintile rose. In other words, the distribution of income became more unequal. Although increased inequality is generally considered to be undesirable, in this case the widening inequality does not necessarily imply that the poor became worse off, because average incomes also increased over the same period. The share of total income received by the poorest quintile fell from 6.1 to 4.5 percent over the period 1975-76 to 1988, but total Thai income rose in real terms from 5,200 to 9.500 baht per capita in constant 1972 prices, which represents an in- crease of about 83 percent. According to this calculation, the absolute real income of the poorest quintile rose by 35 percent (from 317 to 428 baht per capita, again in constant 1972 prices), even though the richest quintile gained proportionately The Economic Setting: Structure and Performanc e of the Thai Economy 59 three times as much: the income of this group rose from 2,564 to 5,216 baht per capita, or 103 percent. Absolute poverty can also be assessed by measuring the proportion of the population whose incomes fall below a designated poverty line. Despite the arbi- trary nature of any such cutoff point, the change in poverty incidence, so mea- sured, may not be especially sensitive to the particular point that is selected. Measurements of this nature (see Suganya and Somchai 1988; Medhi, Pranee. and Suphat 1991; and table 3.12) indicate that absolute poverty in Thailand is princi- pally a rural phenomenon, with the heaviest concentrations in the Northeast re- gion. As for changes in the incidence of poverty over time, the SES data suggest that from 1976 to the early 1 980s the incidence of poverty declined but then wors- ened until the mid- 1 980s, when it declined again. This fluctuating pattern raises some important questions about the impact of growth on poverty. It is obvious that over the long term sustained economic growth is a necessary condition for poverty alleviation; no amount of redistribution could turn a poor country into a rich one. But it is not obvious that growth favors poverty reduction in the short run. Indeed, observers of the Thai economy have frequently asserted the opposite: that economic growth has failed to benefit the poor and may even have harmed them. When the data on economic growth are plotted against the trends in poverty reduction, however, it becomes clear that-over this period, at least-the faster the growth, the greater the poverty reduction (figure 3.9). Rapid growth from 1976 to 1981 coincided with a decline in the incidence of poverty. Reduced growth caused by the world recession in the early to mid- 1 980s coincided with a worsening inci- Figure 3.9. Povertv Reduction and Economic Growth, 1975-88 Poverty reduction (percent per year) 5 5 T (1987-88) 4-a { (I975,/76- 1981) (1982-86) -2 -3 I I + 1-1 2 3 4 5 6 7 8 9 10 11 12 Real GNP growth rate (percent per year) Source: Table 3.12. 60 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth dence of poverty up to 1986. The economic boom of the late 1980s coincided with a markedly reduced incidence of poverty. Of course, it would be absurd to say that rapid growth, and nothing more, is the answer to poverty, and equally absurd to deny that any increase in relative inequality could become a serious social prob- lem. At the same time, the evidence provides no support for the suggestion that rapid economic growth is bad for the poor in absolute terms. On the contrary, the rate of growth may be the single most important determinant of the rate at which poverty declines, even in the short run. Appendix: Thailand's Macroeconomic Data Base The macroeconomic data from Thailand can be divided into two groups: those from the period up to 1954 and those from 1954 to the present time. Macroeconomic Data up to 1954 Official estimates of Thailand's GDP are available only from 1 946. These were pre- pared by the National Income Division, National Economic and Social Develop- ment Board (NESDB). These data are shown in table 3.13. Official estimates of GDP at constant prices are available only from 1951; estimates for earlier years (that is, 1870, 1890. 1900, 1913, 1929, 1938. and 1950) can be found in Sompop (1989). These data were calculated at 1950 prices and are shown in table 3.5. The long- term structural change implied by Sompop's data for the period before 1950 and the NESDB data for the period since 1951 are summarized in figure 3.1(0. The pic- ture that emerges from these data is one of a stagnant economy before 1950 and a rapidly modernizing one thereafter. The historical series constructed by Sompop was an attempt to study the way the Thai economy was affected by the open international trade environment that resulted from the Bowring Treaty of 1855. On the production side, Sompop esti- mated the gross value added of each sector. In the sectors where hard data were available, his estimates were based on the actual output minus intermediate trans- actions. These sectors were rice, rubber, teak, and tin. Estimates for the remaining sectors (other crops, manufacturing, and services) were obtained by extrapolating backward, assuming a constant proportion of value added to GDP in those sectors: 1950 was used as the benchmark year and estimates were adjusted to take into ac- count population growth. Sompop's estimates may therefore understate the degree to which structural change actually occurred. The accuracy of Sompop's estimates depends on the level of disaggregation and the available data. For the years before 1918, no data are available on the output of other crops and other activities and therefore his estimates may be the only available source of information on these sectors. After 1918, however, data on the output of other crops (such as maize, cotton, tobacco, pepper, and livestock) begin to appear The Ecoomric Setting: Sthucture and Perfrttance ofthe Thai Economy 6] Figure 3.10. Shares of Agriculture, Manufacturing, and Services in GDP at Current Prices, 1870-1990 0.8 - 0.6-- 0.4-- 0.2-- 0 1870 1890 1900 1913 1929 1938 1950 1960 19701 1980 1990 El Agriculture 0] Manufacturing U Servnice in the Statistical Yearbook of Thailand and can be used to improve Sompop's esti- mates. According to Sompop, real GDP (at 1950 prices) during the periods 1870- 1913 and 1913-50 grew by only 1.3 and 2.2 percent, respectively, despite the expan- sion of the infrastructure and Thailand's transformation from a subsistence to a mar- ket economy in those years. Thus the question may arise as to whether the results would change if the estimation was based on the expenditure approach. The official estimates of GDP in the earlier years were highly inaccurate. The es- timates for 1946-50 were prepared by Joseph S. Gould. national income adviser, who used the expenditure approach. But according to Paul B. Trescott, Gould's cal- culations understated the actual levels of income and product. He in turn produced "best guess" estimates, which were published in Ingram (1971: 222). Nevertheless, there was only a 5 percent difference between Trescott's estimates and the official ones. Data for the subsequent years were estimated by the National Income Division of the NESDB, but frequent revisions in the 1950s and 1960s made the sequence of statistics intermittent and incompatible year by year. In 1989 GDP estimates were ex- tensively revised and a new series, from 1970 to 1987, published. Table 3.13 shows only the revised estimates from the official sources as compiled by Wilson (1983) for the years 1946-69. and the NESDB new series from 1970. Table 3.13 National Income at Current Market Prices, 1946-89 Agriculture Services GDP GNP GDP per GDP per GNP per GNP per (million Manufacturing (million (million (million Population capita capita capita capita Year baht) (million baht) baht) baht) baht) (million) (baht) growth (%) (baht) growth (%) 1946 6,272 1,146 2,915 10,333 10,333 17 - 597 -597 - 1947 9,549 1,641 4,649 15,839 15,839 18 881 48 881 47.7 1948 11,211 1,706 5,540 18,457 18,457 19 997 13 997 13.1 1949 13,332 2,545 6,322 22,199 22,199 19 1,165 17 1,165 16.8 1950 14,650 3,239 7,706 25,595 25,595 20 1,304 12 1,304 12.0 1951 14,139 2,901 11,170 28,210 28,220 20 1,395 7 1,396 7.0 1952 12,944 3,288 13,289 29,521 29,549 21 1,417 2 1,419 1.6 1953 14,018 3,714 14,496 32,228 32,164 21 1,502 6 1,499 5.7 1954 12,830 3,778 15,436 32,044 31,997 22 1,450 -3 1,448 -3.4 1955 16,568 4,648 18,232 39,448 39,334 23 1,733 20 1.728 19.4 1956 16,586 4,970 19,533 41,089 40,929 23 1,753 1 1,746 1.0 1957 16,990 5,510 22,948 45,448 45,195 24 1,882 7 1,871 7.2 1958 19,099 5,229 22,843 47,171 47,021 25 1.897 1 1,891 1.0 1959 18,964 5,831 25,651 5(0,446 50.310 26 1,969 4 1,964 3.9 1960 21,463 6,759 25,762 53,984 53,885 26 2,046 4 2.)042 4.0 1961 23,111 7,727 28,132 58.970 58,877 27 2,167 6 2,164 6.0 1962 23,689 8,997 31,107 63,793 63,695 28 2,274 5 2,271 4.9 1963 24,564 9,653 33,862 68,079 68.039 29 2,354 4 2.353 3.6 1964 25,008 10,435 39,224 74,667 74,589 30 2,504 6 2,501 6.3 1965 29,383 11,978 42,942 84,303 84,292 31 2,742 10 2,742 9.6 1966 36,921 13,910 50,459 101,290 101,282 32 3,195 17 3,195 16.5 1967 34,641 16,663 56,990 108,294 108,462 32 3,335 4 3,340 4.5 1968 36,616 17,851 62,307 116,774 117,046 34 3,481 4 3,489 4.4 1969 40,321 20,141 68,104 128,566 128,792 35 3,724 7 3,731 6.9 1970 38,163 23,503 85,719 147,385 147,606 36 4,146 11 4,152 11.3 1971 36,666 26,934 89,817 153,417 153,279 37 4,167 1 4,163 0.3 1972 43,130 31,311 95,635 170,076 169,467 38 4,434 6 4,418 6.1 1973 61,523 42,643 117,944 222,110 221,238 40 5.560 25 5,538 25.4 1974 75,420 53,475 150,311 279,206 279,112 41 6,756 22 6.753 21.9 1975 81,521 56.636 165,162 303,319 303,306 42 7,155 6 7,155 6.0 1976 92,460 68,186 185,870 346,516 345,632 43 8,019 12 7,999 11.8 1977 99,970 81,432 222,127 403,529 402,252 44 9,115 14 9.086 13.6 1978 119,638 97,658 270,930 488,226 484,604 45 10.797 18 10.717 17.9 1979 134,148 117.611 307,102 558,861 552,636 46 12.120 12 11.985 11.8 1980 152,852 139,936 365,721 658,509 653,115 47 14,023 16 13.908 16.0 1981 162,987 169,461 427,747 760,195 748,160 48 15,880 13 15.629 12.4 1982 156,839 176,360 486,803 820,002 807,072 49 16,786 6 16,521 5.7 1983 185,628 194,344 530,082 910,054 903.353 50 18.381 10 18,246 10.4 1984 175.190 218,050 580.172 973,412 961,961 51 19,245 5 19,019 4.2 1985 169,895 224,456 620,048 1,014.399 996,802 52 19,583 2 19,243 1.2 1986 178,140 258,644 658,584 1.095,368 1,072,931 53 20,679 6 20,255 5.3 1987 205,592 299,327 748,228 1.253,147 1,230,753 54 23,262 12 22,847 12.8 1988 250,384 373,326 883,267 1.506,977 1,482.207 55 27,4201 18 26,969 18.0 1989 266,379 453.258 1.056,341 1.775,978 1,752,574 55 32,174 17 31,750 17.7 S,ur,. CGonstancc N4 Wils.n. Thailand: A Handbook of Historical Statistics. G K. lall & Co.. Boston. Mass., 1983. National Economic and Social Devclopient Board, National Income of Thailand: New Series 1970-1987. Bangkok t19891. National Economic and Social Dcxeclopment Bnard, Nat1onal Income ofThailand / 1991). unpublished report. National Statistical Office. Statistical Yearbook of Thailand. various issues. 64 Thailand's Macroeconomic Miracle: Stable Adjustment anid Susta ined Growth Macroeconomic Data since 1954 Since the early 1950s Thailand has built up quite a good data system, by the stan- dards of developing countries. The National Income Accounts have been published since 1954. There are regular economywide surveys on employment, agriculture, in- dustry, household income, and expenditure. Information on interindustry transac- tions in the form of input-output tables has been available periodically since 1980. The data on public sector finance are generally available and are of good quality. Data on international trade, finance, and prices are comprehensive and are published regularly. Data on financial flows (flow-of-fund tables) are also compiled. The system by which Thailand compiles macroeconomic data was described in detail in an earlier study (World Bank 1979). The purpose of that report was to assess the reliability and the adaptability of the available macroeconomic data to meet the needs of planning and policy analysis. The World Bank report compre- hensively reviewed the methods employed for deriving national accounts and identified a number of areas for potential improvement in the method of data com- pilation. An important conclusion of the World Bank report, which has direct bear- ing on the present study, was that although the existing macroeconomic data system in Thailand is superior to those of most countries at a similar stage of de- velopment, the quality of the data could be substantially improved with a better data base and with refined estimation methods. The data used in the present study were assembled from a variety of sources, the main ones being the national accounts, the Bank of Thailand's Quarterly Bul- letin, the government's labor force survey, and the consolidated public sector ac- count. Most of these data are consistently and systematically prepared. The main issue that must be raised concerns data quality. There are a number of weaknesses in the existing data system that should be borne in mind, particularly those relating to the national accounts and the public sector accounts. THE NATIONAL ACCOUNTS. The national accounts are prepared annually by the Na- tional Accounts Division (NAD) of the NFSDB. The structure of the accounts con- forms closely to the old United Nations standard accounting framework (old SNA). The main problems with the current system are that a rather weak data base is used for preliminary estimation and the estimatiou procedures are oversimplified. These difficulties are most evident in the derivation of sectoral value added at con- stant prices, estimates of private consumption and of changes in stocks, and in the treatment of external transactions in the national accounts. Most serious are the re- sulting biases in the estimation of sectoral value added,which range from II to 36 percent in nominal terms. According to the World Bank report cited above, the cur- rent national accounts understate true GDP by about 5 to 1 0 percent. The recommendations outlined in the World Bank report were taken up in a major revision of national accounts data by the N \D) ol the NESDB. The revised se- ries for 1970-84, for example, made use of proper weighted averages and indica- tors and better interpolation techniques. A commrniodity balance approach was The Economic Setting: Structure and Pertbrinance oif the Thai Ecsnornv 65 adopted, deflation procedures were improved, and double deflation was used to ar- rive at the constant price GDP series. In another major change, several economic activities were reclassified to achieve greater consistency and to avoid double counting. We were fortunate to have access to these unpublished revised data. Briefly, the level of GDP at constant prices in the new series is 5.4 percent higher than the old one, whereas the level of GDP at current prices in the new level is 3.7 percent higher than the old series. This marginally brings up the average overall growth rate (for 1970-74) from 6.5 percent in the old series to 6.6 percent. The estimated growth rate for agriculture in the new series is lower than in the old series for 1971-76 but is higher for 1977-84. Exactly the reverse is seen for man- ufacturing. In summary, the new data suggest that the overall performance of the Thai economy under the Second and the Third plans was overestimated, while its perfor- mance under the Fourth and Fifth plans was underestimated. The most dramatic change is in the composition of sectoral output. According to the revised national ac- counts series, the share of agriculture fell below that of manufacturing in 1979, and since then the relative importance in production has shifted to manufacturing. It is important to take note of the main weaknesses in the old data series, since they may have an impact on one's interpretation of Thailand's macroeconomic performance and problems. First, the level oft' in manufacturing, construction, transportation. and services was understated (by II to 36 percent in nominal terms). Second, the level of GDP in banking was consistently overstated because interest payments to households were erroneously included in banking value add- ed. Third, estimates of private consumption and government consumption were understated. Fourth, estimates of net changes in stocks in all likelihood included errors in the underlying production and consumption data. And fifth, estimates of net factor income from abroad were incorrect because of the erroneous treatment of net wages and salaty payments. PUBLIC SECTOR DATA. The fiscal data used in this study were drawn from three main sources: the Comptroller-General Department I(GD), the Bank of Thailand's QQuarter/v Bulletin, and the national income accounts. The data are on the whole reliable since they are based mostly on actual transactions. The (GD assembles most of the expenditure data by public sector agencics. The Bank of Thailand col- lects raw data on public finance from various sources and tabulates them in a more refined form. The bank also compiles the consolidated public sector accounts. In preparing the national income accounts data, the NESDB draws heavily on the data compiled by the CGD and the Bank of Thailand in estimating the fiscal activities of the government. The main drawback of the public sector data is that expenditure items are not always assigned to the same categories and thus arc difficult to compare across time. Another problem arises in the accounting of nonbudgetary expenditure and in the estimation of public sector deficits. Theoretically. nonbudgetary expendi- tures refer to all expenditures not financed by hudgetary resources. The hasic data 66 Thailand s Macroeconomic Miracle: Stable Adjustment atid Sustained Growth comprise the consolidated expenditures of the public sector compiled by the Bank of Thailand. But the consolidated data cannot be directly compared with the value added concept of GDP because they include recurrent expenditures of state enter- prises that represent intermediate inputs. Another difference is that the consolidat- ed data pertain to the fiscal year (October to September) whereas the national accounts data are for the calendar year. It is notable that the national accounts approach and the consolidated expen- ditures approach give a different impression of the size of nonbudgetary expendi- ture and public sector deficits. The size of public sector deficits suggested by the national accounts, which do not include the recurrent expenditure of state enter- prises, is slightly higher than the figures implied by the consolidated data. This finding is somewhat surprising. The ratio of public deficits to GDP as reflected by the national income accounts also appears large. This may indicate that estimates of public expenditure in the national accounts are overstated. Previous studies of public sector resource management in Thailand have tended to rely on the consol- idated data since they provide a basis for assessing the overall size of resource con- sumption by the public sector. The figures. however, cannot be conveniently related to the existing body of the national income accounts data. fn this study, we rely primarily on the Bank of Thailand data. Chapter Four The Policy Setting: Role of the Public Sector The role of the public sector in Thailand is best understood by examining its insti- tutional framework and the the ways in which it intervenes in the economy. These interventions can be seen in the country's development plans, the fiscal system, sectoral and trade activities, financial market regulations, labor market regula- tions, the public provision of infrastructure, and the public enterprises. Although Thai economic policies have been far from laissez-faire, the prevail- ing view in Thai political circles is that the government should play only a limited role in the economy. This attitude is strikingly different from that generally pre- vailing in other developing countries. The origins of these attitudes in Thailand were discussed in chapter 2. By the 1980s, the role and the influence of the public sector-particularly of the core agencies controlling macroeconomic policy-had increased. Although the private sector remained the central source of economic dynamism. the govern- ment had become more active in economic affairs. This could be traced in part to a change in political leadership and the perceived macroeconomic difficulties the economy faced, but most of all to the changing intellectual climate in the Thai bu- reaucracy, which had begun to favor a more active economic role for the govern- ment. The government was also responding to demands by the educated general public for the government to take a more vigorous role in initiating and coordinat- ing economic development. Although the performance of public enterprises had been criticized by the 1959 World Bank report (see chapter 2) and by others, these enterprises remained more or less intact, and the main growth of public sector activity was not in these areas. Instead, the composition of public sector activity shifted away from direct involvement in industrial production toward the provision of public infrastructure and services. 67 68 Thailand s Macroeconomic Miracle: Stable Adjustment atid Sustained Grmwth The Institutional Framework The foundations of the current system of administration were laid a century ago, during the reign of King Chulalongkorn. The system was modeled along British lines, but it modified the traditional functions of court into a hierarchical system of government agencies, with administrative power assumed principally by the central government. Senior civil service posts were held exclusively by members of the aristocracy. This structure was also undoubtedly influenced by the central- ized political structure prevailing at that time. Many elements of this administra- tive centralization have survived, surprisingly without radical change, to this day. As noted in chapter 2, an equally important legacy of the Thai bureaucratic system established in the nineteenth century was the attitude that the civil service owes its loy- alty primarily to the king, representing the Thai state, rather than to the current govem- ment. This attitude is deeply embedded and has been a strong force behind policy stability-some would say inertia-even in the presence of radical political change. The public sector in Thailand consists of the central government, the local governments, and the state enterprises. Central Government By far the largest public body is the central government. It is made up of twelve ministries, the Office of the Prime Minister, the Office of University Affairs, and seven independent government agencies, including the Parliament and the Bureau of Crown Property. Apart from supervising the work of departments, offices, and publicly funded agencies directly under them, the central government supervises the work of local governments and state enterprises. The finance of central government is separated into budgetary and nonbudget- ary transactions. The national budget requires the approval of the Parliament. Its expenditure is supported by incomes from six main sources: tax revenue: contri- butions from state enterprises; fines, fees, and proceeds from the sale of goods; domestic borrowing; the issue of new currency; and the use of treasury cash bal- ances. The first three items are budgetary revenues whereas the next three are con- cerned with the financing of budgetary deficits. Note the absence of foreign borrowing as a source of budgetary finance. Receipts from foreign loans do not ap- pear in the budget, and their control is treated separately. Nonbudgetary transactions refer to fiscal transactions outside the annual bud- get. They take two basic forms: expenditure financed by external grants and loans, and expenditure financed by advances from the state treasury deposits. The latter is a special case and requires parliamentary approval, in the form of a special act. Expenditure financed by foreign borrowings is nonbudgetary and therefiore does not require this approval. Three regulations regarding central government expenditure are notable. First, in any fiscal year, the magnitude of any budgetary deficit may not exceed 25 The Policy Setting: Role ot tile Puhlic Sector 69 percent of the expected revenue. Second, direct foreign horrowing by the Ministry of Finance in any fiscal year must be within 10 percent of the expenditure budget. And third. foreign loans for state enterprises that are guaranteed by the Ministry of Finance in any year must not exceed 10 percent of the expenditure budget. Local Governments Local governments are the administrative ann of the central government in the prov- inces. Their administration is the responsibility of the Ministry of Interior, through the Local Administration Department. At present, local governments consist of 126 municipalities, 795 sanitary districts. 72 Changwat (provincial) administrative orga- nizations (CAOs), the Bangkok Metropolitan Administration, and the Pattaya City Administration. The main administrative power rests with the CAOs, whose heads, except in the cities of Bangkok and Pattaya, are appointed rather than elected. Pro- vincial governors are civil servants appointed by the Ministry of Interior. With this direct line of command, the central government is able to control local government administration. Clearly, the Thai system of government is highly centralized. Local governments have relatively small weight in public sector finance- less than 5 percent in terms of expenditure. Their main source of income is the rev- enue from local taxes, revenue from taxes shared with the central government, own income from property, fines, fees and permits, contributions from the central government, and domestic borrowings. Foreign borrowing by local governments is legally possible but must be organized on their behalf by the Ministry of Interior. In practice, no such horrowing has ever taken place. Key Actors and the Existing Policy Framework In Thailand's system of government. ministers are formally responsible for mac- roeconomic policy decisions. In reality, key decisions tend to be made by civil ser- vants in four core agencies: the NESDB, which is the government's main economic planning agency: the Fiscal Policy Office of the Ministry of Finance: the Bank of Thailand: and the Bureau of the Budget (see figure 4.1, in which the core agencies are shaded). Policy decisions at the ministerial level. made either by an individual minister or by ministers acting collectively, usually as a cabinet, rely on informa- tion and analyses made available to them by the departments concerned and the core agencies. The directors of these core agencies ordinarily sit in the Council of Economic Ministers. This council is not a constitutional or even legally mandated body, but a subcommittee of the Council of Ministers (the cabinet). It does not have permanent status, nor is its membership constant. Before they are discussed at the ministerial level, policy options are formulat- ed through coordination and consultation between departments and experts from the core agencies. In recent years. the role and the influence of the core agencies have increased significantly as the government has become increasingly reliant on 71) Thailand r Macrocecyonomic Miracle: Stable Adjatnat aad Suitained Growuth Figure 4.1. Policymaking Structure Cabinet / National Nati _ Committee on Debt Policy State Enterprises ommittee vOffice of Ministry Ministry Prime Ministt f nanc f Interior National Economic Local 1 and Social Fiscal Poicy administration Development ef_ ce Board ~~~~~~~~~department Buareau of : . Local thie Budget ': | government St te Bank of State-; [enterpri Thailand eterpris State enterprise I -S--- ----------------- --------------- state enterprises ---------------------------------- Source: Authors' compilation. them for opinions and analyses. The heads of these core agencies are the central actors in formulating Thailand's macroeconomic policies. The Development Plans Thailand's development plans are best described in relation to the economic con- ditions existing at the time they were formulated. The documents setting out the development plans are significant, even though they are not in themselves particu- larly important as instruments of actual policy. Because circumstances change from those anticipated when the plans are drawn up, departures from the plans are both inevitable and desirable. Nevertheless, they do show the way in which policy imper- atives were perceived at the time the plans were formulated. The Polity Setting: Role of the Public Sector 71 During the 1950s the government became considerably more active in eco- nomic affairs. This was in part the result of a change in political leadership and of the heavy influx of U.S. economic aid following the Second World War. In 1950 the National Economic Council was established to provide the government with staff services in economic studies and the analysis of fiscal and monetary prob- lems. In 1954 the government introduced national accounting and passed the first Industrial Promotion Act, which marked the beginning of active government par- ticipation in industry. One of the recommendations of the 1957 World Bank mis- sion, discussed in Chapter 2, was that a permanent planning agency be set up to undertake economic studies and to prepare national development plans. Planning for economic development was formalized in 1959 when the Na- tional Economic Development Board (NEDB)-now known as the National Eco- nomic and Social Development Board-was established. Since then, seven national development plans have been drawn up. The underlying philosophy of economic planning in Thailand is a commitment to a market economy. Planning has been directed mainly toward securing a smooth functioning of markets with a minimum of direct government intervention or controls. The First Plan (1961-66) was essentially a public expenditure program that accounted to a moderate degree for the source of revenue and the outlays of the government. The main objective of the plan was to encourage economic growth in the private sector through the provision of basic infrastructure. The principal thrust of government involvement was therefore concentrated on expanding infrastruc- ture facilities in transport, communications, power, social and public services, and agriculture. Economic growth during the First Plan period was both rapid and broadly based. The annual GDP growth rate averaged 8.1 percent. In sectoral terms, infra- structure and construction recorded the highest average annual growth rates, at 22.3 and 17.8 percent, respectively. Agricultural output also expanded rapidly, at an average rate of 6.2 percent. These rates were impressive by international stan- dards. Much of this achievement was owed to favorable internal and external trends at the time. New attitudes of the government regarding expanded public ex- penditure, favorable world demand for Thai products, and U.S. military spending in the country did much to promote the success of the plan. During this period, capital expenditure increased at an average annual rate of 18 percent while prices remained fairly stable. Despite an unfavorable trade balance, the overall balance of payments was in surplus because of the massive inflows of foreign capital in the forms of loans and grants. Despite an accelerated population growth rate of 3.5 percent per year, per capita income rose on average by 4.8 percent per year. The Second Plan (1967-71) was still largely a public expenditure program. It incorporated manpower planning but made no attempt to direct resource alloca- tion among sectors. The Second Plan continued the First Plan's task of building infrastructure, particularly in those areas considered conducive to development. The pattern of public expenditure under this plan revealed the government's in- creased emphasis on the slower growth areas, especially the rural sector. 72 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth The growth of real GDP during the Second Plan was less impressive than it had been during the First. The average growth rate of real GDP was 7.5 percent. Output continued to expand rapidly in infrastructure and services while industry main- tained a steady growth rate, averaging 10.1 percent per year. The slowdown in overall economic performance during this period was partly related to the slow- down in foreign investment and U.S. military spending. From 1969 to 1971 the balance of payments moved into a deficit position ior the first time ever, also as a result of the decline in direct foreign investment. Some commentators also pointed to the "disappointing" performance of agri- culture, which grew at an average annual rate of "only" 4.5 percent. The decline in agricultural growth was due in part to the droughts in 1967 and 1968 and in part to fluctuations in the world price of major export commodities. This was also a pe- riod when the expansion of the agricultural land base virtually ceased (Ammar, Suthad, and Direk 1993) and agriculture was shedding labor to the rest of the economy. It has continued to do so-at an increasing rate-ever since. The 4.5 percent growth of agricultural output per year was therefore a more than satisfac- tory outcome and was in any case high by international standards. The Third Plan (1972-76) reflected a moderate shift of emphasis in develop- ment thinking. Although still aiming for higher growth, the Third Plan set specific priorities for reducing the growing disparities hetween urban and rural areas, and between sectors. This shift reflected a growing awareness among NESDB planners of the increasing problems of regional disparity and poverty in the rural areas. The emphasis of the plan was not only on improving public infrastructure and main- taining economic stability, but also on achieving a more equitable distribution of income and social services. Growth during the Third Plan period was disappointing, however, because of the first OPEC oil crisis of 1973-74, the resultant world economic slump, and the continued decline in U.S. military spending. The GDP growth rate averaged 6.2 percent. Again, this aggregate slowdown was attributed mainly to the performance of agriculture, which grew by "only" 3.9 percent per year. The increase in capital investment by the government continued the momentum of expansion in the coun- try's infrastructure. In spite of the growth of industrial production (about 8.6 per- cent per year), it was evident that industry was concentrated in Bangkok, was capital-intensive, and thus was not generating sufficient employment. Neverthe- less, migration from rural areas to the capital continued to be rapid. It was widely thought that the benefits of industrial expansion were not reaching the majority of the populace. Moreover, it was recognized that the impressive aggregate growth since the First Plan had been achieved at the expense of a rapid deterioration in Thailand's land, forest, water, and marine resources. The Fourth Plan (1977-81) was intended to address these problems by re- structuring the national economy. Its immediate objective was to revitalize the economy from the effects of world economic recession and to implement desired structural adjustments. The plan's objectives for the next five years were formulat- ed in terms of eight development goals: the development and conservation of eco- The Policvy Setting: Role of the Public Sector 73 nomic resources and environment; diversification and increased efficiency of production in rural areas; the development of industry; the promotion of exports, imports, and tourism; the development of principal cities and the improvement of Bangkok: and the dispersion of basic services, the dispersion of social services, and social development. The plan spelled out in detail the methodologies, targets, and funding that would be used to reach each of these objectives. At the heart of the plan was an attempt to switch from a growth orientation toward greater social awareness and an emphasis on economic readjustment. The performance of the economy during the Fourth Plan period was again af- fected by changing world economic conditions, particularly the rising price of oil following the second OPEC oil price shock (1979-80), high international interest rates. and declining demand and prices for Thailand's commodity exports. In spite of the unfavorable external conditions, the economy did expand satisfactorily. The growth of real output averaged 7.1 percent per year over the plan period. This achievement was partly a result of the government's attempts to maintain the growth momentum by expanding public investment despite a drastic deterioration in domes- tic savings. The enlarged domestic investment-saving gap was reflected in a serious trade imbalance, and as income from other sources failed to compensate for this def- icit, Thailand's balance of payments remained in deficit throughout the Fourth Plan period. As a consequence, foreign indebtedness increased substantially. To deal with these problems, the Fifth Plan (1982-86) continued to give high priority to economic restructuring, to maintaining economic and financial stabili- ty, and to improving the quality of life of the rural poor. This was meant to be achieved by placing more emphasis on the quality of growth rather than the rate of growth. In the process, a reform of the public development administration sys- tem was to be carried out to facilitate rapid economic development. The growth momentum was thought to lie with industrial development, whose share of output was projected to reach that of agriculture by the end of the planning period. The main thrust of industrial policy was the promotion of export-oriented industries and the dispersion of manufacturing industries to provincial areas. Such a disper- sion was seen as a way of achieving a balance of growth between urban and rural areas. The ambitious Eastern Seaboard Development Scheme, to be located close to the seaside resort of Pattaya, was to have been a central component of this pol- icy. In later years this scheme had to be curtailed substantially. The years 1982-86 coincided with slowed growth as a result of the world re- cession of the early 1980s. Largely as a result of the international oil price increas- es of the 1970s and early 1980s, Thailand's external terms of trade deteriorated from an index of 100 in 1973 to 51 in 1985 and 56 in 1987. Although Thailand had avoided the economic collapse that these external events had produced in other de- veloping countries-including its near-neighbor the Philippines-by 1985 it was experiencing serious macroeconomic problems. To begin with, the current account deficit in the balance of payments was persist- ing at the unsustainable level of 5 percent of GDP. The investment-savings gap had reached a similar magnitude. This represented mainly a decline in savings as a propor- 74 Thuiland \s M4a croecotnomic Miracle: Stable Adjustment atnd Su.qained Growth tion of GDP from 20 to 22 percent in the late 1970s to 16 to 17 percent in 1985. In ad- dition, foreign exchange reserves, as a proportion of GDP, had fallen from 12 percent in 1970 to 3 percent in 1985. This required a $500 million standby loan from the In- ternational Monetary Fund (IMF) in mid- 1985. Its renewal was negotiated a year later. At the same time, the external debt had risen to $ 16 billion, which was equiv- alent to 40 percent of GDP and 146 percent of exports. About $12 billion of this amount was long-term debt, of which $8 billion was public or publicly guaranteed. An additional $4 billion of short-term debt was held mainly in the private sector. The debt-service ratio in 1985 was about 26 percent, up from 17 percent in 1980. To make matters worse, the government's budget deficit had remained at more than 5 percent of GDP over the previous five years. Total public expenditure for central and local governments and state enterprises was about 40 percent of GDP. The central government just managed to finance its current expenditures from its revenues. Virtually all capital expenditures were financed by borrowing. General government savings had fallen from 3.7 percent of GOP (average of 1970 to 1977) to less than I percent in 1985. The overall rate of growth of GDP in real terms was lower in the 1970s than the 1960s, and lower still in the 1980s. Growth in the years 1985 and 1986 was the lowest of any two consecutive years since the 1950s; but this was a decline from a long-term real rate of growth of almost 7 percent to -only" 5 percent. The Sixth Plan (1987-91) was drafted in response to the above problems. Its objectives were to * Promote economic growth to at least 5 percent per year in real terms. v Improve the administrative structure of the government and review its role. (The private sector was to play a greater role and thereby reduce the burden of the government.) * Increase the mobilization of domestic saving from both private and public sectors from the target of 1 8.2 percent under the previous plan to 23.7 percent. * Continue the privatization process and improve the administrative efficien- cy of the state enterprises. (The proportion of their foreign borrowing was also to be reduced.) * Use fiscal and monetary measures to support economic growth and reduce the deficit in the trade and current accounts. (An important fiscal measure was to be the restructuring of the taxation system, which was to increase government revenues and to attract foreign investment.) From 1986 onward Thai macroeconomic policy adjusted sharply to the imbal- ances described above. The adjustments included a large fiscal contraction. The fiscal deficit was transformed into a surplus equivalent to 1.3 percent of GDP in fiscal 1988 and to 4.9 percent in 1990. Cuts in public investment expenditure were a major com- ponent of this adjustment. Fixed capital formation in the public sector declined by three percentage points of GDP from fiscal 1985 to 1988 (to 5.8 percent of GDP). Si- multaneously, Thailand experienced an export boom, concentrated in manufactures. The Policy Setting: Role of the Pub!lic Sector 75 The boom appears to have been a consequence of two mutually reinforcing events, neither of which can reasonably be attributed to deliberate acts of policy on the part of the Thai government: a 30 percent depreciation of Thailand's real effective exchange rate from 1986 to 1990, resulting from the baht being pegged to a depreciating U.S. dollar; and the international relocation of light manufactur- ing industries from elsewhere in East Asia-especially Taiwan (China), Hong Kong, Korea, and Singapore, where labor costs were rising rapidly. The magnitude of the boom was as much a surprise to the Thai economic planners as to anyone else. But the reduction in expenditure on basic infrastructure-including roads, ports, and telecommunications-threatened the medium-term sustainability of the boom because these facilities were becoming badly congested. The Fiscal System The fiscal system can be examined from two perspectives: its methods of raising revenue and handling expenditures. Raising Public Revenue Responsibility for tax policy rests with the Ministry of Finance, specifically the Fiscal Policy Office. Tax revenue comes from both central and local taxes. In 1990 the ratio of total tax revenue (central and local) to GDP was 20 percent, which is low by inter- national standards. The bulk of tax revenue comes from central government taxes. Between 1970 and 1990 government revenue in relation to national income increased from 13 to 20 percent of GDP (table 4.1). The composition of tax reve- nues has changed markedly since the 1960s. Indirect taxes remain by far the most important component, while international trade taxes (import and export taxes) have declined in importance as a share of total taxes but not as a share of GDP. This change has been balanced by increases in the relative importance of income-based and consumption taxes. There is less emphasis on international trade taxes be- cause Thailand has virtually eliminated export taxes and has reduced its average import taxes. An expansion in the volume of imports in relation to GDP partly off- set this effect. Because of this shift, the present tax system has become more de- pendent on indirect domestic taxes. It is significant that, despite the increasing importance of income-based taxes (personal and corporate), direct tax revenues remain of limited importance by international standards. The main features of the current tax system may be summarized as follows: * The system is dominated by indirect taxes with a low elasticity to GDP and a small revenue base. As a result, the average tax rate tends to decrease au- tomatically as GDP increases. 76 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 4.1 Composition of Government Revenue by Major Sources, 1970-90 (percent) Source 1970 1975 1980 1985 1990 Total revenue 100.0 100.0 100.0 100.0 100.0 Incometaxes 11.7 16.1 17.7 21.9 24.8 Indirect taxes 79.1 72.4 74.9 68.3 68.9 Other government revenue 9.2 11.5 7.4 9.8 6.3 Income taxes 11.7 16.1 17.7 21.9 24.8 Personal 6.9 6.9 7.6 12.4 10.1 Corporation 4.8 9.2 10.1 9.6 14.2 Petroleum 0 0 0 0 0.4 Indirect taxes 79.1 72.4 74.9 68.3 68.9 Import duties 28.8 21.5 20.4 19.1 22.6 Export duties 4.5 3.6 3.5 0.7 0.0 Business taxes 19.7 20.3 19.2 18.4 21.9 Sales taxes 16.2 18.7 22.3 23.4 17.5 Other taxes 9.9 8.2 9.5 6.7 6.8 Total revenue/GDP 12.8 13.0 14.5 15.8 20.1 Source: Bank of Thailand, Monthly Bulletin, various issues: Customs Department; and Controller- General's Department. • Personal income tax rates are progressive and have an income elasticity greater than unity. The progressive tax schedule, however, has limited ef- fects on the top income earners because many types of income-including income from bequests and income from interest on bank deposits-are ex- empt from taxation. As a result, the ratio of tax to assessable income is not steeply progressive. * Corporate income tax has limited importance, for close to half of all cor- porations declare losses for tax purposes. In 1984 less than 1 percent of all corporations paid 77 percent of corporate taxes. Without substantial im- provements in collection, corporate income tax will remain an unreliable revenue measure. * Domestic consumption taxes (business and excise taxes) have a regressive structure, and the tax base is small. At present, state enterprises do not pay business tax, and the excise tax covers only nine commodities. This feature makes revenue mobilization costly to the general public as tax rates need to be adjusted upward frequently for revenue purposes. The tax burden is therefore passed on to all consumers regardless of their income positions. * The relative importance of international trade taxes has declined because the government continues to liberalize trade through tariff reforms and the The Policv Setriig: Role ofthe Public Sector 77 abolition of export taxes. Still, tariff rates vary greatly, ranging from 5 to 60 percent ad valorem. Direct government transfers play a small role in Thailand's fiscal structure, in part because the country does not have a welfare system. Furthermore, most of the transfers occur among public sector agencies and not between the government and households. In January 1992 the government implemented a new value added tax (VAT) system. The new tax was designed to overcome some of the problems identified above. The cascading effect of the existing business tax system-the fact that the tax rate effectively increases along the chain of production-was stressed as a reason for implementing the VAT. The rate of the VAT was set at 7 percent, but many industries were exempted, including all of agriculture and any industries producing inputs for direct use in agriculture, such as fertilizer, animal feeds, and pesticides. Businesses with a total annual revenue of less than 600,000 baht (approximately $24,000) were also exempt. Exporters were entitled to a refund of the VAT on proof that the goods had been exported. Despite the VAT reform, the existing tax system hinders effective revenue mo- bilization. An important policy issue at present is how to improve the existing tax system so that revenue can be raised more effectively without jeopardizing private incentives and without creating further inequality in the distribution of income. Public Expenditure There has been a steady shift over time in the allocation of public expenditure be- tween current and capital expenditure (table 4.2). Thailand's badly congested public infrastructure in the early 1990s is partly a reflection of this trend. From 1985 to 1990 public expenditure as a proportion of GDP declined by a full five percentage points-from 20 to 15 percent. Thus in the late 1980s the government experienced record budgetary surpluses, as is apparent from a comparison of tables 4.1 and 4.2. Sectoral and Market Interventions Until the late 1960s, the role of the public sector was confined largely to tax col- lection, public provision of a limited range of public services, and the commercial operation of the public enterprises. As the economy expanded. the authorities came to see the price system as a distorted and unreliable guide to resource allo- cation. As a result, direct intervention by the public sector increased. Some of the measures taken were meant to modify the pattern of resource utilization while some purposely added distortions in order to effect a change in the distribution of economic benefits. These interventions have taken place in five important areas of 78 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 4.2 Composition of Government Expenditure, 1970-90 (percent) Expenditure 1970 1975 1980 1985 1990 Total expenditure 100.0 100.0 100.0 100.0 100.0 Economic classification Current 68.4 77.6 78.0 83.5 81.8 Capital 31.6 22.4 22.0 16.5 18.2 Majorfunctional classification Economic services 29.1 25.3 18.8 15.3 17.5 Social services 26.3 29.1 29.3 29.5 31.2 Defense 17.5 17.9 20.2 21.6 18.9 General administration And services 14.1 14.7 14.6 12.9 14.1 Unallocatable 12.9 13.0 17.0 20.8 18.3 Total expenditure/GDP 17.1 14.9 18.4 19.7 14.8 Source: Bank of Thailand, Monthly Bulletin, various issues. the economy: agriculture, industry, the energy sector (through petroleum pricing policy), trade, exchange rate policy, and regulation of financial and labor markets. Interventions in Agriculture The main impetus behind Thailand's economic growth in the 1960s and the 1970s was the growth of agriculture. By the late 1970s, however, the land frontier was exhausted and agriculture's contribution to production had declined (Martin and Warr 1990, 1994). Nonetheless, agriculture remains the largest source of employ- ment and the largest provider of income for the majority of the population. In the early 1990s, more than half of all Thai households still drew their principal in- comes from agriculture. Although agricultural growth has come mainly from private initiative, gov- ernment intervention, particularly in the pricing system, has had a considerable impact (Ammar and Suthad 1989). The most important intervention, and the one with the longest history, had been the taxation of rice exports, which in its modem form began immediately after the Second World War and was suspended in 1986. Over these decades, the heavy export tax on rice had depressed rural incomes by reducing the farmgate prices of paddy and the rural wage rate (Chirmsak 1984). It The, Policy, SeXtui,: Role of/tlce Public Sect or 79 also impeded technological change by altering the price-cost ratio in the rice sec- tor. A high export tax (15 percent of the f.o.b. price) is still levied on rubber (Am- mar and others 1993). The compulsory rice reserve scheme that began in 1973 (a period of rice shortage) and ended in 1982 had a similar effect. Under the scheme, exporters were required to sell to the government a proportion of their rice (fixed in relation to the amount of rice exported) at a price set lower than the domestic price. This system was designed to provide the government with a cheap supply of rice for resale to the general public. The policy was thus similar to an ad valorem export tax and it further depressed farmgate prices for paddy. The government has also imposed periodic export quotas on agricultural products. In the past, both rice and maize were occasionally subjected to such con- trols. At present, they apply only to cassava. since the government has agreed to limit exports of this product to the European Economic Commission. The Ministry of Commerce is responsible for the allocation of export quotas, and its allocations from year to year are often politically motivated. Apart from rice, maize, cassava, and rubber, agricultural products subject to government regulation have included swine, castor oil seeds, and tobacco. In most of these cases, government regulations have introduced monopolistic elements into the markets that have reduced economic efficiency. In the swine market, for example, a condition for setting up a private slaughterhouse is that property rights on land and buildings be transferred to the local government, in accordance with the Animal Slaughtering and Meat Sale Control Act of 1960. In this way, public slaughterhouses. which are managed by local governments, gain a monopoly on the market. Interventions in IndustrY: Protection and Export Pr(mnotion The structure of Thailand's protection policy was discussed in chapter 3 under "Product Markets." Aside from this, the main feature of government intervention in industry has been the promotion of private investment, administered through the Board of Investment (BOI), which was established in 1959. The BO] uses a combi- nation of various investment promotion schemes, tariff policies, tax regimes, and trade and price controls to direct the pattern of private investment. During the 1960s and the early 1970s, industrial policies stronglv favored import substitution. Import tariffs were raised significantly to protect local industries, and the strongest incentives were directed at the production of final products based on imported in- termediate and capital goods. With the passage of the Investment Promotion Act of 1972, the emphasis of industrial promotion policy then supposedly shifte(d away from import substitu- tion toward exports. Export promotion has since become the central theme of ef- forts to promote private investment in industry. At least, that has been the expressed intention. The 1977 Investment Promotion Act modified the 1972 act and introduced the promotion of trading companies 80 Thailand s Ma(roeco(n(mic Miracle: Stable Adjustment antid Sustained Growth Prospective investors must first apply for promotion privileges and obtain a BOI promotion certificate. A list of industries eligible for promotion privileges is drawn up by the BOI using the national development plan as a broad framework. The incentives offered typically include tax and tariff exemptions, a guarantee of government protection from nationalization and from direct competition by state enterprises, and guarantees of rights of profit and of capital repatriation. The range of incentives differs between the industries, depending on priority rankings in the promotion policy. The promotion policy as practiced has been criticized for its fre- quent changes, the BOI's use of its discretionary powers in granting promotional privileges, and the extent of the incentives offered, which often differ among firms within the same industry. In addition to the BOI, the Ministries of Industry, Commerce, and Finance, as well as the Bank of Thailand, formulate and administer policies that directly affect industrial development and are ostensibly aimed at export promotion. The Cus- toms Department provides a scheme for the exemption from or refund of import duties paid on inputs used in the production of goods for export. Because of ad- ministrative obstacles to and delays in the actual payment of drawbacks, few ex- porters actually claim the refunds to which they are entitled. The exemption scheme is also said to be administered in a discretionary manner. The Fiscal Policy Office of the Ministry of Finance operates a comprehensive tax refund system, called tax rebates, for all taxes incurred in the production of goods subsequently exported. These taxes are estimated on a case-by-case basis using input-output methods and information on the cost structure of the exporter. The rebate is paid in the form of tax credit certificates, which can he used against other tax liabilities. The Bank of Thailand provides export-oriented firms with a rediscount facility at subsidized interest rates. The Ministry of Commerce provides technical assistance through its Export Service Center. The Ministry of Industry controls the establishment and expansion of facto- ries and production plants and the use of local contents in production. At present, twenty-three categories of industry are subjected to factory control, and four in- dustries-motor vehicle assembly, motorcycle production, electric wire and cable, and steel production-are subject to minimum local content requirements. Like many other countries, both developed and less developed, Thailand has favored supporting infant industries thought to be capable of becoming successful exporters after a short period of protection. An important question to consider is the extent to which this policy contributed to Thailand's record of economic growth, which was led by manufactured exports. We explore this issue by examining the statistical relationship between the trade performance of industries and the government's interventions to promote them. Our index of trade performance is the net export performance ratio of each industry, as described in chapter 3 and summarized in table 3.8. We are especially interested in the behavior of this index over time and its correlation with measures of industrial policy interventions. The analysis focuses on five instruments of in- tervention that can be quantified from available data: industry protection, the allo- The Polith, Sertting: Role of the Public Sector 81 cation of subsidized loans through the Industrial Finance Corporation of Thailand (IFCT), the promotion of industries through the Board ol Investment, the allocation of tax exemptions by the Customs Department, and the allocation of tax rebates by the Fiscal Policy Office of the Ministry of Finance. The effectiveness of these instruments is assessed on the basis of the export performance of Thailand's leading industries. Sixteen industries were selected for the analysis. To allow for the fact that these industries vary in size, we divided each industry's allocation of the instru- ment shown by its share of total value added, summed across all sixteen industries (table 4.3). Thus, an industry for which the resulting ratio exceeds unity receives a higher share of the total export promotion incentive concerned than its share of total value added, and so forth. Table 4.4 shows the correlation coefficients between the measures of export performance and each of the above five measures of industrial policy interven- tions. Table 4.5 shows the correlation between the changes over time in each in- dustry's trade performance and the changes in each of the five policy instruments. The results show that export performance is negatively related to all five measures. Moreover, the change over time in net export performance is negatively related to the change in all five instruments. Industries whose export performance worsened over time received increasing levels of support. but those whose performance im- proved tended not to receive such support. In the case of import protection, these results are hardly a surprise. Protection is explicitly an incentive to import-competing production and a disincentive to ex- porting. More surprising is the even higher negative correlation in the case of the BOI instruments and the tax rebates as well. Far from being instruments of export promotion, the IFCT loans, Customs Department tax drawbacks, Fiscal Policy Of- fice tax rebates, and especially the BOI investment promotion schemes are similar to industry protection in their allocation across industries. At the industry level, it is clearly the poor performers that are promoted hy these measures. The political economy behind these results is presumably that industries that are well organized for lobbying purposes put proportionately more resources into the behavior that secures bureaucratic support than industries that are less well or- ganized. But the former are not necessarily performing well in economic terms- our results suggest the reverse. Thus policy measures intended to promote exports are captured by the system of rent-seeking and in fact support roughly the same poorly performing industries as are favored by the system of protection. Petroleum Pricing Policy Thai economic growth in recent decades has been associated with industrializa- tion, farm mechanization, urbanization, and the greatly increased use of transport services. Consequently, energy demand has grown rapidly: from 1974 to 1990 the rate was in excess of 1 2 percent per year. At the time of the first oil shock, imported petroleum products accounted for 86 percent of total commercial energy used, but Table 4.3 Instruments of Policy by Industry, 196-90 __ _ _ Effective rates of Share of total IFCT Share of total BOI Share of total tax Share of total tax protection (percent) approved loans promoted projects drawbackfunds rebate futids Industtr 1974 1984 1987 1960-79 1980-85 1986-90 1983-85 1987-89 1986 1987 1989 /986 1987 1989 Agriculture 5 9 13 0.94 1.14 1.49 0.86 0.70 0.06 0.04 0.02 0.07 0.06 0.05 Tobacco 2.067 7 10 0.18 0.01 0.00 - - - - - - - Textiles 46 23 25 0.44 0.31 0.24 0.18 0.22 2.32 2.46 2.33 2.06 2.03 1.93 Wood products -42 58 65 0.46 2.06 1.37 0.71 2.13 0.16 0.20 0.14 0.42 0.59 0.6 Furniture 183 1 2 0.36 0.91 1.10 - - - - - - - - Paper 10 29 36 4.06 2.34 2.65 0.52 2.80 1.88 1.12 0.84 0.99 1.20 1.23 Printing -2 -2 2 0.72 0.23 0.56 0.38 2.08 Rubber 48 7 14 0.32 0.40 0.94 1.52 1.74 8.22 8.43 7.39 1.84 1.77 2.07 Chemicals 52 35 45 3.45 2.87 2.24 1.93 1.62 0.49 0.37 0.61 0.76 0.65 0.72 Petroleum -j( 5 -6 0.26 0.07 0.56 - - - - - - - _ Nonmetallic X8 2( 24 3.54 4.46 1.47 3.38 0.98 0.69 0.95 0.67 0.58 0.64 0.72 Metal products 29 13 20 1.66 2.61 4.47 2.48 1.13 1.01 0.64 0.95 1.34 0.94 1.14 Machinery 5 23 25 0.00 1.94 1.87 0.49 1.54 0.68 0.30 1.77 1.63 2.52 6.11 Electrical machinery 217 21 26 2.95 0.25 1.01 3.00 2.59 - - - - - - Transport 151 40 153 0.33 0.20 0.08 - - 0.08 0.18 0.12 0.10 0.14 0.35 Other manufactufing 498 18 18 0.97 1.15 0.15 0.99 1.31 0.21 0.23 0.28 0.92 0.98 0.54 Note: All industrv shares are expressed in relation to the industrv s share oJ'total value added. Source: IFCT approved loans data: frorm the Bank of Thailand. Datafor value added is takenfrom National Income of Thailand, Office of the National Economic and Soc ial Development Board, Bangkok. Effective rates of protection are drawnfrom the data underlving tables 3.2 and 3.3. Table 4.4 Correlation Coefficients across Industries: Trade Performance and Industrial Policy, 1970-89 Tax Tax Effective rate of protection IFCT loan allocation BO! projects drawbacks rebates Period 1974 1984 1987 1960-69 1980-85 1986-90 1983-85 1987-89 1986-89 1986-89 197(-74 -0.06 -0.02 -0.08 -0.16 -0.03 -0.02 -0.24 -0.39 -0.16 -0.46 1975-79 -0.07 -0.11 -0.14 -0.18 -4.09 {).08 -0.26 -0.47 -0.12 -0.39 1980-84 -0.06 -0.16 -0.15 -0.16 -0.11 -0.11 -0.23 -0.52 -0.11 -0.40 1985-89 -0.04 -0.14 -0.15 -0.17 -0.15 -0.25 -0.28 -0.52 -0.03 -0.35 Source: Calculated from data on NEPR by industry in table 3.8 and levels of policy instruments by industry in table 4.3. > Table 4.5 Correlation Coefficients across Industries: Changes in Trade Performance and Changes in Industrial Policy Change in Change in BOI NEPR over Change in EPR Change in IFCTIoan allocation promoted projects the period: 1974-8 1974-84 1984-87 1960- 79 to 1986-(0 196(-79 to 1980-85 1980-85 to 1986-90 1983-85 to 1987-89 1985-89/ 1980-84 -0.09 -0.09 0.01 -0.34 -0.13 -0.27 -0.19 1980-84/ 1975-79 -0.10 -0.10 0.07 -0.46 -0.27 -0.26 -0.43 1975-79/ 1970-74 0.06 -0.06 0.17 -0.38 -0.40 -).03 0.20 Source: Calculated from data in tables 3.8 and 4.3. 84 Thailands Macroeconomic Miracle: Stable Adjustment aInd Sustained Growth by 1985 this had fallen to 44 percent. The changes in energy use have been a con- sequence of structural changes in the Thai economy, international price changes, and government pricing policy. Thailand remains the fifth largest net importer of petroleum products among developing countries, after Brazil, India, the Republic of Korea, and the Philippines. The government taxes petroleum products at widely different rates. Gasoline is heavily taxed, diesel oil and liquefied petroleum gas (LPG) are moderately taxed, and fuel oil and kerosene are scarcely taxed at all. l ocally refined products, by far the most important, are taxed through a combination of excise taxes, municipal taxes, and a variable levy known as the "oil fund." There is no import tax on crude oil. Imported refined products are taxed by means of import duties, business and municipal taxes, and the oil fund levy. Changes in the oil fund levy make it possi- ble for the government to regulate retail prices of petroleum products independent- ly of ex-refinery and import prices. In 1973 the ex-refinery rates of tax were about 100 percent on gasoline, slight- ly lower on LPG and about 30 percent on fuel oil. These tax rates were reduced af- ter the first oil shock in an effort to keep retail petroleum prices down. The political importance of petroleum product pricing is shown by the subsequent events: the increase in the tax on petroleum products in 1980. which had been allowed to fall through the 1970s, led directly to the fall of the Kr iangsak government (see chap- ter 2). The new government, led by General Prem Tinsulanonda, promptly reduced the taxes on petroleum products to about two-thirds of their average levels under the previous government. Petroleum taxes have been an important source of revenue for the Thai gov- ernment: they rose from about 7 percent of total tax revenue in 1974 to 13 percent in 1980, and since 1981 have remained at about I 0 percent (Praipol 1993). Excise taxes account for about 85 percent of this revenue and import taxes for most of the remainder. The oil fund revenues have been used in part to subsidize the use of fuel oil by the Electricity Generating Authority. The net contribution to central govern- ment revenue has been small. These subsidies became especially important after the second oil price shock. From time to time, subsidies from the oil fund have been granted to all petroleum products, except gasoline. In 1990 the retail price of gasoline was about 100 percent above import prices, the price of LPG was 30 per- cent higher, and that of kerosene and fuel oil was roughly at import parity. Quantitative Restrictions on International Trade Apart from tariffs and export taxes, the trade regime in Thailand includes a number of restrictive measures such as quantitative import and export controls. At present, there are import bans on eighteen commodities, and special permission is required to import another thirty. The Ministry of Commerce imposes and supervises the im- port controls. Commodities under control include those produced in the socialist countries, weapons and strategic firearms, rice, and sugar. The controls on rice and sugar are designed to prevent reimporting after the products have been exported. The Policy Setting: Role of the Public Sector 85 Export controls are placed on thirty-eight commodities, sixteen of which are out- right bans. The export controls are meant to ensure domestic supplies for local con- sumption at low prices. Items such as paper, pesticide, sheets of flat iron, polyfiber, and cement are regulated to ensure that local supplies are available. Under the Price Setting and Antimonopoly Act of 1979, the Ministry of Com- merce also administers price controls on "essential products." In 1986 thirty-four com- modities came under price control. Such control has helped to keep down the cost of living, but it has been at the expense of shortages of these products in retail stores. Exchange Rate Control Exchange rate management is in the policy domain of the Bank of Thailand, but input from the Ministry of Finance is important. From 1955, when the multiple ex- change rate system was abolished (Corden and Richter 1967) until the devalua- tions of 1981, the baht was maintained at more or less fixed parity with the U.S. dollar (20 to 21 baht per U.S. dollar). The main argument for this peg was the sta- bility and the confidence it was believed to provide. Between 1955 and 1977, the dollar depreciated in relation to other currencies and the baht-dollar rate was ad- justed five times, but these adjustments were all very small. The volatility of the dollar and the enlarged trade deficits in the late 1970s fol- lowing the second oil shock prompted the Thai government to reconsider its ex- change rate policy. Although much of Thailand's trade is denominated in U.S. dollars, less than 20 percent of its export and import transactions are with the Unit- ed States. Pegging to the overvalued dollar in the late 1970s had merely increased the country's balance of payments deficits. In March 1978 the Bank of Thailand announced that the baht would no longer be tied to the dollar but to a basket of currencies in which the dollar would be a major component. The new system was, however, short-lived. In November 1978, a system of daily fixing was introduced, in effect putting the baht back into parity with the dollar, at 19.8 baht per dollar. In 1979 and 1980, Thailand encountered severe trade deficits, which reached a record 4.6 percent of GDP. The deficits resulted from an artificially strong dol- lar-to which the baht was still pegged and which had mitigated against Thai ex- ports-and a strong growth in domestic spending. In response, the government devalued the baht twice in 1981, bringing the rate to 23 baht per U.S. dollar. The government also abandoned the daily fixing system. Such a drastic move, partic- ularly the second devaluation, which then was the largest in recent history (8.7 percent), proved to be politically unpopular and led to the resignation of a deputy finance minister. To build up confidence in the baht after two successive devalua- tions, the government introduced a currency swap in 1981 to guarantee that those bringing in foreign funds would not have to pay back more than the amount they had borrowed from abroad when their loans came due. The abandonment of the daily fixing system was tantamount to returning to the old single-rate, fixed exchange system. In 1984 the baht was again devalued- this time by 14.5 percent-to curb the growing trade deficit. The fixed exchange 86 Thailands\ Macroeconomic Miracle: Stable Adjustment aind Sustained Growth rate system was also abandoned and replaced by a supposedly flexible exchange rate system, under which the baht is tied to a basket of currencies. This switch was to provide greater flexibility in the management of the country's foreign exchange. It is obvious that the basket is heavily dominated by the U.S. dollar. Approximate parity to the U.S. dollar has been maintained in spite of the dollar's realignment in relation to other currencies. This matter is explored further in chapter 9. Financial Markel Regulations Regulation in the financial market is the responsibility of the Bank of Thailand, with the approval of the Ministry of Finance. Financial regulations act both to sta- bilize the economy and to preserve the commercial viability of the financial sys- tem. The present regulations have four distinctive teatures. First, entry to the financial industry, through the opening of new financial in- stitutions-commercial banks, finance companies. and insurance companies-as well as branches of foreign banks, is regulated by the Ministry of Finance through licensing permits. In the case of commercial banks, this control, which is strictly enforced, is aimed at insulating the existing institutions from new competition in order to ensure their long-term viability and stability. In recent years, this tight control, together with the government's unwillingness to let ailing commercial banks collapse, has proved costly: government protection enabled mismanaged banks to continue their operation at public expense. Second, until recently the Bank of Thailand regulated interest rates, by fix- ing the maximum deposit rates and maximum lendinig rates. Since June 1989 controls on the rates for time deposits have been relaxed. Commercial banks have been encouraged to vary their own rates voluntarily within the bounds pre- scribed in response to their liquidity positions. Individual banks, fearing that they would lose their market shares, competed for deposits by offering the ceil- ing rates for deposits. In situations of excess liquidity, this reduced bank profits. As a result, liquidity problems were prolonged until the Bank of Thailand even- tually intervened, by adjusting the ceiling rates The ceiling rates can thus be seen in part as a device for limiting the banks' capacity to exploit their oligopo- listic power. A third novel feature is that Thailand has no direct controls on capital in- flows, whereas until recently capital outflow has been tightly regulated. In these circumstances, local liquidity becomes highly rcsponsive to changes in foreign interest rates and the exchange rate. Since domestic rates are normally regulated at levels slightly higher than the foreign rates, commercial banks and large com- panies can freely adjust their liquidity position by borrowing abroad. This flex- ibility has helped ease the pressure on foreign reserves in times of balance of payments problems. Fourth, regulations within the banking system are also plentiful. Besides the usual monetary policy controls, there are regulations requiring new bank branches to fulfill a variety of conditions. including compulsory holdings of The Policy Setting: Role of the Public Sector 87 government bonds. This makes the market for government bonds a captive one. There is also an institutional regulation known as "agricultural credit policy," which requires commercial banks to lend a fixed proportion of their previous year's deposits (currently set at 13 percent) to agriculture. This regulation, which came into effect in 1975, is designed to enlarge the flow of private credit to agriculture. Labor Market Regulations Government regulations in the labor market are under the jurisdiction of the Min- istry of Interior, to which the Labor Department belongs. Labor regulations take two basic forms: the worker protection scheme and minimum wage regulation, which is more relevant to this study. In response to an International Labour Office appeal for workers' rights and protection, the Labor Department introduced minimum wage regulation in 1973. It was intended to guarantee workers a daily income that would be sufficient to meet their basic needs. The regulation fixes a standardized minimum daily wage for in- dustrial workers in Bangkok and in the provinces. The figure is revised annually by representatives from the government, the employees, and the employers. From 1973 to 1987, the figures were revised fourteen times. The 1987 rate was 73 baht for Bangkok and the five nearby provinces; 67 baht for the provinces of Chonburi, Saraburi, Nakorn Rashasima, Chiengmai; and 61 baht for other provinces. The effect of minimum wage regulations on labor market hiring has been slight, since only large companies and state enterprises adhere to them. According to a survey conducted in 1986, less than one-third of the firms in operation paid their workers the minimum level or higher. This means that fewer than half of the employed unskilled workers were paid the minimum wage. Firms paying less than the legal minimum are mostly small and medium-size firms. Obviously, the threat of unemployment leads workers to accept lower pay in order to secure employ- ment. Only the large firms and state enterprises have formal worker organizations strong enough to police the full administration of the minimum wage regulations. Despite the limited coverage of the minimum wage regulations, revisions to the minimum wage do have an impact, by raising the entire wage structure within the formal sector, including that of the salaried employees. An inflationary impact from such revisions would therefore seem a genuine possibility. Public Provision of Infrastructure During the country's first two development plans (1961-71), the main theme of the planning process was the need for basic infrastructure. By the 1980s, the na- ture of public provision had changed significantly. The emphasis of public invest- ment shifted from infrastructure to development projects. The idea was to develop 88 Thailands Macroeconomic Miracle: Stable Adju,wnent and 5u.stained Growth strategic industries that would help strengthen the economy's ability to absorb ex- ternal shocks as well as to provide a foundation for resource-based industrializa- tion. The fertilizer project, the steel project, and the Eastern Seaboard Project were examples of such development. The provision of infrastructure during the first two development plans was concentrated in roads, irrigation, power, and telecommunication. The expansion of the road network in 1960s, which was tacitly linked to an American-supported counterinsurgency program, had a considerable impact on agricultural develop- ment and overall economic growth. It provided farmers with direct access to ex- ternal markets, which significantly increased farmigate prices for cash crops, as well as access to a vast amount of uncultivated land. The extension of the land frontier was instrumental in the agricultural growth of the 1960s. Investment in irrigation was designed to provide a degree of water control in wet-season agriculture. It took the form of building large dams and waterways. These activities provided a source of power supply and a basis for investing in power, telecommunication, and electrification projects. Considerable effort also went into developing airports and harbors. Some of these projects were undertak- en because of the security fears associated with the Vietnam War. By the middle of the 1970s Thailand's infrastructure compared lavorably with that of other de- veloping countries. Through the provision of basic infrastructure in the 1960s, the government provided a stimulus for private investment and economic growth. The main draw- back of this effort was its unequal distribution of benefits. The irrigation system, for example, was concentrated in the central plains and the lower north and lacked a proper feeder system, with the result that the main beneficiaries were large farm- ers in close proximity to the irrigation sites. Similar problems were observed in the distribution of power, telecommunications, and electricity. In the Third Plan (1972-76), the pattern of public expenditure was modified slightly to give greater emphasis to the distribution problem. One reason for this shift was the move in 1974 toward representative government. More emphasis was given to providing rural areas with services, notably, electrification, rural health centers, and family planning. Between 1974 and 1976, the share of public expen- diture on health, education, and agriculture rose considerably. In the Fourth Plan (1977-81), however, public expenditure on agriculture, health, and education declined while that on industry, energy, transport, defense, and administration increased. This change was the result of a number of economic and political developments. Following the military coup of 1976, control of the government's finances was back in the hands of the military and the bureaucrats. In 1977 a special government decree significantly extended the public sector's di- rect access to foreign borrowings. For the first time, state enterprises were able to borrow directly from abroad, with government guarantees, in order to finance their capital investment. This decree led to a dramatic expansion in defense expenditure and in the role of state enterprises. Between 1978 and 1983, there was a steady in- crease in capital expenditure by the state enterprises, financed in particular by for- The Peicv Selling: Rotle f the Public Sector 8'9 eign borrowing. The bulk of this spending was on energy-related activities because of the need to develop alternative local sources of energy so as to reduce the economy's dependence on imported energy. In 1977, at the time the decision was made to liberalize the public sector's for- eign borrowing, there was no definitive policy on the management of public debt. Regulations on foreign debt merely limited the amount each state enterprise could borrow, but not the aggregate for the public sector. The enlarged foreign debt com- mitment became a serious policy problem in the early 1 980s, when the economy went into a recession following a slump in primary commodity exports. The cur- rent account deficit reached 7.1 percent of GDP in 1983. and the debt-service ratio reached 23.3 percent and was still rising. This development, which was considered a threat to the country's financial stability, prompted the government to revise its foreign borrowing policy and to establish the National Debt Reform Committee. One of its new policy measures was to set annual limits on the public sector's for- eign borrowing. Several of the large development projects being planned. most of which relied on foreign funds, were either scaled down or postponed owing to lack of foreign exchange. Public Enterprises As in many other developing countries, public enterprises play a significant role in the Thai economy (Jones 1982). The sector has growni rapidly in recent years and its role in resource allocation has become an important concern of economic policy. Far from being confined to infrastructure, the activities of Thailand's state enterprises stretch into many areas of business, including manufacturing, trans- port, hotels, services, trade, and finance. The Bank of Thailand is also formally a state enterprise. At present, there are sixty-eight state enterprises, seventeen of which operate in infrastructure. Most state enterprises began as special projects (revolving funds) under central government departmenlts with their own staffs and financial accounts, then slowly graduated to the status of state enterprises when their activities expanded. Historv Until the end of absolute monarchy in 1932, state business ventures consisted mainly of royal trade monopolies in a small number of luxury commodities and the provision of basic utilities. This situation continued until the beginning of World War II. whereupon Thai nationalists in the government began expanding the role of state trading companies in reaction to the Chinese domination of domestic commerce. This was a move to reassert Thai control over domestic trade (Silcock 1967: 260). 90 Thailand's Macroeconomic Miracle: Stable Adjustment (nd Sustained Growrh After the war, public enterprises became increasingly important in other sec- tors of the Thai economy because of the income they could provide for military officials and civilian politicians. By 1957, a large part of the country's industrial capacity was controlled by public enterprises. The industries they dominated in- cluded tobacco, paper, sugar, gunny bags, timber, tin, metal cabinets, pharmaceu- ticals, batteries, tanneries, textiles, cement, spirits, glass, rubber footwear, alum, and shoe polish (World Bank 1959: 90-91). The growth of the public enterprises had been haphazard. Not only was there little economic rationale for the choice of industries entered by the public enter- prises, but they had become uncontrolled in their operation, finance, and invest- ment behavior. Furthermore, their practice of lending public funds to one another made their accounts difficult to interpret (Kraiyudht 1993). The end result was that these enterprises became major vehicles for the purchase of political patronage. This was especially important for those Chinese businessmen whose interests were directly threatened by the public enterprises. On taking control of the government in 1959, Field Marshal Sarit Thanarat es- tablished a government committee to investigate the country's state enterprises. The recommendations-based on its findings and on later, similar studies-were not implemented. For example, a 1965 study by the United States Agency for In- ternational Development, (USAID) on behalf of the Thai government recommend- ed the sale of thirty state enterprises to private industry, almost all of which still remain in the public sector (Kraiyudht 1993). Because the management of a state enterprise comes under the jurisdiction of its parent ministry, its chairman is a political appointee. Following the move to- ward constitutional government in 1974 and the increased number of civilian pol- iticians assuming ministerial portfolios, the political influence in state enterprises has continued to increase. In 1986, fifty-six out of sixty-eight chairman of state en- terprises were members of political parties. Through their control of state enter- prises, the political parties have therefore been exerting increasing influence on economic decisions. In the past, these chairmanships were dominated largely by senior bureaucrats and military officers. Role in the Economy The recent increase in the overall importance of public enterprises within the Thai economy is shown in table 4.6. In 1970 the public enterprise sector was smaller than the central government, as measured by recurrent expenditures, capital ex- penditure, and revenue. By 1988, however, it was larger than the central govern- ment by all three measures.2 The size and growth rate of public enterprises appears quite different when one looks at employment. Thailand's public enterprises tend to be capital inten- sive, in comparison with the central government. They employed almost 300,000 persons in 1988, which was approximately one-quarter the number employed by the central government. The rate of increase of employment in the public enter- The Policy Setting. Role/ of the PblMic Sector 9J/ Table 4.6 Economic Importance of Public Enterprises, 1970 and 1988 (perc-ent) Recurrent expendituire! Capital expenditure! GDP GDP Reenuue/GD)P Public Central Public Cetiral Public Centrall Year enterprises government enterprises goiernnuet entcrprise.s geoerninent 1970 9.5 12.1 1.9 5.9 11.2 13.7 1988 18.9 15.9 7.7 2.7 20.8 16.3 Source: Bank of Thailand. Monthly Bulletin, various issues. prise sector from 1970 to 1988 was 7 percent per year, which was slightly smaller than the 7.8 percent registered by the central government. Trading Profits/Losses aid Pricing PolicyX The public's perception of these enterprises is that they incur huge losses. This is only partly correct. It is true that some important public enterprises regularly lose large sums. The main examples are in the transport sector: the Bangkok Mass Tran- sit Authority incurs losses averaging the equivalent of roughly $30 million a year, and the State Railway of Thailand about $23 million. Nevertheless, in most years. the aggregate revenues of public enterprises exceed their aggregate expenditures. The problem with these calculations is that the expenditures of the state enterprises do not include proper allowance for a return on the public sector's capital invest- ment in them. In any case, some of the state enterprises included in these statistics operate as profit-maximizing monopolists and generale large operating surpluses. Examples are the Thailand Tobacco Monopoly and the State Lottery Bureau. The structure of public enterprise supervision is summarized in figure 4.2. Each enterprise has a board of control and is under the supervision of a govern- ment ministry. In general, the board has formal responsibility for setting prices, but the supervising ministry may also play a role. The more important the enter- prise, the greater the role of the ministry. A few of the most important enterprises must have any changes in their prices approved by the cabinet. This makes it po- litically difficult to raise prices in key public enterprises. In addition to the trans- port sector mentioned above, this applies to the water, electricity, and telephone utilities. The cabinet also approves the price structure of all petroleum-related products. For these key public enterprises, the redistributive effects of changes in prices and the political implications of these effects dominate discussions of tariff rates. Decisions to change prices are made only rarely, and long after economic circumstances have changed. In the past, this slowness to adjust prices has ac- counted in large part for the losses incurred by key public enterprises. The role of the boards in setting prices has been increasing. In the face of financial problems, 92 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 4.2. Structure of Public Enterprise Supervision Cabinet Fiscal Policy Office Ministry ofFinance _ ______ e Supervising National Economic Comptroller- Minstry and Social General's \ _ -- - Development < Office \ Board \ _ _ t/ ~~~~~~Budget National Public C Debt Policy enterprise Committee National Bureau of State Enterprise the Budget Committee L L_7~~_ Au ditor 1 General's| L Otfice Source: Authors' compilation. the Prem government placed more power over pricing within the boards of the public enterprises. Investment Behavior and Foreign Debt In 1985 capital expenditure by state enterprises accounted for more than 70 per- cent of total (consolidated) public sector expenditure. State enterprises' capital budgets are subject to the review and approval of the National Committee on State Enterprises. whose secretariat function is shared by the National Economic and Social Development Board, the Budget Bureau, and the Fiscal Policy Office. It is at this stage that the investment plans of state enterprises are scrutinized in detail. This practice allows the NESDB to tailor the prop(osed investment plans to the pri- orities laid down in the national development plan. Once approved, the projects The Polic y Setting: Role of the Public Sector 93 seeking budgetaiy support in the form of a contribution from the central govern- ment receive further review by the Budget Bureau. Such support, if granted, be- comes part of the central government's expenditure budget. Public enterprises accounted for about two-thirds of the public sector's for- eign debt outstanding in 1985. Servicing this debt absorbs roughly half the total debt-servicing burden of the public sector. In recent years, recognition of this problem has caused the investment plans of all public enterprises to come under close scrutiny from the NESDB. The implications of enterprises' investment plans for foreign debt servicing receive the closest examination. Not only are public en- terprises now required to have foreign and domestic borrowing plans examined by the National Debt Policy Committee, but all such loans are then negotiated and signed by the Finance Ministry's Fiscal Policy Office. Regulation of External Public Sector Debt The history of Thai public debt regulations is summarized in table 4.7. Before the 1960s, Thai governments avoided extemnal borrowing. This implied that a sound investment program would have to be canceled if it called for heavy borrowing from abroad. Beginning in the early 1960s, with the launching of the first devel- opment plan and its heavy public investment in infrastructure, the government rec- ognized that foreign borrowing would be necessary to finance such investment. Despite this change in policy, the government continued to favor financial conservatism, as reflected in the 1959 proclamation limiting the size of the budget deficit to within 25 percent of its expenditure. In the following year an external public debt law was setting a 5 percent ceiling on the permissible debt-service ra- tio, defined as the total debt service as a percentage of total exports, as measured by the Ministry of Finance, and limiting the size of foreign debt service to less than 13 percent of forecast revenue. In 1964, the ceiling on the debt-service ratio was raised to 7 percent in anticipation of heavy borrowing for public investment. Also in 1960, the National Debt Policy Committee (NDPC) was established to monitor and regulate the foreign borrowing activities of government departments. Review by the NDPC became mandatory for public sector projects requiring for- eign loans. This committee reports to the cabinet and is chaired by the finance minister. Its main responsibility is to ensure that the conduct of foreign borrowing by the public sector is within the framework set by the cabinet. Foreign loan pro- posals of all public sector agencies are compared and ranked in order of priority. It is here that political lobbying for foreign loan quotas is most intense. In 1960 the NDPC consisted of five members from the Finance Ministry, the Budget Bureau, the National Economic Development Board, the Bank of Thai- land, and a foreign representative. In 1964 the number was increased to nine and included the president and the general secretary of the National Economic Devel- opment Board. The British-trained economist Puey Ungphakom, governor of the 94 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 4.7 A Chronology of Public Debt Policy, 1959-90 Year Policy 1959 Government budget deficit limited to less than 25 percent of government expenditure. 1960 Ceiling on public sector debt service ratio (DSR) set at 5 percent. Foreign debt service of public sector limited to less than 13 percent of forecast revenue. National Debt Policy Committee (NDPC) set up to monitor and regulate foreign borrowing. 1964 Ceiling on DSR raised to 7 percent. 1976 Total public sector foreign borrowing Ihnited to less than 10 percent of total government expenditure. Ministry of Finance empowered to borrow from abroad. 1977 Ceiling on DSR raised to 9 percent, of which 2 percent was reserved specifically for military borrowing. NDPC empowered to control public enterprises' foreign borrowing. 1981 Ministry of Finance empowered to negotiate foreign loans for military procurements. Ceiling on DSR of 9 percent set to include borrowing for both military and nonmilitary purposes. 1984 DSR ceiling temporarily raised to 11 percent for the period 1984-87 to accommodate refinancing program. 1986 A limit on total public sector foreign borrowing for all purposes established at I billion dollars per year. 1990 Limit on public sector foreign borrowing raised to 2.5 billion dollars per year. Source: Authors' compilation. Bank of Thailand between 1959 and 197 1, was closely involved in implementing the government's financial discipline plan. Dr. Puey was at times director of both the Budget Bureau and the Fiscal Policy Office of the Ministry of Finance, as well as governor of the Bank of Thailand. His influence on policy was profound and reflected his conservative attitudes regarding financial management. The decisions of the Foreign Debt Policy Commission were not directly in- fluenced by politicians, since most of the members were civil servants. To avoid corruption and mismanagement of the infrastructure funds, the commission en- couraged foreign borrowing from sources such as the World Bank, because these funds would be closely monitored and supervised. Since 1982, the Foreign Debt Policy Commission has consisted of ten members, six of them from the Ministry The Policy Setting: Role o*f the Public Sector 95 of Finance. They are also empowered to regulate public enterprises' foreign bor- rowing. Aside from the Minister of Finance and his deputy. the members have all been civil servants, with the result that the influence of politicians and the military in manipulating foreign borrowing has been greatly reduced. Most military procurement was in the form of military aid from the United States after Thailand signed an agreement with the U.S. government on suppress- ing communism. In 1954 and 1963, for example, foreign borrowing for military purposes amounted to only $0.9 million and $1.5 million, respectively. Military foreign borrowing resumed in 1975, when the U.S. military bases were withdrawn from Vietnam. Borrowing for military purposes substituted for military aid from the United States, especially in 1975 and 1976. when Vietnam occupied Cambo- dia. A new regulation on foreign borrowing imposed in 1976 stipulated that for- eign borrowing had to be less than 10 percent of government expenditure and was to be undertaken through the Ministry of Finance, which was to act as the repre- sentative of the Thai government. In 1977, owing to an increase in foreign borrowing for military procurement, the ceiling on the debt-service ratio from foreign borrowing for military purchases was set at 2 percent, leaving the 7 percent ceiling on the debt-service ratio for other purposes intact. Recognizing that public enterprises were responsible for the mounting foreign debt, the Foreign Debt Policy Commission was empowered to control their foreign borrowing. Between 1972 and 1977 the public and total debt- service ratio were below 3 and 13 percent, respectively. Since 1978. the ratio has increased rapidly because of foreign borrowing for military purchases and bor- rowing by public enterprises. From 1978 to 1982, the total ratio ranged from 14.6 to 16.6, while the public ratio was rising. Public debt was the major cause of grow- ing external debt. After the second oil shock in 1979. Thailand sought help from the IMF for standby-arranged funds. In 1981 the Prem government empowered the Ministry of Finance to negoti- ate all foreign loans for military procurement. In the following year, it decreed that the ceiling on the debt-service ratio was to remain at 9 percent, but no distinction was made between military and nonmilitary borro\hing. The implication of this regulation was that military purchases financed by foreign loans would have to be considered together with other investment projects under the 9 percent limit. Since the military could negotiate loans directly, this placed them in direct conflict with the Ministry of Finance. Usually, military loans were obtained with fairly high in- terest rates and shorter grace periods. In 1978 the share of interest payments on public debt for military purposes was 14 percent. This share increased to 34 per- cent in 1979, 47 percent in 1980, and 56 percent in 198 1. In 1984 the Prem government raised the ceiling on the debt-service ratio from 9 to II percent for the period 1984-87 to accommodate its refinancing pro- gram. In 1986 the amount of foreign borrowing allowed in any one year was set at a maxiinuni of $1 billion. In effect, the ceiling discourages foreign borrowing for large investment projects of heavy industries. In the early 1990s two other important regulations were passed: total debt service could not exceed 9 percent 96 Thailand s Macroeconomic Miracle: Stahble Adjusimen; and Sustained Growth of expected export earnings, and total new borrowings in any year could not ex- ceed $2.5 billion. In 1986 the foreign debt of public enterprises was more than 60 percent of to- tal public debts (table 4.8). The acceleration of the public debt in 1978 and 1979 was due to military borrowing. Although military foreign borrowing slowed after 1979, the debt of public enterprises accelerated from 41.4 percent in 1978 to 60.7 percent of total public external debt in 1980. As a result. the ratio of total outstand- ing foreign debt to GDP rose rapidly after 1978. At the same time, the growth of government sector and public enterprise slowed from 1981 to 1984. The refinanc- ing scheme raised the growth rate of public debt somewhat in 1985 and 1986. Table 4.8 Growth Rate and Components of External Public Debt (percent) Government Public enterprises Reserves! Total out- debt oustand- Growth Growth standing ing disbursed Year rate Share rate Share debt/GDP only 1970 16.2 (57.3) -6.0 (42.7) (4.8) (279.8) 1971 -3.0 (59.3) -10.8 (40.7) (4.7) (247.5) 1972 3.7 (53.7) 30.2 (46.3) (4.6) (288.8) 1973 13.8 (49.2) 36.4 (50.8) (4.2) (332.9) 1974 -0.3 (43.4) 25.8 (56.6) (3.9) (427.8) 1975 -0.7 (35.0) 41.6 (65.0) (4.3) (325.7) 1976 12.3 (41.9) 10.9 (58.1) (4.9) (256.9) 1977 15.5 (34.9) 55.4 (65.1) (5.8) (197.3) 1978 81.3 (40.8) 41.4 (59.2) (8.0) (140.6) 1979 55.6 (41.8) 49.1 (58.2) (10.2) (109.7) 1980 29.0 (36.6) 60.7 (63.4) (12.3) (74.4) 1981 22.8 (35.4) 28.5 (64.6) (14.8) (53.1) 1982 18.6 (35.4) 19.3 (64.6) (16.4) (43.6) 1983 15.5 (35.8) 3.8 (64.2) (17.0) (36.5) 1984 10.2 (36.5) 7.0 (63.5) (20.2) (35.6) 1985 24.3 (37.8) 18.4 (63.2) (2 3.1) (30.5) 1986 21.9 (37.2) 24.9 (62.8) (25.3) (34.3) Note: Figures are percentage annual growth rates, except those in parentheses, which are percentage shares. Soure: Calculated from data from the Budget Bureau Part Two Macroeconomic Adjustment to External Shocks Chapter Five The Shocks The word "shock" implies that something has happened to the environnment in which economic activity takes place that is, first, a surprise, and second, serious enough to require a response. If the change had been fully anticipated. then eco- nomic agents would presumably have adjusted their behavior in advance to its im- plications, once the change came to be expected. If the change was not serious, then it would not be important to analyze its effects. The macroeconomic shocks experienced by Thailand during the study period were both unanticipated and serious. This clhapter concentrates on those originat- ing outside Thailand. notably. the two OPEC-induced petroleum price increases of 1973-74 and 1979-80. An attempt is also made to quantify the importance of these shocks for Thailand's aggregate economic pertfrmance. In addition, the dis- cussion gives some attention to the impact of the terms ot' trade shocks on national income and on short-term and long-term economic growth. Econometric analysis demonstrates that the effect on growth was moderatc. Balance of Payments Shocks The external shocks affecting Thailand were measured in terms of their direct impact on the balance of payments, expressed both in U.S. dollars and as a pro- portion of Thailand's GNP (see tables 5.1 and 5.2. -espectively). The shocks ex- perienced since 1970 can be divided into six types according to their area of impact: * Nonffiel primary exports. Here the shock was temporary, occurring in 1973-74. It is notable that the value of these exports was maintained at a relatively constant share of GNP after the shock had abated. * Imports of petroleuim products. This component of' the balance of pay- ments was by far the most volatile ot' those listed here. Changes in the 94 Table 5.1 Magnitude of External Shocks Affecting Thailand's Balance of Payments, 1970-90 (millions of U.S. dollars) Exports of nonfuel Petroleum product Workers' Tourism U.S. Net direct Year Total exports primary products Total imports imports remittances income government investment 1970 686.0 629 1,269.6 112 9.8 104.3 29.5 42.8 1971 802.5 706 1,288.8 131 13.5 106.2 25.0 38.9 1972 1,045.7 864 1,472.8 150 761.1 130.7 20.1 68.6 1973 1,515.6 1,202 2,039.5 226 122.6 164.6 17.3 77.8 1974 2,405.6 2,006 3,104.7 617 222.3 186.8 11.2 188.3 1975 2,176.9 1,785 3,166.1 699 64.9 219.9 8.2 85.6 1976 2,958.9 2,399 3,502.3 818 16.2 195.6 5.2 79.1 1977 3,454.1 2,804 4,706.1 1,024 33.1 225.8 2.6 106.1 1978 4,043.8 3,042 5,405.9 1,124 18.3 437.3 5.3 49.7 1979 5,234.1 3,883 7,515.4 1,599 34.9 550.1 4.6 51.3 1980 6,447.3 4,578 9,278.6 2,868 93.9 867.5 4.5 186.3 1981 6,884.4 4,981 9,899.2 2,984 64.5 983.3 8.1 291.6 1982 6,834.9 4.915 8,405.2 2.642 89.7 1,038.2 10.1 188.6 1983 6,307.7 4,284 10,186.0 2,481 172.1 1,089.1 18.4 356.2 1984 7,340.1 4,780 10,248.9 2,426 78.5 1,155.6 22.1 407.1 1985 7,058.3 4,222 9,327.5 2,088 71.8 1,169.7 10.9 161.2 1986 8,801.6 4,821 9,341.8 1,230 88.7 1,419.0 9.5 261.6 1987 11,590.2 5,449 13,272.8 1,717 123.7 1,944.9 8.6 183.2 1988 15,786.1 7,597 19,726.1 1,536 77.0 3,118.2 13.5 1,102.5 1989 19,856.9 8,383 25,210.9 2,286 80.5 3,753.3 19.2 1,692.3 1990 22,811.8 8,019 32,548.7 3,064.5 90.5 - 22.0 2,445.3 Source: Bank of Thailand; World Bank. World Debt Tables 1991. Table 5.2 Magnitude of External Shocks Affecting Thailand's Balance of Payments, 1970-90 (percentage ofG NP) Exports of noni - Petroleum Juel primary product Workers' Tourism U.S. Net direct Year Total exports products Total imports imports remittances income government inv estment 1970 9.67 8.87 17.89 1.58 0.14 1.47 0.42 0.6 1971 10.89 9.58 17.49 1.78 0.18 1.44 0.34 0.53 1972 12.83 10.61 18.08 1.84 9.34 1.60 0.25 0.84 1973 14.13 11.20 19.01 2.11 1.14 1.53 0.16 0.73 1974 17.56 14.64 22.66 4.50 1.62 1.36 0.08 1.37 1975 14.63 12.00 21.27 4.70 0.44 1.48 0.06 0.58 1976 17.46 14.16 20.67 4.83 0.10 1.15 0.03 0.47 1977 17.52 14.22 23.87 5.19 0.17 1.15 0.01 0.54 1978 16.97 12.77 22.69 4.72 0.08 1.84 0.02 0.21 1979 19.34 14.35 27.77 5.91 0.13 2.03 0.02 0.19 1980 20.22 14.36 29.10 8.99 0.29 2.72 0.01 0.58 1981 20.08 14.53 28.87 8.70 (.19 2.87 0.02 0.85 1982 19.48 14.01 23.95 7.53 0.26 2.96 0.03 0.54 1983 16.06 10.91 25.93 6.32 0.44 2.77 0.05 0.91 1984 18.04 11.75 25.19 5.96 0.19 2.84 0.05 1.00 1985 19.23 11.50 25.41 5.69 0.20 3.19 0.03 0.44 1986 21.57 11.82 22.90 3.02 0.22 3.48 0.02 0.64 1987 24.22 11.39 27.74 3.59 0.26 4.06 0.02 0.38 1988 26.93 12.96 33.66 2.62 0.13 5.32 0.02 1.88 1989 29.10 12.28 36.94 3.35 0.12 5.50 0.03 2.48 1990 28.43 10.00 40.57 3.82 0.11 - 0.03 3.05 Source: Calculated from table 5.1. 102 Thailands Macroeconomic Mirac le: Stable Adju.'ttnent zind Sustained Growth value of these imports affected all the other external shocks throughout the 1970s and 1980s. The exploitation of natural gas from 1981 onward is be- lieved to have reduced dependence on petroleum imports through the 1980s; both for this reason and because of declining petroleum prices, im- ports of petroleum as a share of GNP declined through the 1980s. * Workers' remittances. The changes in workers' remittances were small throughout the two decades, except for 1972 and, to a lesser extent, 1973 and 1974. This phenomenon was essentially a statistical artifact. It includ- ed large transfers of funds from Indochina-Vietnam, Cambodia, and Laos-just before the Communist victories in those countries. The Bank of Thailand treated them as "unrequited transfers for unknown purposes." They were not really workers' remittances. * Incomefrom tourism. This source of income grew steadily throughout the two decades. By the mid-1980s it was more important to the balance of payments than Thailand's total imports of petroleum products. * U.S. government military expenditure. Atter the Communist victories in Indochina in 1975, this expenditure became unimportant, but even before this its aggregate importance was small. At its peak, in 1975, it was equiv- alent to only one-sixth of Thailand's petroleum import bill. * Net direct investment. This investment did not become really important un- til after 1987, when it rose from less than one to more than two percentage points of GNP. The shocks can also be divided by time period and the nature of the effect. Shock 1, an increase in the price of primary products, occurred in 1973-74. It had a positive effect. As already mentioned, it was a temporary event. Shock 2 was sparked by the 1973-74 increase in petroleum prices, but to sep- arate it from the first it is treated as having spanned the period 1975-78. Its effects were negative. This shock initially coincided with, and was somewhat masked by, the first shock. In contrast to that event, however, the oil price shock was not tem- porary. The increase in petroleum import costs was permanent. Shock 3, which was felt during the period 1979-85. was initiated by the pe- troleum price increases of 1979-80 and the international interest rate increases of 1980-82. It, too, had negative effects. Shock 4, during 1986-90, came in the form of a decline in petroleum prices in the late 1980s, the rising prices of primary commodities, and the movement of light manufacturing enterprises from Northeast Asia to Thailand. This shock had positive effects. All four shocks were initiated by a change in international prices. The effects of these shocks are traced in figures 5.1 to 5.3. One effect was a long-term down- ward trend in the conventional terms of trade (figure 5.1). The 'overall terms of trade" corresponds to the conventional concept of an index of the international prices of Thailand's merchandise exports in relation to those of its merchandise imports. When the contribution of petroleum prices alone is considered-by ex- The Shocks 103 Figure 5.1 Overall Terms-of-Trade Index and Petroleum Terms-of-Trade Index, 1970-90 Index (1970 I100) 140.00 - 120.00 100.00 80.00 ~ \ \~ Overall terms of trade 60.00 Petroleum terms of trade .. / * 20.00 - 0.00 I--------_____________---_______ 4----_______ 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Source: Bank of Thailand. amining the ratio of the index of international merchandise export prices described above to an index of the international price paid by Thailand for its petroleum im- ports-this measure of the terms of trade experienced a dramatic and permanent change during the period of our study. The movement in the nominal prices of Thailand's imported and exported commodities is also telling (figures 5.2 and 5.3). Between 1973 and 1976 there were significant movements in the relative prices of Thailand's imports and ex- ports, as captured by the terms-of-trade indices, but also substantial inflation in the nominal U.S. dollar prices of both categories of commodities. Figure 5.3 decomposes the index of Thailand's nominal import prices into its petroleum and nonpetroleum components, as follows. At time t, the unit value in- dex of Thailand's import prices is given by pjl = ,p,l + ( I _Ca,) pIN (5.1) PIt t I+~cr where Pl0 denotes an index of oil import prices, pIN denotes an index of the prices of nonoil imports, and ct is the share of petroleum in the value of total merchan- dise imports. All three price indices are assumed to be equal in some base year. 104 Thailand's Macroeconomic Mirac le: Stable Adjustmeni and Sustained Growth Figure 5.2. Import and Export Price Indices, 1965-90 Index (1972=1) 6.00 -_ Import price , 5.00 _ , 4.00 ' Export price / 3.00 2.00 1.00…- - - 0.00- I I I I 1965 1970 1975 19)8(l 1985 1990 Source: Bank of Thailand. Figure 5.3 Petroleum and Nonpetroleum Components of Import Price Index, 1970-90 Price index (1972=1) 14 12 Petroleum imports 10 B 6 / ,,.fl----... .............. . . . 2 -...fl . ....... Nonpetroleum import prices 1970 1975 1980 1985 1990 Source: Bank of Thailand. The Shocks 105 Then the index of the prices of nonpetroleum imports can be estimated from data on Pf, Pl° and cc from pIN I IX pIO (5.2) Pt P l l-at The results of these computations are shown in figure 5.3. They confirm that the increases in import prices occurring in 1973-74 and 1979-80 were dominated. in proportional terms, by the surge in petroleum prices. But in both periods, non- petroleum import prices rose significantly as well. Since nonpetroleum imports account for more than 80 percent of Thailand's import bill, these nonpetroleum price increases contributed significantly to the inflation that Thailand experienced in these two periods, along with most of the rest of the world. These import price data also need to be examined in relation to the composi- tion of Thailand's total imports, where significant changes have occurred since 1970. The country's dependence on imported petroleum products has declined markedly, at least partly because of natural gas discoveries in the Gulf of Thailand and changes in domestic petroleum pricing policy (Praipol 1993). These changes in the composition of total imports suggest that data on import prices alone should not be used to conclude whether the increase in petroleum import prices or non- petroleum import prices, both in relation to export prices, was the more significant event. We return to this issue below. The episodes described by shocks I to 4 above represent the external changes that had the largest implications for Thai national iicome. Of course, there were other shocks, unique to Thailand-such as the withdrawal of U.S. military aid af- ter 1975 and the loss in foreign exchange earnings, following the withdrawal of U.S. military forces from Vietnam, connected with their "recreation" spending in Thailand-but these were less important for Thailand's balance of payments than the four shocks summarized above. From 1972 to 1974, Thailand's petroleum and petroleum product imports, of which crude oil represented almost 90 percent by value. increased from 9 percent to 19 percent of the value of total merchandise imports. The resulting negative shock to Thailand's balance of payments was equal to about $426 million at 1973 prices, or 3.9 percent of GDP. Following the second major price increase, in 1979- 80, petroleum imports increased again, this time from 21 percent of total imports in 1978 to 30 percent in 1980. This shock caused the balance of payments to de- teriorate by another $640 million, or 2.4 percent of 1979 GDP. From 1982 to 1990 the average price (in U.S. dollars) of Thailand's petroleum imports was cut in half. Over the same period the unit value of Thailand's exports rose by 12 percent. Nevertheless, the significant reduction in petroleum prices through the early and mid-1980s only partly mitigated the effects of the two oil price shocks of the 1970s. The fact was that bh 19 90 imported petroleum. in rela- tion to the unit value of Thailand's exports, still cost Thailand thlree times as much per barrel as it did in 1970. 106 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Effects of the Shocks on National Income Between 1971 and 1990 real income growth deviated significantly from its trend growth rate of 6.7 percent (see figure 3.1). As the following paragraphs show, this deviation clearly seems to be related to the two oil price shocks and to the high international interest rates following the second of these shocks. Figure 5.1 reveals that Thailand experienced two major trade shocks: a mod- erate decline in the nonoil terms of trade, and a proportinately much larger decline in the oil terms of trade. To assess the relationship between the changes in the terms of trade and national income, we decomposed these changes into an "oil terms-of-trade" and a "nonoil terms-of-trade" component and then estimated the extent to which changes in each caused a gain or a loss in national income. We estimated this income loss by taking the difference between the actual val- ue of imports and the estimated counterfactual value of imports (see tables 5.3 to 5.6). The latter is based on the assumption that the terms of trade remained con- stant at their preshock level. The method then expresses the value of the trade shock as a sequence of lump-sum changes in national income at constant prices. This is done by constructing a sequence of counterfactual imports: estimates of what the value of imports (in constant prices) would have been in the absence of the shock. The counterfactual value of imports is estimated by multiplying the ac- tual value of imports by the terms of trade index, with the latter set at unity in the base period, 1970. The difference between the actual and counterfactual value of imports represents the net income loss due to the terms of trade deterioration. The calculation starts from 1970 and extends to 1991. The results of our calculations are expressed in U.S. dollars, at constant 1970 prices. The nonoil terms-of-trade effect is calculated for any year i as N QNM pNM /pX NM p NMIp X (5.3) L Qt (pNM/p) QN ( o0 o - QNM (pNM /pX ) - QNM where QNM = the actual real value of nonoil imports in year t (valued at 1970 prices); PA'm = the nonoil imports price index for year t (1970 = 100): Pt = the exports price index for year t (1 970 = I 00): and P NM, P X = the price indices 0 P0 of nonoil imports and exports in the base vear of 1970. We normalize such that p "M = P X = 1. The actual real value of nonoil imports is calculated in table 5.3 by deflating the current value of nonoil imports with the nonoil import price index. Thus, the term QSN"'(PN+'Af/PX) is the equivalent value of exportable goods that has to be sacrificed to obtain the current value of nonoil import in year t. The term QNM (P6NMIP X)= QNM is the counterfactual value of nonoil imports, and is equivalent to the value of exportable goods that would have been sacrificed to ob- tain the observed quantity of imports if the nonoil terms of trade had remained at their base-year level. The Shocks 107 Table 5.3 Magnitude of the Nonoil Terms-of-Trade Shocks, 1970-91 Nonoil Counter- Export imports Nonoil Actual factual Nonoil price price terms nonoil nonoil terms-of-trade indexa indexa of tradeb imports' importsd gaine Year (1) (2) (3)C (4) (5) (6) 1970 100.00 100.00 100.00 1,157.55 1,157.55 0.00 1971 89.86 118.45 75.86 977.41 741.43 235.97 1972 98.70 118.08 83.59 1,120.28 936.45 181.83 1973 151.35 129.07 117.27 1,405.10 1,647.71 242.61 1974 208.92 188.32 110.94 1,321.02 1,465.54 144.52 1975 197.11 196.85 100.13 1,253.28 1,254.93 1.65 1976 191.71 205.92 93.10 1,303.55 1,213.57 -89.97 1977 195.85 221.49 88.42 1,662.41 1,469.97 192.44 1978 211.74 244.05 86.76 1,754.50 1,522.17 232.33 1979 253.02 267.73 94.50 2,209.79 2,088.33 121.46 1980 299.07 293.17 102.01 2,186.62 2,230.63 44.01 1981 308.07 341.49 90.21 2,024.98 1,826.80 198.19 1982 286.23 354.08 80.84 1,627.65 1,315.75 311.91 1983 331.25 340.11 97.40 2,265.47 2,206.46 -59.01 1984 287.71 351.68 81.81 2.224.43 1,819.80 -404.63 1985 295.52 380.21 77.72 1,904.05 1,479.92 -424.13 1986 303.78 396.00 76.71 2,048.44 1,571.39 -477.06 1987 325.21 426.56 76.24 2,711.80 2,067.50 -644.30 1988 352.35 487.84 72.23 3,731.67 2,695.27 -1,036.41 1989 362.64 518.24 69.98 4,418.21 3,091.69 -1,326.53 1990 370.23 533.33 69.42 5,530.83 3,839.43 -1,691.40 1991 382.49 558.56 68.48 6,153.67 4,213.88 -1,939.79 a. 1970= 100. b. (3)=(1)+ (2)x 100. c. Millions U.S. dollars at 1970 prices. d. (5) = (3) x (4) + 100, millions of U.S. dollars at 1970 prices. e. (6) = (5) - (4), millions of U.S. dollars at 1970 pfices. Source: Authors' calculations using raw data from Bank of Thailand, with the base year readjusted to 1970. Similarly, table 5.4 estimates the extent to which changes in the oil terms-of- trade shock caused a loss of income to the country. This is called the oil terms-of- trade effect and is calculated for any year t according to the similar formula: 108 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 5.4 Magnitude of the Oil Terms of Trade Shocks, 1970-91 Counter- Actual oil factual oil Oil terms of import imports trade gain Export Oil Oil (million (million (million index import terms UJSs) IS$) US$) imports price index of trade (at 1970 (at 1970 (at 1970 (1970=100) (1970=100) (1970=1100) prices) prictes) prices) Year (1) (2) (3) (4) (5) (6) 1970 100.00 100.00 10(.00 11 .00 112.00 0.00 1971 89.86 112.74 79.70 11tl.2() 92.61 -23.59 1972 98.70 110.60 89.25 135.63 121.04 -14.58 1973 151.35 134.16 112.81 168.46 190.04 21.59 1974 208.92 436.30 47.88 141.42 67.72 -73.70 1975 197.11 479.48 41.11 145.78 59.93 -85.85 1976 191.71 520.18 36.85 157.25 57.95 -99.30 1977 195.85 559.53 35.00 183.01 64.06 118.95 1978 211.74 553.55 38.25 203.05 77.67 125.38 1979 253.02 772.27 32.76 207.05 67.84 139.22 1980 299.07 1,306.31 22.89 219.55 50.26 169.28 1981 308.07 1,589.85 19.38 187.69 36.37 151.32 1982 286.23 1,655.24 17.29 159.61 27.60 132.01 1983 331.25 1,448.82 22.86 )71.24 39.15 132.09 1984 287.71 1,400.45 20.54 173.23 35.59 137.64 1985 295.52 1,553.33 19.02 134.42 25.57 108.85 1986 303.78 851.07 35.69 144.52 51.59 92.94 1987 325.21 947.35 34.33 201.75 69.26 132.49 1988 352.35 810.48 43.47 162.14 70.49 91.65 1989 362.64 923.45 39.27 287.36 112.85 174.51 1990 370.23 1,127.40 32.84 331.80 108.96 222.84 1991 382.49 1,143.97 33.44 305.04 101.99 203.05 Note: Calculation of columns (3), (5) and (6), as in table 5.3. Source: Authors' calculations using raw data from Bank of Thailand, with the base year readjusted to 1970. 5 4) LP = QpM (PfM /PX) - QPM (pj'M,p X = QPM (pPM /pX) QPM where QPM - the actual real value of oil imports in year t (valued at 1970 pric- es); PPM = the oil imports price index for year t ( 1970 = 1 00); and p PM, Pox 7hu Shocks 109 the oil imports and exports price indices at the base year of 1970. Again we nor- malize such that p PM - p X .- The actual real value of oil imports is the current Xalue of oil imports deflated by the oil imports price index. The term QPMA1(pPf4/PXI is the equivalent value of exportable goods that has to be sacrificed to obtain the current value of oil im- ports in year t. The term QPM (P)M 1pX ) QPM is the counterfactual value of oil imports and is equivalent to the value of exportable goods that would have had to be sacrificed to obtain the observed volume of imports if the oil terms of trade index had remained at its level in the base year. It must be stressed that these calculations relate to actual outcomes in relation to counterfactual outcomes. The estimated income et'fects of the terms-of-trade changes presented above should not be confused with the change in totail income that actually occurred in relation to the base period. The changes in total income reflected the impacts of many factors, of which the terms of trade was only one component. Rather, the results above represent an attempt to isolate the impact that the observed changes in the terms of trade had on national income, other things be- ing held constant. They are estimates of the difference between what occurred-as reflected in the observed data-and hypothetical estinates of what would other- wise have occurred without the terms-of-trade shock, or the counterfactual. In constructing a counterfactual, we assumed that the volume of imports was unaffected by the shock. Obviously, this method is somewhat crude. If the shock had not occurred, the allocation of resources and of real consumer expenditures would have been different from those observed, and in this counterfactual situa- tion national income would have been increased as a result of those reallocations. The volume of imports would presumably have been greater as a result. Insofar as our estimated counterfactual does not allow for this fact (in assuming the ob- served volume of imports to have been unaffected by the shock), it may thus un- derestimate the counterfactual value of imports and hence overestimate the value of the income loss due to the shock. Nevertheless, the error involved seems likely to be small. Table 5.5 summarizes our estimates of the total income loss resulting from both the nonoil and oil terms-of-trade changes and their combined effect. Table 5.6 expresses these results as proportions of GDP. It should be kept in mind that the re- sults for the later periods are less reliable than the earlier results because the coun- terfactual used in the calculations is based on the year 1970. This base period therefore becomes increasingly irrelevant as it recedes into the past. Thus only the results for the first three shocks are presented. From 1971 to 1985 there was a loss of Thai national income of about $3.5 bil- lion (in 1970 prices) as a result of the total terms-of-trade loss. This was equivalent to about 1.3 percent of total GDP over this period. Only about 40 percent of this loss was directly attributable to the effect of the oil terms-of-trade shock. This is a significant result. Other discussions of the terms-of-trade decline experienced by Thailand since the early 1970s have concentrated on the effects of the oil price 110 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growvth Table 5.5 Counterfactual Effects of the Terms of Trade on National Income, 1970-91 (millions oj U.S. dollars at 1970 prices) Nonoil terms of Oil terms of Total terms of trade effect trade effect trade effect Year (1) (2) (3)a 1970 0 0 0 1971 -236 -24 -260 1972 -184 -15 -198 1973 243 22 264 1974 145 -74 71 1975 2 -86 -84 1976 -90 -99 -189 1977 -192 -119 -311 1978 -232 -125 -358 1979 -121 -139 -261 1980 44 -169 -125 1981 -198 -151 -350 1982 -312 -132 -444 1983 -59 -132 -191 1984 -405 -138 -542 1985 -424 -109 -533 1986 -477 -93 -570 1987 -644 -132 -777 1988 -1,036 -92 -1,128 1989 -1,327 -175 -1,501 1990 -1,691 -223 -1,914 1991 -1,940 -203 -2,143 Shock 1, 1973-74 387 --52 335 Shock 2, 1975-78 -513 -429 -943 Shock 3, 1979-85 -1,475 -970 -2,445 Whole period, 1971-85 -2,021 -1,490 -3,511 a. (3) = (1) + (2). Note: The summary data at the bottom of the table shows the estimated income effects for changes in the terms of trade during each shock period. Source: Calculated from tables 5.3 and 5.4. The Shocks 111 Table 5.6 Conterfactual Effects of the Terms of Trade on National Income, 1970-91 (percentage of GDP) Nonoil terms of Oil term.s of Total terms of trade effect trade effect trade effect Year (1) (2) (3)a 1970 0.00 (.00 0.00 1971 -3.48 A).35 -3.82 1972 -2.60 -0.21 -2.81 1973 3.07 0.27 3.34 1974 1.72 -0.87 0.84 1975 0.02 -0.97 -0.95 1976 -0.93 -1.03 -1.96 1977 -1.81 -1.12 -2.93 1978 -2.02 -1.(9 -3.10 1979 -1.00 -1.15 -2.16 1980 0.35 -1.34 -0.99 1981 -1.57 -1.20 -2.77 1982 -2.51 -1.(6 -3.57 1983 -0.44 -0.9k) -1.43 1984 -2.91 -0.99 -3.90 1985 -3.38 -0.87 -4.25 1986 -3.51 -0.68 -4.20 1987 -4.23 -0.87 -5.10 1988 -5.92 -0.52 -6.45 1989 -6.90 -0.91 -7.81 1990 -7.91 -1.04 -8.95 1991 -8.45 -0.88 -9.33 Shock 1, 1973-74 2.40 -0.32 2.08 Shock 2, 1975-78 -1.28 -1.07 -2.36 Shock 3, 1979-85 -1.66 -1(.9 -2.75 Whole period. 1971-85 -0.75 -0.55 -1.30 a. (3) = (1) + (2). Note: The summary data at the bottom of the table indicates the total income effect for each shock period, as reported in table 5.5, divided by the sum of GDP in the years covered by the shock. Source: Calculated from tables 3.4. 5.3 and 5.4. 112 Thailandi Macroeconomic Miracle: Stable Adjustment tmdl Sustiined Growth rises of 1973-74 and 1979-80. According to our calculations, changes in other components of the terms of trade, taken together, were more important. Shock I was equivalent to a positive income windfall of 2.1 percent of GDP per year, and shocks 2 and 3 were equivalent to negative windfalls of 2.4 and 2.8 percent of GDP per year, respectively. From the data on GDP growth (see table 3.4), the average GNP growth rate for the years 1970 to 1991 inclusive was 7.3 percent. The effects of the terms-of-trade shocks would seem to be large enough to explain the full magnitude of the deviations around trend growth rates that actually occurred. For the years 1974 and 1975, the average growth rate was 4.55 percent, 2.75 percentage points below trend; and for the years 1979-85, the average rate was 5.5 percent, 1.8 percentage points below trend. These are significant changes to the external environment, but they would not seem large enough to produce macroeconomic crises unless they were handled poorly. Effects of the Shocks on Economic Growth We now use an alternative analytical approach to ilvestigate more systematically the effects of the external shocks on Thailand's growth performance. Whereas the type of analysis presented in the previous section can address only the short-run income effects of the trade shocks, this statistical analysis makes it possible to dis- tinguish short-run from long-run growth effects. To study long-term effects systematically, it is necessary to separate the effect of variations in the terms of trade from the effect ot variations in the conventional determinants of economic growth, growth in the inputs of primary factors of pro- duction. Our analysis aims to improve upon the usual growth accounting analysis by distinguishing between the economic variables that affect national output-the inputs of capital, labor, and land-on the one hand, and changes in the country's external trading environment, on the other. The shock consists of changes in the latter variables, not the former, but the former are changing over time as well. In the usual analysis, an attempt is made to identify a simple relationship be- tween G)P and the terms of trade, ignoring chaiiges in the levels of physical inputs. The impact of the shock is that it alters the economic productivity of national fac- tors of production. It is therefore appropriate to distinguish between those changes in national output caused by variations in the levels of inputs of factors of produc- tion and those changes caused by externally ilduced enhancement or diminution of their productivity. Growth Accounting The objective here is to determine the extent to which variations in the rate of Thai economic growth can be explained by the growth of the conventional primary fac- tors of production-capital, labor and land-or by other factors, such as techno- The Shocks 113 logical change or external price movements. We applied the standard techniques of growth accounting to Thai data in an attempt to answer this question. Our estimating equation was derived by hypothesizing that the terms of trade enter the aggregate production function. Thailand's international trade involves in- termediate goods and capital goods, not merely consumption goods. Thus real na- tional output depends not only on the supplies of physical factors of production and their productivity but also on the terms on which the domestic economy is able to trade with the rest of the world. That is, for given levels of the primary factors of production, real GDP also is affected by variations in the terms of trade because these variations affect the quantities of intermediate goods (like petroleum) that can be purchased in exchange for a given quantity of domestic output. They thus affect the net output that can be produced with a given stock of primary factors. Let the aggregate production function be (5.5) Y =fiK, L, H, P*, t) where Y is GDP at constant prices, K is the aggregate capital shock at constant pric- es, L is the aggregate labor force, H is the total area of cultivated land, P* is the terms of trade (ratio of export prices to import prices), and t denotes time. SHORT-TERM GROWTH EFFECTS. To capture the short-term effects that changes in the terms of trade have on aggregate output we estimated the above equation in Cobb-Douglas form. We used Thai annual data for 1961-89, drawn from the World Bank's World Tables, except for the capital stock series, which is a new se- ries recently released by Thailand's National Economic and Social Development Board. The data on GDP and real factor inputs used in this exercise are displayed in figure 5.4; real GDP, the real capital stock, and the labor force are all indexed to unity in the initial year of our data set, 1961. The terms-of-trade series is as de- scribed above. Constant returns to scale were imposed via the restriction that the coefficients on the variables K, L, and H sum to unitv. It is then possible to rear- range the estimating equation as (5.6) ln(Y/H) = OxO + aKln(KIH) + LLln(L/H) + ,BlnP* + at The estimated parameters and relevant diagnostics (t-statistics in parentheses) were (5.7) ln(Y/H) = -0.33 + 0.27 ln(K/H) + 0.52 ln(IJH) + 0.12 InP* + 0.036t (-0.50) (1.82) (2.21) (2.26) (7.50) where R2 = 0.99. DW statistic = 2.05, and log likelihood = 54.67. The estimated coefficient on the terms-of-trade variable is positive and signif- icantly different from zero at the 5 percent significance level. When the lagged val- ue of this variable is included in the regression, its estimated coefficient is 114 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 5.4. Indexes of Real GDP, Real Capital Stock, Labor Force and Land, 1961-89 Index (1961=1) 10 T 8 6 4 2 1961 1966 1971 1976 1981 1986 1989 -U-- GDP --f Capital -5~-Labor -(- Land Source: Table 5.7. considerably smaller and insignificant. The estimated elasticity of current GDP with respect to the current terms of trade is 0. 14. That is, a 10 percent rise (fall) in the terms of trade leads to a 1.4 percent rise (fall) in GDP. From 197010o 1990, Thai- land's terms of trade declined by 40 percentage points, from an index of 100 to an index of 60, an annual rate of decline of 2.5 5 percent. Our estimates imply that this decline reduced Thailand's rate of growth by an average 0.35 percent per year. Al- ternatively, the lump-sum reduction in GDP implied by a 40 percent decline in the terms of trade was 5.6 percent of mean GDP over this period. Using these estimated parameters, it is now possible to study total factor pro- ductivity growth on an annual basis by applying the equation dinY dinK dlnL dlnH (5.8) Ft = t- - H-At where F, denotes the estimated rate of total factor productivity growth in year t and UK aL and XH denote the estimated coefficients above, and where the implied val- ue of cxq is 0.21. These results were summarized for five periods: 1970-72, the period before shock 1; 1973-74, which captures the impact of the first shock; 1975-78, which captures the impact of the second shock; 1979-85, which captures shock 3; and The Shocks 115 1986-89, which captures the impact of shock 4 (see table 5.7). The impacts of the two negative terms-of-trade shocks caused by the two OPEC price increases are clearly evident by comparing the results for 1975-78 (shock two) and 1979-85 (shock 3) with the other three periods-before and after the two oil price shocks, respectively. Total factor productivity (TFP) growth was apparently reduced by the two oil price shocks, but the effect was far from catastrophic. Significantly, TFP growth remained positive in spite of the shocks. LONG-TERM GROWTH EFFECTS. The lagged value of the dependent variable may be included as a regressor in order to distinguish between short- and long-run effects. The Cochrane-Orcutt autoregression technique was again used to correct for auto- correlation. The estimated equation was: Table 5.7 Growth Accounting, 1970-89 Whole Shock I Shock 2 Shock 3 Shock 4 period Variable 1970-72 1973-74 1975-78 1979-85 1986-89 1970-89 Growth data (percentper Year) GDP 4.42) 6.83 8.27 5.33 9.41 6.87 Labor force 1.97 0.95 5.80 2.53 3.28 3.15 Capital stock 2.37 2.96 4.66 6.15 8.26 5.07 Land 5.00 4.98 0.39 0.69 1.29 1.81 Terms of trade -5.22 10.35 -9.24 -3.61 0.75 -2.58 TFP 1.71 4.50 3.62 3.35 5.16 3.47 Contribution to overall growth (percent) Labor 23.34 7.28 39.99 22.18 18.56 23.96 Capital 14.54 11.70 15.24 31.03 23.77 19.99 Land 23.47 15.13 0.97 2.70 2.85 5.49 Terms of trade change -14.00 17.96 -13.25 -8.04 0.94 -4.45 TFP 38.65 65.89 43.80 44.09 54.82 50.56 Source: Author calculations, with data drawn from World Bank. World Debt Tables, various issues, and National Statistical Office, Quarterlv Bulletin of Statistih u varcious issues. 116 Thailand s Macroeconomic Miracle: Staible Adjustment and Sustained Growth (5.9) In( Y/H) = 0.31 + 0.09 ln(K/H) + 0.71 ln(IIH) + 0.16 InP* + 0.027t + 0.25 In( Y/H)I (0.47) (0.55) (2.99) (2.50) (3.19) (1.17) where R2 = 0.99, the DW statistic = 2.13, the Durbin h-statistic = 0.74, and log likelihood = 54.58. The estimated short-run elasticity of real GDP with respect to the terms of trade is 0.16, which is slightly higher than indicated above and is again significant- ly different from zero. The advantage of this specification is that the long-run elas- ticity is given by (31(1-6), where 6 is the lagged dependent variable. Thus the estimated long-run elasticity is 0.21. The long-term decline in the terms of trade over the twenty years from 1970 was 2.55 percent per year, which implies a long- term reduction in real GDP of 0.54 percent per year. According to these estimates. GDP growth would have been 0.54 percent higher than it was but for the terms-of- trade decline that actually occurred. From shocks I to 4 the terms of trade changed at annual rates of 10.35, -9.24, -3.61, and 0.75 percent, respectively (table 5.7). Combined with our estimates of the short-run and long-run elasticity of GDP with respect to the terms of trade, these imply short-run changes in GDP growth of 1.66, -1.48, -0.58, and 0.12 percentage points, respectively. The implied long-run changes are 2.15, -1.94, -0.76, and 0.16 percentage points, respectively. Conclusions The most important external shocks to affect Thailand from 1970 to 1990 were the two rounds of OPEC-induced oil price increases of the 1970s. The effects of these two shocks on Thailand's balance of payments were equivalent to 4 percent of 1973 GDP and 2.4 percent of 1979 GDP, respectively. Shocks of this size seem large enough to cause serious problems but are not necessarily large enough to produce crises, unless badly managed. We found that the size of the shocks was large enough to explain fully the de- viations of Thai GDP growth around its trend. This may be taken to mean that the shocks were large, or that the shocks caused GDP to fluctuate by roughly the mag- nitude of the effect of the shocks on the balance of payments, and no more. That is, the shocks were not destabilizing. Our analysis of Thailand's total factor productivity growth indicates a positive relationship between total factor productivity growth and the terms of trade (ratio of export prices to import prices). Total factor productivity was affected in the short run by terms-of-trade movements in the predicted direction, but this effect was mild and short-lived. Adjustment to these short-run effects was rapid in that the estimated long-run effects were only 1.33 times as large as the estimated short-run effects. In the long run, total factor productivity continued to increase, despite a long-term de- cline in Thailand's terms of trade. The estimated long-term reduction in GDP growth caused by the long-term decline in the terms of trade was 0.54 percent per year. Chapter Six Thailand's Macroeconomic Response Thailand's aggregate economic responses to the shocks described in chapter 5 can be examined by reviewing, first, the various alternative responses that Thailand could have adopted, second, the pattern that would have been optimal and, third, Thailand's actual responses. In this chapter, Thailand's responses are examined in the light of internal and external balance, the changes in aggregate saving and in- vestment, and the level of external debt. The analysis is conducted at an aggregate, economy-wide level and leaves the discussion of specific instruments of govern- ment policy to subsequent chapters. Analyzing the Adjustment Response It is helpful to begin with a simple analytical framework. The discussion opens with a taxonomy of alternative patterns of adjustment responses that emphasiz- es the concepts of internal and external balance. We then present a simple in- tertemporal framework for analyzing the pattern of optimal adjustment in the face of external shocks like those experienced by Thailand. As this exercise shows, the crucial issue is whether the shock is expected to be temporary or per- manent. Alternative Adjustment Responses Suppose there is a negative terms-of-trade shock. National income falls and along with it, tax revenue. Government expenditure may or may not change. Absorption may fall, but not as much as income. The current account will thus move into def- icit. The government's possible responses can be summarized in terms of three simple stylized cases, which can be called the neutral response, the response 117 118 Thailand s Macroaconomic Miraile: Stable, Adju.atmenl (aid Sustalin(ed Growth aimed at restoring internal balance, and the responise aimed at restoring external balance. NEUTRAL RESPONSE. Under this "nonadjustment" response, the disequilibria in in- ternal and external balance are allowed to persist. The internal disequilibrium is such that as income falls, some domestic resources will become idle. The extent to which this consists of unemployment depends in part on the downward flexibil- ity of real wages and the flexibility of firms' employment and output decisions. The external disequilibrium will be given by (6.1) M-X=(I"- Sp)+((G-T) where M denotes gross imports, X denotes gross exports, I' denotes private sector investment, Sp denotes private sector saving. G denotes government expenditure, and T denotes government revenue. The left-hand side of this expression is the ex- ternal deficit. The right-hand side states that this imbalance is equal to the sum of the private sector and public sector deficits--the two expressions in parentheses. The objective of the neutral private sector response is to save less in relation to in- vestment as a resul]t of the shock. The neutral public sector response will be to run a budget deficit as tax revenues fall. RESPONSE TO RESTORE INTERNAL BALANCE. When GDP falls, the government may respond by stimulating domestic demand so as to maintain growth. The fiscal means of achieving this outcome include a tax cllt, an expenditure increase, or a subsidy to petroleum imports. The result of this response will be a restoration of internal balance at the expense of worsening public sector and external deficits. The private sector outcome will be to maintain consumption and investment spending, at the expense of dissaving. The dissaN ing of individual agents will be financed by domestic and foreign borrowing. and the dissaving of the public sector will correspond to budgetary deficits, also financed by borrowing. In aggregate, foreign borrowing will be the ultimate sourc. of tinance. RESPONSE TO RESTORE EXTERNAL BALANCE. In this case, when income falls, ab- sorption must fall more than in either of the above two cases. Total investment must fall. If there is a private sector deficit, (Ip > Y'P, then this must be offset by a public sector surplus (G < T). Optimal Aggregate Response Which of the above responses is most appropriate for a country faced with the kind of external instability Thailand has experienced? When a simple two-period theo- retical model is used to explore this question. it seems that the answer will depend on whether private and public sector agents expect the shock to be temporary or permanent. The Thai experience in this regard is discussed later in the chapter. Thiilan(d s Macroeconomic Response 1/9 ADJUSTMENT TO EXTERNAL SHOCKS: A SIMPLE FRAMEWORK. Figure 6.1 presents a simple two-period theoretical model of the adjustment of saving, investment, and the balance of payments in response to an external shock. The axes show aggre- gate national consumption and the production of a single commodity in each of two periods: period 0. the present, shown on the horizontal axis; and the future, period 1. shown on the vertical axis. The initial output in each of the two periods is shown by point A. The initial output in period 0, v'v) may be consumed or invest- ed. Investment allows output in period I to be increased. The schedule PP de- scribes these investment possibilities. In a closed economy, in which borrowing from or lending to other countries is not possible, consumption and production must both occur on the PP schedule. If social preferences have the form described by indifference curve U, then con- sumption and production will each occur at point B. Investment in period 0 is giv- en by y0 - c0, which is also the level of saving. This enables future consumption (period I ) to exceed output in that period (dissaving). If an international capital market is now introduced, borrowing from or lend- ing to other countries becomes possible. The line marked KK describes these pos- sibilities. Its slope is -( I + r), where r is the international rate of interest, in real terms. The optimal production point is now point C and consumption ideally oc- curs at point D. allowing a level oftsocial welfare U", higher than U. to be attained. There is now a difference between saving and investinent in period 0. Saving is given by I0)- ct) and investment is higher at Y(1- z) . The difference between investment and saving in period 0 corresponds to c( z( . That is. it is optimal for investment to be Y0 - c( and for this to be financed in part by domestic sav- ing (yo - c0 ) and in part by borrowing trom abroad (c(1 - . ). The amount bor- rowed from abroad is equivalent to a current account deficit and a capital account surplus. This deficit is given by the national accounting identity (6.2) D=Mt- Xt=>,-St where Dt is the deficit, in year t, and M, , X,1 and S, denote imports, exports. in- vestment, and saving in year t, respectively. Before using this diagram for the analysis of trade shocks, it is useful to note that a balance of payments deficit or surplus is not in itself necessarily either de- sirable or undesirable. In the case shown, optimal utilization of the opportunities provided by international capital markets and domestic investment opportunities requires that a current account deficit exist in period 0. There is nothing undesir- able about this. Of course, the borrowings must be repaid in the future, but social welfare is raised in the process. Obviously, the actual level of any current accoLunt deficit that might be ob- served in period (1 could in principle be either larger or smaller than this optimal amount. This point is important because it is commonl for observers of macroeco- nomic statistics to describe any increase of a current account deficit as a "worsen- ing" and any reduction as an "improvement." Our analysis shows that this 120 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth approach is unsatisfactory. An increase or a reduction in a current account deficit from one year to the next could be either desirable or undesirable, depending on whether it represented a move toward or away from an optimal intertemporal al- location of resources. Similarly, the analysis in figure 6.1 suggests that for a net borrower there is an optimal level of debt in the present period. A current account deficit in any year is a flow item that adds to the existing stock of debt (minus re- serves). This is desirable if the present level of debt is suboptimal, and undesirable if present debt is excessive. OPTIMAL ADJUSTMENT. The foregoing apparatus is now used to analyze the opti- mal adjustment to an external shock, such as an improvement in the country's terms of trade. A distinction between temporary and permanent shocks is central to the analysis. A temporary shock is taken to mean one that improves the value of production possibilities in period 0, leaving the value of period I potential output Figure 6.1. A Model of Savings, Investment, and Balance of Payments Future Ci, yi \K\ K Y --- ------------ y~~~~~~~~ C ", -- -,1- - - - -- - - ___ ___ __ ___ ___ __ ___ ___ __ __ __ ___ ___ __ ___ ___ __ ___ __ Present z o c C"O Y', Co,Yo Source: Authors':compiation. Thailand' Mocroet onomic Re.Vponse 121 unchanged from before. A permanent shock will be one that raises production pos- sibilities in both periods. The optimal response to these two types of shocks will be different. A temporary. positive shock may be represented in figure 6.1 as a rightward movement of the endowment point, A. and the associated investment possibilities schedule, PP. Each point on the new PP schedule will have the same slope as one on the old PP schedule immediately to its left. A permanent. positive shock will corre- spond to a movement of point A outward along a ray from the origin, and a corre- sponding proportional movement of each point on the PP schedule outward from the origin. The resulting diagrams are complex, and it is sufficient here to summarize the characteristics of an optimal adjustment as they would be observed in period 0. We hold the international real interest rate r constant throughout the analysis. In response to a temporary and positive shock, the optimal level of investment would be unchanged. The intuitive reason for this outcome is that the optimal level of investment depends on the shape of the PP schedule and the international inter- est rate, neither of which is affected by a temporary shock. Only the position of the PP schedule changes; however, saving in period 0 rises. The intuitive reason is that present consumption is only an imperfect substitute for future consumption. Smoothing consumption over time is therefore desirable, and saving a high pro- portion of the temporary windfall is a way of achieving this outcome. It follows that the current account deficit, the difference between aggregate investment and saving, must decline. A permanent shock will induce an increase in investment, because the change in the shape of the PP schedule implies improved productivity of investment. Sav- ings will also rise, but not as much as in the case of a comnparable temporary shock, that is. one that has the same effect on output in period 0, because output possibil- ities in period I are also improved and the necessity for consumption smoothing is thus reduced. The current account deficit may rise or fall, but any decline would be smaller than that observed under a comparable temporary shock. An example of a positive trade shock in the case of a petroleum importer like Thailand would be a reduction in petroleum prices. The optimal patterns of adjust- ment to a negative trade shock will, of course. be the reverse of those described. Thailand's Aggregate Response Using the above theoretical framework, what can be said of Thailand's aggregate adjustment response to the four shocks identified in chapter 5? Changes in Internal and External Balance Data on the attainment of internal and external balance in Thailand are presented in figures 6.2 and 6.3, respectively. Figure 6.2 resembles a Phillips diagram except that 122 Thailand s Macroeconomic Miraicte: Stable Adjustment aind Sustained Growth Figure 6.2. Income Shocks and Internal Balance, 1971-90 1974 25 Inflation (percent) 1980 20- - 15 197a 1981' 1978 1977 /,1990 1982 5197 1989 19872 \ ; 1983 1976 --- 1988 11 86 19 1984 -4 -2 0 2 4 6 8 Deviation from growth trend (percent) Saurce. Table 3.4 and authors' calculations. the horizontal axis shows deviations from trend growth instead of rates of unemploy- ment. As with most developing countries, Thailand's data on unemployment are un- reliable, and deviations from trend GDP growth rates are the best proxy for internal balance. The vertical axis of figure 6.2 shows the annual rate of inflation and the hor- izontal axis shows deviations from the trend rate of growth of real GDP. Figure 6.2 is thus intended to reveal departures from internal balance. Figure 6.3 shows the ratio of the current account deficit to GDP on the vertical axis and (as with figure 6.2) de- viations from the real GDP growth trend on the horizontal axis. The figure therefore plots the relationship between Thailand's external and internal balance. As figure 6.2 indicates, from 1972 to 1974 inflation surged. The reason was the surge in nominal import and export prices, together with Thailand's fixed ex- change rate. Beginning in 1974, growth rates fell below the trend rate of 6.7 per- cent, but when inflation abated in 1975, income growth also recovered. Growth was above the trend from 1976 to 1978 and inflation remained moderate. Shock 3 coincided with reduced growth (during 1979-81) and increased inflation, espe- cially in 1980. The inflation quickly subsided, but growth remained below the trend until 1982. Inflation remained low in the rest of the decade, but growth rates did not recover until 1987. The two oil price shocks triggered not only large relative price movements among Thailand's imported and exported commnodities, but also substantial in- Thailandl's Macroet onomio Response 123 Figure 6.3. Income Shocks and External Balance, 1971-90 Current account deficijtl,DP Deviation from growth trend 1986 ~~~~~~~~~~~~~percent) 4 -2 4 6 - 1 972 1971973 19 7 dt\2 1971 1 1976 90 9 97 1977~~98 1980 -~ 198 1979 1981 1983 -8 Source Table 3.4 and authors' calculations. creases in their nominal prices (see figure 5.2). The nominal price of imported pe- troleum and of nonoil imports and exports also increased. Since the Thai exchange rate was fixed in terms of the U.S. dollar during these periods, the external infla- tion was imported into Thailand (see figure 3.7 and table 3.4). Following each of the oil price shocks, inflation surged to roughly 20 percent but then quickly abated. in contrast to the pattern in many other developing countries. The reasons for the temporary nature of the inflation in Thailand are examined in chapter 8. From figure 6.3 it is apparent that the shocks disrupted Thailand's external balance for extended periods. Both shock 2 and shock 3 led to large current ac- count deficits, and these were allowed to persist until 1985. It seems clear that maintaining external balance was accorded a lower priority than achieving low in- flation (which was first on the agenda) or maintaining growth. Changes in Aggregate Saving and In vestment Shock 1, the 1973-74 commodity price boom, was too short-lived to be analyzed in the present framework, but the response to the shock 2 (1974-76) round of oil 124 Thailand's Macroeconomic Miracle: Stable Adjuntment and Sustained Growth Figure 6.4. Investment-Savings Gap, 1970-90 Percent 0.4 0.35 0.3 Investmcnt/GNP 0.25 0.25 Savings/GNP 0.15 0.1 0.05 1970 1975 1980 1985 1990 Source: Bank of Thailand. price increases, is clearer. Investment was maintained, and savings fell (figure 6.4). The shock was treated as if it were temporary. Consumption and investment were maintained on the implicit assumption that the shock would not last. As in- come growth recovered through the late 1970s, both the investment/GDP and sav- ings/GDP ratios increased, but in parallel. The contribution of foreign saving to the financing of total investment remained steady. The response to the shock 3 (1979-80) round of petroleum price increases was initially similar. The shock was again treated as being temporary. From 1979 to 1985 investment rates were again maintained, with the brief exception of 1982, and saving rates fell slightly. But by 1985 the resulting external imbalance was be- coming a serious concern. From 1986 onward, adjustment took two forms. First, saving rates rose and investment rates fell. Thereafter, the adjustment story is com- plicated by shock 4, caused by the export and investment-led boom of the late 1980s. The GDP growth rate rose sharply, and both savings and investment rates responded. The larger response of savings, however, virtually eliminated the in- vestment-savings gap over the interval 1986 to 1989. From 1990 the current ac- count again began showing a growing deficit. From figure 6.4 it is clear that the main culprit behind the rising saving-invest- ment gap of the late 1970s was the rapid increase in aggregate investment as a pro- portion of GDP and simultaneous stagnation in aggregate savings. Public fixed investment was responsible for this expansion in investment (figure 6.5). From 1980 to 1986, however, the main explanation for the high investment-savings gap Thailand's Macroeconomic Response 125 Figure 6.5. Composition of Gross Domestic Investment, 1970-90 Percentage of GNP 0.4 0.35 0.3 Change in stocks 0.25 0.2 0t5 _ - w' _,- j!1 005 0 ---+ -+-H+--4- +-++ i. -w W z 1970 1975 1980 1985 1990 Source. Bank of Thailand. was a dramatic decline in gross national savings, which dropped trom 20 / percent of GDP in 1980 to 17.1 percent in 1985 (table 6. 1). As a share of GDP, gross national savings fell by more than one-sixth. Many experts in Thailand and foreign observers attributed the growth in the external debt over the first half of the 1980s to the behavior of the public sector deficit. Public sector savings had indeed declined (table 6.1 ), but far more signif- icant was the one-third drop in household savings, which had fallen from 13.7 per- cent to 9.0 percent of GDP. The decline in household savings as a share of GDP accounted for well over half of the total savings-investment gap. One possible explanation for this drop is that the slowdown in income growth was seen as a temporary phenomenon and therefore was not matched by a slowdown in consumption growth. Another, more speculative suggestion is that the contraction of rural incomes resulting from low commodity prices squeezed aggregate household savings, because rural Thais have a higher marginal propensity to save than urban dwellers. Such a squeeze could hardly be large enough, however, to explain the decline in savings that ac- tually occurred. Public sector savings, operating through fiscal outcomes, without doubt had a stabilizing effect (see chapter 7), but the household sector made a far greater con- tribution in this regard. through its saving behavior (figure 6.6). Personal savings are highly responsive to changes in growth-they arc more sensitive than total in- 126 Tlailand ' Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 6.1 National Savings, 1980-85 (percentage of GDP, current prices) Savings 1980 /981 1982 1983 1984 1985 Net private 14.8 12.2 1 1.5 8.7 8.1 8.7 Households 13.7 11.2 1(.9 7.9 8.3 9.0 Corporations 1.1 1.0 0.6 0.8 -0.2 -0.3 Net public 0.8 0.9 -0.4 1.3 1.0 0.4 General government 0.6 0.6 -0.8 0.5 0.0 -0.3 State enterprises 0.2 0.3 0.4 0.8 1.0 0.7 Net national 15.6 13.1 11.1 10.0 9.1 9.1 Depreciation 7.4 7.5 7.8 8.0 8.0 8.0 Statistical discrepancy -2.0 -3.0 -0.5 -2.1 - - Gross national 20.7 17.8 18.4 15.9 17.1 17.1 Source: NFSDB. Bank of Thailand, and World Bank mission ehtimates. Reproduced from World Bank (1986). Figure 6.6. Composition of Domestic Saving, 1970-90 Percentage of GNP 0.35 Public saving 0.25 0.2 Corporate saving 0.15 0.1 Personal saving ' 0.05 epr _ 0 1 970 1972 1 974 1976 1978 980 1982 1 984 1986 1988 1990 Source. Bank of Thailand. Thailand \ Mamroeconomic Response 127 vestment or public sector savings-and this point is central to an understanding of the long-term macroeconomic stability of the Thai economy. Changes in External Debt The military government that came to power in 1976 introduced some important changes in economic policy. As explained in chapter 4, restrictions on the foreign borrowing of state enterprises and the military itself were relaxed, and public ex- penditure expanded. The latter move caused the current account deficit to increase to 5.7 percent of GDP. From 1977 to 1985 the investment-savings gap averaged 6.4 percent of GDP, compared with an average of 2.1 percent from 197() to 1976. By the early 1980s Thai policymakers were greatly concerned about the growth of the external debt. Table 6.2 summarizes the movement of Thailand's external debt from 1970 to 1990 and its composition. EXTERNAL DEBT. 1980-85. Perhaps the closest that Thailand has come to a "crisis" in the past two decades was in 1982-85. The external debt situation was by then a serious public issue. The subsequent export boom relieved this pressure, but inso- far as this boom seems to have owed more to external developments than to any changes Thailand itself made, the situation could just as well have been very dif- ferent. The post-1985 boom might not have occurred. Table 6.3 provides useful background data to the 1985 debt "crisis." Thailand's credit rating is one of the highest among developing countries. This, combined with prudent management of the composition of Thailand's debt, has enabled Thailand to borrow on favorable terms, as reflected in the relatively low interest rates and long maturity of Thailand's debt. In the period 1980-85, Thailand had ready access to international bond markets, a rarity among developing countries. Moreover, a high proportion of its debt was held with official rather than private institutions. Nevertheless, the size of the foreign debt had become a serious concern. By 1985 public long-term debt was in excess of $8 billion and private debt just over $4 billion, with roughly another $4 billion in total short-term debt. The total debt stock in 1985 was $16 billion, which represented 41 percent of GDP and 146 percent of exports. The debt-service ratio (total debt service, includinig obligations to the IMF, divided by total exports, including factor services) had reached 26 percent, an alarm- ing figure for Thailand's conservative financial managers. Tables 6.4 and 6.5 show the time path of the growth of this debt. Note that in years when the public sector borrowed heavily, the growth of private borrowing was smaller, and vice versa. This relationship appears to be at least partly the result of the way in which domestic and foreign borrowing are used to finance the public sector's deficit. Whenever concern is raised about foreign borrowing, public bor- rowing has concentrated on domestic sources, forcing the private sector to borrow offshore to avoid the resulting high domestic interest rates. Because of this crowd- ing-out effect, public or private foreign debt in isolation could be a misleading in- dicator. Total debt is more relevant. 128 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 6.2 Total Foreign Debt, 1970-90 (millions of U.S. dollars) Interest Year Private Publir Total payments 1970 401 348 749 34 1971 425 368 793 41 1972 505 407 913 43 1973 461 459 920 56 1974 648 528 1,176 86 1975 736 623 1,360 106 1976 785 830 1,615 107 1977 880 1,151 2,031 127 1978 931 1,788 2,719 205 1979 1,243 2,713 3,957 318 1980 1,751 3,953 5,704 510 1981 2,099 5,077 7,175 697 1982 2,296 6,021 8,318 695 1983 2,663 6,865 9,528 703 1984 3,372 7,425 10,797 809 1985 3,370 9,406 12,776 880 1986 3,117 10,954 14,071 1,038 1987 2,837 12,911 15,748 1,093 1988 3,016 12,363 15.379 1,144 1989 4,658 11,743 16,401 1,064 1990 7,341 11,253 18,594 1,501 Accumulation 1974-78 469 1,329 1,799 631 1979-83 1,733 5,077 6,809 2,923 1984-90 4,677 4,387 9,065 7,529 1971-90 7,741 11,600 19,342 11,223 Source: Bank of Thailand. Thailand'S Ma, roeconomic Response /29 Table 6.3 Indicators of Indebtedness, 1970-85 Indicator 1970 1975 1980 1981 1982 1983 1984 1985 Stock of debt (U.S. billions of dollars) Total 1.09 2.28 8.11 10.10 11.34 13.15 14.49 16.06 Long-term 0.72 1.36 5.80 7.22 8.45 9.66 10.92 12.21 Public long-terma 0.32 0.62 4.10 5.13 6.14 7.00 7.57 8.16 Private long-term 0.40 0.74 1.70 2.10 2.32 2.66 3.37 4.05 Short-termb 0.37 0.92 2.30 2.88 2.88 3.49 3.55 3.85 Debt ratios (percentage) Total debt/ GDP 16.6 15.5 24.2 28.0 30.8 32.6 34.5 40.6 Total debt/ exports' 92.9 76.7 94.6 109.2 120.5 142.5 138.6 146.0 Public 27.3 20.9 47.9 55.4 65.2 75.9 72.4 74.2 Private 34.2 24.9 19.9 22.7 24.6 28.8 32.2 36.9 Short-term 31.4 30.9 26.9 31.1 30.6 37.8 34.0 35.0 Debt se nice/exports' Total debt 17.1 15.1 17.3 17.4 13.89 22.9 24.8 26.1 Long-term debt 14.0 12.0 14.6 14.3 15.9 19.1 21.4 22.3 Public 3.4 2.4 5.1 6.7 8.3 10.2 11.9 11.7 Private 10.6 9.6 9.5 7.6 7.5 8.9 9.4 10.6 Short-termd 3.1 3.1 2.7 3.1 3.1 3.8 3.4 3.8 Total debt- service ratio including IMFe 17.1 15.1 17.5 18.3 19.3 23.3 25.4 26.3 a. Long-term debt is debt with over 12 months maturity. b, World Bank staff estimates for 1970 and 1975 (oticial data not asailable). c. Exports of goods and services. d. Interest payments only. e, Includes repurchase obligation to the IMF. iNMF debt is not included elsewhere in the table. Source: World Bank Debtor Reporting System (MtIT data : B.l.S. and government data (short-term debt); INi (repurchase obligations. 1985 data are staff estinaLtes. Based sn World Bank (1986). 130 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 6.4 External Debt, 1967-85 (millions of baht) Debt-export Year Government Private Total ratio 1967 38 49 87 8.5 1968 45 71 116 11.3 1969 43 95 138 12.8 1970 40 125 165 15.1 1971 41 159 200 16.5 1972 45 140 185 12.1 1973 55 216 271 13.0 1974 61 207 268 8.7 1975 76 295 371 13.0 1976 87 299 386 10.8 1977 123 329 452 11.0 1978 200 644 844 17.0 1979 296 698 994 15.4 1980 442 859 1,301 15.7 1981 668 818 1,486 16.5 1982 869 792 1,661 18.1 1983 952 897 1,849 20.5 1984 1,065 1,032 2,097 20.4 1985 1,154 1,221 2,375 21.3 Source: Bank of Thailand The composition of Thailand's foreign debt in 1985 was favorable (table 6.6). New commitments in 1984 were secured at an average interest rate of 8.7 percent in Thailand, compared with an average 10. I percent for middle-income oil import- ers. The average maturity of Thai debt was 17.2 years with a grace period of 6.9 years, compared with 12.2 years and 4.1 years, respectively, for these oil import- ers. In addition, official sources accounted for 63 percent of Thailand's debt, while the average for middle-income oil importers was 39 percent. This left scope for increased private sector borrowing when necessary. The international financial community evidently perceived Thailand to be a creditworthy borrower. The atti- tude among Thai government officials was that Thailand underborrowed in the early 1970s but that-following the change in regulations in 1977 permitting state enterprises to borrow abroad-borrowing was excessive in the late 1970s and ear- ly 1980s. Thailand 's Ma roe( or0imic Response 131 Table 6.5 External Public Debt, 1970-86 Public debt outstanding Debt Public (inillions of outstanding! Interest debt-servwice Year U.S. dollars) GDPa paw!nentsb ratio 1970 325.7 5.0 1.4 4.4 1971 359.6 5.2 1.5 3.2 1972 386.0 4.9 1.2 2.7 1973 441.5 4.2 1.2 2.5 1974 512.9 3.8 0.9 1.9 1975 616.1 4.2 1.2 2.4 1976 822.7 5.0 1.2 2.4 1977 1,119.2 5.8 1.4 2.9 1978 1,819.2 8.0 1.8 3.7 1979 2,827.1 10.6 2.4 4.8 1980 4,070.1 12.4 3.1 5.0 1981 5,127.1 14.6 4.3 6.8 1982 6,137.9 17.2 5.1 8.5 1983 7,000.4 17.9 5.7 10.2 1984 7,540.7 18.6 5.4 12.0 1985 9,836.7 26.7 5.9 14.7 1986 11,022.6 27.4 6.4 16.7 a. Percentage of disbursed debt outstanding only. b. As percentage of exports of goods and services. Soarce: World Bank, World Dehi Tables, various issues. EXTERNAL BORROWING AND THE PETROLEUM PRICE SHOCKS. Because Thailand's external position changed markedly after 1985, it is helpful to focus on the adjust- ments that occurred until then. The two questions of particular interest are whether it is possible that Thailand did not adjust to the oil price shocks, except by borrow- ing, and how the increase in Thailand's external debt compares with the increased petroleum import bill that the oil shocks produced. Consider, first, the value of Thailand's actual imports of petroleum products from 1973 to 1985 and, second, what their value would have been if petroleum prices had maintained the relationship to the export unit price index of the indus- trialized countries that existed in 1972. Suppose that Thailand had financed this difference by external borrowing. How large would the resulting stock of debt have been by 1985? The result is an accumulated principal of approximately $15.5 billion (table 6.7). This calculation assumes that all interest is serviced but the 132 Thailand s Macroeconomic Miracle: Stable Adjustmenit and Sustained Growth Table 6.6 The Structure of External Public Debt for Selected Countries, 1984 and 1985 Average, middle- Terms and Average, income oil structure Thailand Korea Malaysia Philippines East Asia importers Terms on new commitments, 1984 Interest rate 8.7 9.7 9.4 9.0 9.3 10.1 Maturity (years) 17.2 11.9 15.4 14.3 14.4 12.2 Grace period (years) 6.9 4.4 9.1 4.0 5.5 4.1 Structure of outstanding debt, Januanr 1985 Percentage from official sources 63.0 39.3 21.9 47.8 44.6 39.2 Percentage at floating rates 29.4 46.8 61.6 41.0 39.6 46.9 Source: Word Bank, World Debt Tables /985 and 1986; and World Bank ( 1986). principal accumulates. The actual increase in Thailand's total stock of external debt from 1972 to 1985 was the remarkably similar amount of $14.5 billion, which is 94 percent of our hypothetical increase in debt. Furthermore, Thailand's international reserves, measured in terms of months of import coverage, were depleted between the early 1970s and 1985. In 1972 re- serves were equivalent to 7.9 months of merchandise imports, compared with 2.4 months in 1985. The difference in 1985 was worth roughly $4.1 billion. Thus the depletion of reserves plus the increase in the external debt of $18.6 billion-more than accounts for the increased cost of Thailand's petroleum imports since the first oil shock. The former exceeds the latter by 20 percent. It is likely that Thailand's domestic adjustments caused the actual quantities of petroleum imports to be lower than they might have been, but these aggregate statistical calculations are surprisingly consistent with the hypothesis that the pe- troleum price increase was financed by a combination of external borrowing and a depletion of reserves. These calculations suggest that by 1985 Thailand's domes- tic macroeconomic adjustment to the oil price shocks had been deferred. Thailand s NAtacroec onomic Response 133 Table 6.7 Foreign Exchange Cost of Petroleum Price Increases, 1972-85 (U.S. millions ojfdollars) Petroleum Actual price Export unit Proportioial Cumulative pett-oleutia (1972 = valuea price Annual cost cost imports 100) (1972=100) increaseh increaseC increase Year (1) (2) (3) 14) (5) (6) 1972 135 100 00 ( n.a. 1973 209 134 119 0.108 23 23 1974 588 470 149 0.683 402 425 1975 708 468 168 0.642 455 88( 1976 830 467 191 (.591 491 1,371 1977 1.041 503 206 (.591 615 1,986 1978 1,118 516 233 0.549 614 2,600 1979 1,592 689 269 0.610 972 3.572 1980 2.859 1,164 303 (0.740) 27116 5,688 1981 2'974 1.320 291 ).779 2.318 8,006 1982 2.624 1,359 281 0.794 2,081 10,087 1983 2.466 1.190 272 ).772 1.902 11.989 1984 2.410 1,156 264 ).771 1,859 13.848 1985 2.090 1.159 264 01.772 1,613 15,461 a. Index of industrial countries. b. (4) =12) - (3)1 + 12). c. (5) (4) x (1). Sounrc: World Bank (1980. 1986); and IMF International f itiinaiaiel Statistics, various issues. Adjusttnent of Petroleum Imports The 1973-74 oil price shock exposed Thailand's vulnerability to fluctuations in international petroleum prices. Although it seemed obvious to all that reduced im- port dependence was desirable, domestic political conflicts delayed the changes in domestic pricing policy that were required to achieve this end. The prices of vir- tually all Thailand's major commercial energy products are controlled by the gov- ernment. These include all petroleum products. natural gas, electricity, and lignite (a mineral product used primarily to generate industrial energy). Petroleum prod- ucts are especially sensitive in this regard. Kerosene is used by the rural poor for cooking and lighting. Fuel oil is used industrially and, critically, in the generation of electricity, itself a major item in the urban cost of living. In 1970 imported petroleum. principally crude oil, accounted for more than 80 percent of the total consumption of petroleum products in Thailand and 60 per- cent of total energy use (Praipol 1993: 297). But the increase in the international price of crude oil occurring with the first oil shock was not passed on to domestic consumers. The combination of import taxes and excise taxes on petroleum that 134 Thailand's Macroeconomic Miracle: Stable Adjustment (Ind Sustained Growth had previously been a major source of government revenue was scaled down to stabilize domestic retail prices. As a result, domestic prices were not adjusted sig- nificantly in real terms throughout the aftermath of the first oil shock. The shock was implicitly treated as a temporary event. The second oil price shock of 1979-80 showed that this assumption was no longer tenable. When the domestic prices of petroleum products and electricity were raised in 1980, the reaction was so strong that it brought down the Kriangsak govern- ment. The Prem government that succeeded it promptly reversed these price re- forms. The oil fund that had been established by the government in 1974 to tax the windfall profits of petroleum companies on old stock at the time of the first oil shock went into deficit because it became an administrative instrument for financ- ing subsidies on the consumption of petroleum products. In the early 1980s all pe- troleum products were subsidized in this way except for gasoline (Praipol 1993). The heaviest subsidies were on kerosene, fuel oil, and LPG. The subsidies on fuel oil delayed the move on the part of industrial users to natural gas and the subsidies on LPG encouraged costly refits of commercial vehicles to facilitate its use in place of gasoline. Through the mid-1980s, natural gas finds in the Gulf of Thailand reduced Thailand's dependence on imported petroleum. In particular, the Electricity Gen- erating Authority of Thailand moved increasingly away from heavily subsidized Figure 6.7. Petroleum Imports at Current and Constant Prices, 1970-90 Percent 35- 2C~urrent prices 15 j Constant prices S0 X 0 - I _ ___ _ _ _ _ _ 1970 1975 1980 1985 1990 Source: Bank of Thailand and IME Thailand s Macroeconomic Response 135 fuel oil and toward natural gas. By 1990 imported petroleum had declined to 40 percent of total petroleum product use (including natural gas), in comparison with 80 percent in 1970. But Thailand's import dependence had not declined as much as these data might suggest because total petroleum use had increased rapidly in the interim (see figure 6.7, which shows petroleum products as a share of total im- ports in both current and constant 1972 prices). Indeed, these calculations reveal that, at constant prices, petroleum as a share of total imports changed little over the entire twenty-year period. The dramatic changes in petroleum products as a share of imports (at current prices) was due almost entirely to movements in the international prices of petroleum products in relation to other imports and owed almost nothing to quantity adjustments. Moreover, almost twenty years after the first oil price shock, petroleum imports had resumed roughly their 1972 share of total imports, measured at both current and constant prices (figure 6.7). It follows that by the end of this period the ratio of pe- troleum import prices to import prices in general had returned to roughly their 1972 level. But Thailand's terms of trade had declined by more than 20 percent (see table 3.4). Although the increase in petroleum prices in relation to other import prices had vanished, export prices in relation to imports had declined significantly. Conclusions Thailand's aggregate adjustment to its terms-of-trade shocks was quite different from what the theoretical literature would describe as optimal adjustment. Our analysis suggests that optimal adjustment depends on whether the shocks are seen as temporary or permanent phenomena. In the case of a negative shock, if it is tem- porary, then it may be optimal to borrow abroad to smooth consumption during the life of the shock. If it is permanant, aggregate absorption will have to be adjusted, and the full value of the income loss should not be borrowed. Analysis of Thai- land's actual borrowing in relation to the magnitude of the two OPEC oil price shocks shows that the value of additional external borrowing almost exactly matched the value of the loss in the terms of trade due to the shocks. That is, Thai- land behaved as if the shocks were strictly temporary. It is difficult to describe this as optimal adjustment. Previous studies of Thailand's adjustment to the petroleum price shocks have made much of the decline in Thailand's dependence on imported petroleum. Since the two oil price shocks of the 1970s and early 1980s, Thailand's imports of pe- troleum products have indeed declined as a proportion of total imports. Neverthe- less, this outcome was almost entirely a result of changes in the international prices of petroleum products in relation to other imports. When the data are exam- ined at constant prices, there is little change in the share of petroleum products in total imports. Quantity adjustments played almost no role. Again, this could not reasonably be described as optimal adjustment. 136 Thailands Macroeconomic Miracle: Stable Adjustment (Ind Sustained Growth Balance of payments shocks of the type considered in chapter 5 had the po- tential to produce severe macroeconomic problems. Although problems occurred, they could not be considered crises. But since macroeconomic crises did occur in other countries affected by these and similar shocks, it is helpful to review the cir- cumstances that c ould have produced crises in Thailand's case. MID-TO-LATE 1970s. By 1974-75, Thailand had experienced the following exter- nal shocks: a deterioration in the terms of trade caused by the first oil shock (ini- tially cushioned by an export commodity price boom, which had vanished by 1975); the withdrawal of U.S. military grants; and the withdrawal of U.S. bases and associated spending by U.S. military personnel. Remittances from Thai workers abroad, increased tourism, and the growth of manufactured exports then became important, but not until the late 1970s. In the meantime, Thailand was experiencing a serious problem because the govemnment decided to defer adjustment and to borrow internationally. In the mid-I 97()s Thai- land and the foreign commercial banks, then flush with petrodollars, discovered one another. To facilitate foreign borrowing, the Thanin govemnment in 1976-77 eased the restrictions on foreign borrowing by the public sector (see chapters 2 and 4). Was this an irrational move? In the mid- 1970s the stock of Thai foreign debt was fairly low and the country could sustain a higher level. From the foreign banks' point of view, Thailand had been a reliable debtor in the past and seemed to have good economic prospects. Moreover, in the mid-1970s few observers, in- cluding economists, recognized that the international oil price increases would be permanent. Nor did they anticipate the increases in international interest rates in the late 1970s and early 1980s. Given the information available at the time, Thai- land's borrowing strategy could not be described as reckless, or even as unwise, but it was clearly somewhat uncontrolled. In addition, in the mid-1970s Thailand was experiencing severe domestic po- litical problems. Politically, this was one of the most turbulent periods of Thai- land's modern history. The democratic governments of Kukrit Pramoj and Seni Pramoj were on shaky ground. They had come to power in 1973 through a popular revolt against the Thanom military government, but the military remained hostile and politically strong. The democratic period was to end with a military coup in 1976, and the new government was itself replaced by a yet another military coup in the following year. In this political environment. deferred adjustment was obvi- ously the preferred strategy of the politicians. International interest rates were low, and a borrowing constraint was not binding because of Thailand's credit rating. The pressure to defer adjustment by borrowing internationally could not be resist- ed, and this strategy was to prevail until the early 1980s. EARLY-TO-MID I 980s. Thailand's substantial debt burden was the consequence of the adjustments made in the 1970s and early 1980s. But national savings had de- clined alarmingly in the early 1980s at the very time that repayment of the debt was becoming a problem. The resulting balance of payments deficit was clearly Thailand 's Macroeconomic Response 137 unsustainable. The public sector was also unsustainably in deficit. In particular, the government's tax base was narrow, evasion was widespread, and the antiquated public revenue system seemed generally incapable of raising the increased revenue the government now required. These problems are not yet fully resolved today, but by the early 1 990s they had abated considerably. Strong policy measures had been taken (including some fiscal reform), foreign borrowing was curtailed, and several large-scale public investments were either abandoned or scaled down. Chapter Seven The Role of Fiscal Policy At a theoretical level, fiscal policy is more likely to be effective under a fixed ex- change rate regime than under a flexible exchange rate. The opposite applies to monetary policy. Fiscal policy therefore seems a good candidate for a significant stabilizing or destabilizing role in the Thai context, as has indeed been suggested in the literature on Thailand's adjustment experience (Robinson, Yangho, and Ranjit 1991). Whether the manner in which Thailand's fiscal aggregates respond- ed to external shocks exacerbated or ameliorated the short-term macroeconomic effects of those shocks is the subject of this chapter. Fiscal Adjustment: 1970-90 As noted in chapter 4, Thailand's consolidated public sector consists of the central government, local government, and the public enterprises. The fiscal position of the consolidated public sector is dominated by that of the central government (ta- ble 7. 1). The deficits of public enterprises have on average been smaller than those of the central government and have tended to move with them. Local government deficits are insignificant. This chapter concentrates primarily on the central gov- ernment. Unless otherwise stated, terms such as "public sector expenditure," "public sector revenue," and "public sector deficit" thus refer to the fiscal positions of the central government. I A Brief History of Fiscal Adjustments Throughout the 1960s and into the early 1970s, Thailand's public sector deficits was small. The Thai economy expanded steadily, as did revenue. The goal of fiscal policy was to modify the indirect tax system with a view to promoting exports and 138 7The Role *j Fiscal Poli'v 139 Table 7.1 Selected Fiscal Indicators, 1970-90 (ratio to GDP) Central government Public Consolidated enterprise public sector Year Expenditure Revenue Defi it deficit deficit 1970 18.32 13.11 5.22 0.22 5.52 1971 18.24 12.85 5.39 0.26 5.71 1972 17.63 12.37 5.26 0.29 5.75 1973 14.88 11.81 3.07 0.59 3.71 1974 13.07 13.80 -0.73 0.43 -0.25 1970-74 16.43 12.79 3.64 0.36 4.09 1975 15.49 13.31 2.18 0.62 2.85 1976 17.76 12.73 5.02 1.24 6.30 1977 18.01 13.55 4.46 1.03 5.53 1978 18.83 13.57 5.26 1.78 7.12 1979 19.54 14.24 5.30 1.40 6.63 1980 21.27 14.94 6.33 2.62 8.99 1975-80 18.48 13.72 4.76 1.45 6.24 1981 21.70 14.61 7.09 2.55 9.60 1982 21.53 13.91 7.62 2.01 9.53 1983 21.12 15.38 5.74 1.51 7.20 1984 21.80 14.46 7.34 0.93 8.24 1985 23.15 14.57 8.58 1.69 10.24 1981-85 14.59 7.27 1.74 8.96 21.86 1986 20.84 14.28 6.57 0.95 7.46 1987 18.25 15.09 3.16 0.47 3.45 1988 16.56 16.79 -0.23 0.71 0.46 1989 21.63 18.42 3.21 0.26 3.86 1990 24.78 20.07 4.71 -0.07 4.94 1986-90 20.41 16.93 3.48 0.46 4.03 Source: National Economic and Social Development Board. Vatioinal inorme of ThaI/and. various issues. supporting import-substitution industrialization. Public expenditure, which was concentrated mostly on infrastructure, expanded moderately. This pattern was to change significantly after the 1973-74 oil price shock. For most of the period since 1970, the Thai public sector has been in deficit, running at an average of 3 to 4 per- cent of GDP. The exceptions were 1974 and the boom years after 1988, which 140 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 7.1. Government Expenditures and Revenues, 1970-91 Percent of GDP 21 20 19 18 1 7 16 15 14 13 12 1111- - 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Fiscal year --- Expenditure (G)YGDP - Revenue (R)/GDP Source: NESDB, National Income of Thailand, various issues. brought budgetary surpluses. From an examination of the size of government ex- penditures in relation to revenues since 1970, expressed as proportions of GDP (fig- ure 7.1), and in relation to GDP growth and inflation (figure 7.2), it appears that the years of budgetary surplus (largest deficits) coincide roughly with years of most rapid (slowest) income growth. A more systematic test of this proposition appears below. Between 1970 and 1974 the ratio of public expenditures to GDP declined from 18 percent to 13 percent. Public revenue as a percentage of GDP also declined, though at a slower rate than expenditure. and thus the ratio of the deficit to GDP declined as well. The world commodity boom of 1973, together with the subsequent oil price shock, raised revenues. High prices of commodity exports and high domestic inflation caused tax revenues to increase substantially (by 28 percent in 1973 and 48 percent in 1974). But high domestic inflation thwarted the plan to expand public expenditure, holding the expansion to the levels of 7 and 11 percent in 1973 and 1974, respectively.2 These changes resulted in a public sector surplus in fiscal 1973-74 and sub- stantially improved the financial position of the public sector. The government took this opportunity to reduce indirect tax rates so as to lessen their inflationary effect on production costs and incomes. From 1973 to 1976, a series of tax reduc- tion schemes was implemented to alleviate the burden of inflation and to offset the slowdown in GDP growth. Notwithstanding the tax-cut policy, tax revenue as a pro- portion of GDP rose by two percentage points in 1974, thereby producing a bud- getary surplus at the same time that Thailand was experiencing its most rapid consumer price inflation since the 1940s. Thie Role of Fiscal Policv 141 Figure 7.2a. Public Sector Deficit and Real GDP Growth Rate, 1970-90 Percent 15 13 I ' 9 7 -3 - 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Real GDP growth rate ---------- Deficit/GDP Figure 7.2b. Public Sector Deficit and Rate of Inflation, 1970-90 Percent 25 20 '5 0 -5 I I I I I I I I III I I I I I I I I _ 1970 1972 1974 1976 1978 198( 19X2 1984 1986 1988 199( Rate of inflation ---------- Deficit/GDP Source: Tables 3.4 and 7.1. After 1975 public sector expenditure increased signiticantly, particularly in the areas of defense and administration, which in 1976 and 1977 saw the average annual rate increase by 29 and 13 percent. respectively. Capital expenditure by the state enterprises increased even faster, at annual r ates of 21 percent in fiscal 1975, 35 percent in fiscal 1976, 61 percent in fiscal 1977. and 59 percent in fiscal 1978. As mentioned in previous chapters, these increases in consolidated public sector expenditure were facilitated by regulatory changes that increased the de- 142 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth fense loan allocation and authorized the Ministry of Finance to give financial guar- antees on foreign borrowing by state enterprises. In aggregate, between fiscal 1975 and 1978 public expenditure rose at an average annual rate of 24 percent, while revenue rose 19 percent. In fiscal 1978, the public sector's cash deficit was 16 mil- lion baht, or 3.3 percent of 1978 GDP. The foreign indebtedness of the public sector had increased from $87 million in 1976 to $200 million in 1978. In 1975 the government decided to use an overt fiscal stimulus in an effort to counteract the slowdown in income growth resulting from the first oil shock. Public investment increased in 1975 and 1976, while domestic consumption demand was stimulated by tax reductions. By 1976 the deficit had expanded and GDP growth had returned to its preshock levels. A series of tax increases was then implemented in 1977 and 1979. The change in fiscal stance slowed the growth of the public deficit, despite the continued rise in public expenditure: from 13 percent of GDP in 1974 to 21 percent in 1980. From 1975 to 1980, the deficit-to-GDP ratio rose an average 4.8 percent a year, up from an average 3.6 percent over the period 1970-74. The public finance problems that followed the second oil price shock of 1979- 80 were more serious than those produced by the 1973-74 shock. In the second case, the shock was followed in 1981-82 by a slump in primary commodity prices, which subsequently led Thailand into a recession and significantly reduced tax revenue. Fur- thermore, international real interest rates rose significantly in the early 1980s, and to- gether with the expanded public debt from the late 1970s they served to increase the public sector's debt-servicing burden. The oil price shock caused domestic inflation to increase dramatically: it peaked at 25 percent during one-quarter of 1980, with an average rate for the year of 20 percent. By the anti-inflationary standards of Thai- land's economic managers, these rates of inflation were unacceptably high. The Ministry of Finance responded to the second oil shock by increasing the already high rates of subsidy contained in public enterprise prices and public util- ity charges. The subsidies were expected to help control inflation, even though their impact on the price level was a once-only event. That is, their impact on the rate of consumer price inflation could be sustained beyond the initial year of im- pact only if they were continually increased. The new subsidies greatly increased the transfers from the central government to the state enterprises. Between fiscal 1979 and 1982, the expenditure of the state enterprises continued to expand at a rate slightly higher than that of the central government, 21 percent for state enter- prises and 19 percent for the central government. From 1981 to 1985 the government's fiscal stance continued to be expansion- ary, as reflected in a rising deficit, which reached 66 billion baht in fiscal 1982, or 7.9 percent of GDP. The highest level was 9 percent of GDP in 1985, up from an average of 4 percent in the 1970s. The enlarged public sector deficits of the early 1980s led to a serious balance of payments problem. It was clear that the public sector deficits had to be reduced. Accordingly, the period between 1980 and 1986 was marked by repeated cuts in planned expenditure and a series of tax reforms to boost revenue. But all this did little to close the deficit gap. There is no doubt that the government wished to reduce the 7he Role of Fiscal Polity 143 fiscal deficit, but this objective was not achieved. Seemingly, the slowdown in GDP growth was the main factor making revenue mobilization difficult. Public expenditure continued to grow faster than revenue, and the deficit widened correspondingly. The growth in consolidated public sector expenditure continued to be dominated by debt repayments, capital expenditure of state enterprises, defense, and administration. From 1987 on, Thailand's fiscal position changed radically. Restraints on spending, partly a product of the fiscal concerns of the early to mid- I 980s-and more particularly, the increased tax capacity caused by the abrupt acceleration of real GDP growth in the late 1980s-resulted in a strong fiscal position, allaying many of the deep concerns of only a few years before. To determine whether the behavior of Thailand's fiscal aggregates was stabi- lizing or destabilizing, it is necessary to examine the statistical relationships be- tween the main fiscal aggregates and variations both in aggregate income and in the price level. Table 7.2 shows the correlation between real public expenditure, real public revenue, and the real public sector deficit, on the one hand, and the cur- rent and lagged values of GDP growth rates and rates of inflation, on the other, for the period 1970-90. It is clear that the size of the budgetary deficit is negatively related to GDP growth. The above evidence suggests that Thai fiscal aggregates have behaved in a sta- bilizing manner. The next question to consider is whether this behavior was the re- sult of discretionary policy decisions, made in response to short-term fluctuations in real income, or an automatic outcome of the structure of the Thai expenditure and revenue systems. Discretionary or Automatic Fiscal Stabilizers? Thailand's fiscal history over the past two decades suggests that the short-term in- tentions of the fiscal authorities to expand or contraci the public sector deficit may have had only marginal effects on fiscal outcomes. Exogenous forces, determining Table 7.2 Correlation between Fiscal Variables, Income Growth, and Inflation GDP growth GDP growth Inflation Inflationi (current (previous (current (previous Change period) period) period) period) Total expenditure/GDP 0.32 0.56 0.009 0.03 Total revenue/GDP 0.69 0.68 -0.09 -0.20 Total deficit/GDP -0.70 -0.54 0.13 0.29 Note: Data cover the period 1970-90. Sourne: Authors' calculations from data in tables 3.4 and 7.1 144 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth income growth and short-term changes in rates of inflation, appear to have had a much greater short-run impact on fiscal aggregates. In the industrialized countries, both discretionary adjustments in government fiscal policy and autonomous adjustments in fiscal outcomes are considered poten- tially important sources of stabilization (see, for example, Dornbusch and Fischer 1992). The operation of the discretionary component depends on the way govern- ment policy responds to external shocks. Autonomous adjustments, the outcome of automatic stabilizers, operate through the structure of the tax and revenue systems themselves, and not through short-term discretionary changes in government policy. Textbook descriptions of automatic stabilizers emphasize the role of personal income taxes on the revenue side and welfare payments on the expenditure side. In the economies that make up the Organisation for Economic Co-operation and Development (OECD), personal income tax revenues are the dominant sources of revenue. These revenues rise as real incomes or rates of inflation rise, but in a greater proportion than the increase in nominal income because of the progressive income tax schedules. In contrast, welfare and social security expenditures rise as incomes fall because income relief is required most in times of recession. The result is that the public sector deficit declines during periods of rapid growth, as tax revenues rise and welfare expenditures decline as proportions of GDP. The public deficit rises in periods of recession as income tax revenues decline and welfare payments rise. As a consequence of these processes, the countercycli- cal behavior of the government's fiscal deficit plays a stabilizing role, reducing the impact of externally induced fluctuations in national income. Automatic stabilizers are seldom mentioned in the context of developing countries. This is not so surprising in one sense. It' personal income taxes on the revenue side and welfare payments on the expenditure side were truly the impor- tant variables to look for, one would not expect automatic stabilizers to be signif- icant at all for developing countries such as Thailand. Personal income taxes are only a small source of revenue in such countries, and, most of them, like Thailand, have no public social security systems. Thus it would seem that automatic stabi- lizers are unlikely to play an important role in developing countries. The present chapter will demonstrate that for Thailand, the reverse is true: au- tomatic stabilizers are important on both the expenditure and the revenue sides of the public sector accounts. Indeed, they are far more important sources of short- run stabilization than discretionary changes in fiscal policy. To some extent at least, it is must be expected that tax revenues and expendi- tures will be responsive to short-term changes in economic activity and that some degree of automatic stabilization will result. If the government's forecasts of in- come growth and inflation are inaccurate-and they must inevitably involve some errors-then the actual levels of spending and taxation revenue will presumably be somewhat different from their planned levels. As a result, the planned level of deficit will be different from the actual level. An unexpected expansion (slow- down) of economic activity presumably causes tiscal planners to underestimate (overestimate) future revenues. The Role ofFisc-al Policy 145 Because planned expenditure of the Thai government is constrained by fore- cast revenue, through the operation of the 1959 Budgetary Law (see below), the size of the actual budget deficit will then be anticyclical. In boom years. the planned level of deficit would tend to exceed the level that is finally experienced, if it is assumed that the boom was not fully anticipated by the planners. Similarly, during a slump, the planned level of deficit will be less than the actual deficit. This mechanism would produce a stabilizing feature from the revenue side of the bud- get. But how important is this effect in relation to discretionary policy changes? It is possible to study the behavior of fiscal intentions in relation to fiscal out- comes because the planned levels of revenue and expenditure are declared in the government's annual budget documents in advance of fiscal outcomes. The magni- tude of the planned deficit reflects discretionary fiscal policy, which reacts to changing macroeconomic policy variables such as deviation from the trend growth rate, the inflation rate, and the ratio of the current account deficit to GDP. The pur- pose here is not to study the way fiscal intentions are formed, but to compare fiscal intentions with fiscal outcomes and to relate both to income growth and inflation. We define the unplanned deficit as the difference between the planned and actual deficit. The unplanned revenue, unplanned expenditure, and unplanned deficit can be thought of as a result of the responsiveness of automatic fiscal stabilizers. Figure 7.3 shows the relationship between the planned and unplanned deficits, expressed as percentages of GDP. The underlying data are summarized in table 7.3. The government planned to run a budget deficit in each year between 1970 and 199(0, except for 1989. The surpluses that occurred in 1974, 1988, and 1990 were Figure 7.3. Actual, Planned, and Unplanned Deficits, 1970-90 Percent of oGDP -4 2 -6 1970) 1972 1974 1976 1978 198) 1982 1984 1986 1988 1990 --…---Unplanned Actual . I'lanned Source: Table 7.3 and author>' calculations. 146 Thailand's Macroeconomic Mirac le: Stable Adjustment and Sustained Growth the result of unexpected increases in revenue. The planned deficit, which reflects the discretionary intention of the government. apparently varied less than the un- planned deficit (table 7.4). The variability of the unplanned fiscal deficit/GDP ratio and the unplanned fiscal impulse exceed that of the corresponding planned vari- ables. Stabilization can now be examined over the short and intermediate terms. Short-Run Stabilization The first point to note is the relationship between year-to-year variations in fiscal policy and year-to-year fluctuations in income, inflation, and the external balance. Expansionary fiscal policy is taken to mean that the ratio of the deficit to GDP is Table 7.3 Central Government Budget: Planned versus Actual Level, 1970-90 (millions of bahf) Fiscal Ex:penditure Revenue Deficits/expenditure year Planned(G,2)a Actual(Ga)b Planned(R,,)a ActuaI(Rb,) I(Gp-Rp)1/Gl f(G,-R,)/GaI 1970 27,300 23,617 19,020 17.909 0.30 0.24 1971 28,645 26,978 21.800 190(88 0.24 0.29 1972 29,000 28,905 21,700 21.165 0,24 0.29 1973 31.600 30,937 23,300 25.344 0.26 0.18 1974 36,00)0 34.629 26,520 37,921 0(.26 -0.10 1975 48,000 43,541 38,500 38,229 ().20 0.12 1976 62,650 53,686 48,675 42,731 0.22 0.20 1977 68,79() 63,470) 50,470 51,710 ().27 0.19 1978 81,000 74,716 62.000) 62,022 0.23 (0.17 1979 92.000 86,157 72,000 75,109 (0.22 0.13 1980 114,557 114,287 92,68(0 92,147 (.19 0.19 1981 140,000 129,941 120,000) 11(0,329 (0,14 0.15 1982 161,000 152,168 140,00(0 113,810 (0.13 0.25 1983 177,000 165,100 151.00() 130,448 0.15 0.17 1984 192,0)00 177,402 160,0(0(0 147,847 (0.17 0.17 1985 213.000 197,468 178,000) 15,196 (0.16 0.19 1986 218,00)0 204,016 185.00() 16(.254 ().15 ().19 1987 227.50(0 207,817 185,50() 192,484 ).18 0.07 1988 243.50)0) 220,655 199,50(0 245,646 0).18 -0.11 1989 285.500) 266,310( 262,500 316,370 0.08 -0.19 1990 335,00(0 308,32(0 31(0,000 41)3,113(1 0.017 -(.31 a. Based on Thailand, Ministry of Finance, Thailand's Bud et in Brief, various issues. b. Based on Bank of Thailand, Monthlv Bulletitt tables 23 and 24. 7he Role of Fiscal Polici 147 Table 7.4 Statistical Properties of Fiscal Variables, 1970-90 (percent) Coefficient Variable Mean Minimum Maximum Variance of variation Revenue/GDP 14.71 12.00 20.00 4.01 0.14 Expenditure/GDP 16.74 13.00 19.70 3.10 0.11 DeficitIGDP 2.06 -5.20 5.00 7.86 1.36 Current expenditure/GDP 13.66 11.17 15.87 2.41 0.11 Capital expenditure/GDP 3.64 2.68 4.59 0.39 0.17 Actual fiscal deficit/GDP 2.07 -5.08 4.76 7.14 1.29 Planned fiscal deficit/GDP 3.22 -1.74 5.62 2.07 0.45 Unplanned fiscal deficit/GDP -1.15 -6.30 2.13 3.91 -1.73 Actual fiscal impulse -0.41 -4.12 3.05 3.34 -4.45 Planned fiscal impulse -0.22 -4.66 2.96 1.81 -6.11 Unplanned fiscal impulse -0.19 -4.44 3.30 4.70 -11.38 Sourre: Authors' calculations from data in tables 3.4 and 7.3. larger than in the previous year. Fiscal impulse is defined as the change in the ratio of the deficit to GDP, and an expansionary fiscal policy will thus mean a positive fiscal impulse. The unplanned fiscal impulse is defined as the difference between the actual and planned fiscal impulse. The relationship between fiscal impulses and important macroeconomic variables can be explored from the information provided in table 7.5. FISCAL STIMULUS AND AGGREGATE INCOME. In table 7.5 the variable "growth de- viation" denotes the difference between the actual and trend real GDP growth rates. A positive value of this variable indicates strong economic growth, and a negative value a slowdown. Years are classified into three groups according to economic growth performance: low, medium, and high. The behavior of these variables is clearer from figure 7.4. Deviations from the trend growth rate (normalized GDP growth rate) are negatively correlated with the actual fiscal impulse (figure 7.4, panel a). Fiscal impulse was stronger during pe- 148 Thailand s Macroeconomic Miracle. Stable Adjustment and Sustained Growth Table 7.5 Policy Goals and Budget Deficit, 1970-90 (percent) Growth Inflation CADb/ Budget deficit (percent of GDP) Year deviationa (CPI) GDP Actual Planned Unplanned 1970 -1.31 -0.09 3.53 3.13 5.62 -2.49 1971 -2.84 0.44 2.37 4.76 4.46 0.30 1972 -3.73 4.82 0.62 3.79 4.29 -0.51 1973 2.06 15.56 0.45 3.19 3.74 -0.54 1974 -3.45 24.33 0.64 -0.93 3.40 -4.32 1975 -2.95 5.30 4.08 2.12 3.13 -1.02 1976 1.58 3.78 2.59 4.48 4.03 0.45 1977 2.10 7.60 5.55 3.20 4.54 -1.33 1978 2.64 7.90 4.80 2.60 3.89 -1.29 1979 -2.49 9.91 7.62 2.38 3.58 -1.19 1980 -3.02 19.67 6.44 3.42 3.32 0.09 1981 1.47 12.70 7.37 2.55 2.63 -0.08 1982 -3.74 5.23 2.82 4.69 2.56 2.13 1983 -0.55 3.75 7.28 3.07 2.86 0.21 1984 -0.67 0.85 5.08 3.0)5 3.29 -0.24 1985 -4.29 2.43 4.13 3.69 3.45 0.24 1986 -2.88 1.85 -0.59 3.51 3.01 0.50 1987 1.67 2.50 0.74 1.14 3.35 -2.21 1988 5.42 3.80 2.69 -1.69 2.92 -4.61 1989 4.25 5.36 3.63 -3.60 -1.74 -1.86 1990 2.20 5.98 9.05 -5.08 1.22 -6.30 a. Deviation from real GDP growth trend (6.7 percent). b. CAD = Current account deficit Source: National Economic and Social Development Board. National Income of Thailand, various issues; Thailand, Ministry of Finance, Thailands Budget in Brief various issues; Bank of Thailand, Monthly Bulletin, tables 23 and 24. riods when the real growth rate fell below its trend path and weaker during eco- nomic expansion. A decomposition of the actual impulse into its planned and unplanned components reveals that the planned level of fiscal impulse (figure 7.4, panel b) does not exhibit an anticyclical pattern as strong as the unplanned fiscal impulse (figure 7.4, panel c). A striking negative correlation can be observed be- tween unplanned fiscal impulse and growth deviation. The key to this inverse re- The Role of Fisc al PolicY 149 Figure 7.4. GDP Growth Compared with Actual, Planned, and Unplanned Fiscal Impulse, 1971-90 Percent a. Actual impulse 6 4 2 0- -4 -6 - - 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 - Normalized real GDP growth rate - Actual fiscal impulse Percent b. Planned impulse 6 4 2 -2 -4 vv -6 - 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 -N Normalized real GDP growth rate -----Planned fiscal impulse Percent c. Unplanned impulse 6 4 2 -2 -4 -6 - - i 4 - i i i 1 i 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 -N Normalized real GDP growth rate ----- Unplanned fiscal impulse Source: Tables 3.4 and 7.3, and authors' calculations. 150 7hailand s Macroeconomic Mira( le: Stahle Adju.vtanlend and Sustained Growth lationship is clearly that tax revenues are more responsive than expenditures to changes in aggregate income (see table 7.2). The correlation between these variables is shown in table 7.6. Although it is obviously difficult to draw reliable inferences frotn short sample periods, the re- sults suggest that to the extent that short-term stabilization with respect to income occurred, it was due entirely to the role of automatic stabilizers. Planned fiscal im- pulses were weakly destabilizing with respect to current income. Clearly, to un- derstand Thailand's fiscal stabilization with respect to incotne, it is vital to consider the operation of automatic stabilizers.3 FISCAL STIMUL[JS AND THE PRICE LEVEL. The hehax ior of the actual fiscal impulse during inflationary periods can be observed in tigure 7.5. Again, there is a negative correlation between the two variables, almost as strong as that between growth de- viation and fiscal impulse (table 7.6). During inflationary periods with strong eco- nomic growth, tax revenues increased, capital expenditures declined, and the actual fiscal stance became less expansionar!. If inflation is associated with an economic slowdown, as it was in the case of the two oil price shocks, tax revenue collection will not increase sufficiently to reduce the fiscal impulse. This reduces the negative correlation between the fiscal impulsC and inflation. FISCAL STIMULUS AND EXTERNAI. BALANCE. It has been argued that in Thailand the public deficit drives the current account deficit. According to the monetary approach to the balance of payments, excessive growth in the money supply cre- ates an increased demand for imports as an outlet for the excess money supply. In the absorption approach, if aggregate demand or domestic absorption rises faster than output, net exports will be negative. Thus, if a public deficit stimu- lates aggregate demand, the current account will deteriorate if the income elas- ticity of demand for imports is high. Furthermore, if the public deficit produces pressure on the price level, net exports will dlecline because of the deterioration of price competitiveness. Table 7.6 Correlation between Planned and Unplanned Fiscal Variables, Income Growth, and Inflation, 1970-88 GDP growth GDP growth Inflation Inflation (current (previous (current (previous Fiscal impulse period) period) period) period) Actual -0.35 -0.38 -0.32 0.27 Planned 0.18 -0.(9 -0.25 -0.38 Unplanned -0.39 -0.37 -0.24 0.37 Source: Authors' calculations from data in tables 3.4 and 7 5. The Role of Fisc al Poulics 151 Figure 7.5. Inflation Rate Compared with Actual, Planned, and Unplanned Fiscal Impulse, 1971-90 Percent a. Actual impulse 25- 20 -- \ f 10- 5 - I-± _ __ 1972 1974 1976 1978 1980 1982 19X4 1986 1988 1990 - Inflation rate ---- Actual fiscal impulse b. Planned impulse Percent 25 20 -D 10 _5 4- _ + +--+ -t - -- I - 1972 1974 1976 1978 1980 1982 10)84 1986 1988 1990 -Inflation rate ----- Planned fiscal impulse Percent c. Unplanned impulse 25 T 20 -5- 4 +-- _ +, 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 - Inflation rate -----Unplanned fiscal impulse Source: Tables 3.4 and 7 3, and authors calculations. 152 Thailand s Macroeconomic Miracle: Stable Adjunvrment and Sustained Growth In Thailand's case, the income elasticity of demand for imports is high, and the current account deficit typically expands during economic booms. If the public deficit was driving these events, one would expect to observe a positive relation- ship between fiscal impulse and the current account deficit. The relationship be- tween the unplanned fiscal impulse and current account deficit as a percentage of GDP is illustrated in figure 7.6. Between 1970 and 1977, the relationship between fiscal deficits and exter- nal deficits was positive. The data for this period are consistent with fiscal defi- cits driving short-term outcomes on the current account. That is, that growth was led by public sector demand. After 1977, however, the relationship was clearly negative, owing to the fact that strong economic growth also generates higher tax revenue from business and sales taxes. Furthermore, increased imports also resulted in high tariff revenue. Since the strong economic growth was due to ex- port and investment growth rather than fiscal stimulus, the unplanned fiscal im- pulse moved inversely with the current account deficit. The fiscal impulse variable is affected by other macrovariables that also affect the current account deficit simultaneously. Again, it is evident that fiscal impulse does not increase the current account deficit. It provides a stabilizing feature with respect to external balance, since it tends to be weak when the current account deficit is large and stimulates aggregate demand during periods of current account surplus. Figure 7.6. Current Account Deficit and Unplanned Fiscal Impulse, 1971-91 Percent - t t t + - - -t - - t - t -t e -( t / -6 1971 1973 1975 1977 1979 19X1 19X3 1985 1987 1989 199 1 - Current account deficit/GDP ---- Unplanned liscal impulse Source: Table 7.5 and authors' calculations. The Role of Fiscal Polico 153 Table 7.7 Estimates of the Reaction Functions of Fiscal Policy Variables, 1972-90 Reaction function (Y, - Y1*) Y,1 din(P) CADIGDP R2 DW Da/GDP -0.30 -0.10 -0.08 0.80 2.1 0 (4.80) (2.10) (0.71) DP/GDP -0.14 -0.004 0.03 0.37 2.1 0 (2.90) (0.13) (0.28) D,IGCDP -0.19 -0.08 -0.12 0.48 2.1 0 (3.10) (1.50) (0.84) G,/GDP -0.004 0.076 -0.06 0.40 1.7 0 (0.33) (4.00) (1.40) R /GDP 0.14 0.16 0.20 0.41 2.1 0 (2.20) (2.60) (0.40) Note: t-statistics are shown in parentheses. Y* = trend income level growing at 6.7 percent; D = bud- get deficit: G = government expenditure; R = govemment revenue: and ).. = Da, Dp. Subscripts: a = actual; r> = planned: and u = unplanned. Source: Authors" calculatons from data in tables 3.4, 7.3, and 7.5. The regression results in table 7.7 relate the magnitudes of actual, planned, and unplanned fiscal aggregates to deviations from trend growth, inflation, and the ratio of the current account deficit to CDP. The results confirm that automatic fiscal stabilizers played an important role in Thailand. The coefficients of the ratio of the unplanned deficit to GDP are negative with respect to all explanatory variables, al- though not all are statistically significant in this equation. An unexpected boom causes an unplanned rise in revenue and a decline in the public deficit. The auto- matic stabilizing effect operates more strongly through the revenue side than through variations in expenditure. Inflation also causes a rise in unplanned revenue and expenditure, but its im- pact on revenue is stronger, and it causes the actual deficit to decline. The coeffi- cients of the current account/GDP ratio (CAD/GDP) are not significant in any of these equations; at most, there may be a weak negative relationship between the ratio of the unplanned deficit to GDP and the current account deficit. Our interpre- tation-namely, that the period falls into two discrete parts, in which the relation- ship between these two variables is reversed-is consistent with this result. 154 Thailands Macroeconomic Miracle: Stable Adjustmenit and Sustained Growth Internediate-Run Stabilization Two intermediate-run issues are of interest here. To what extent did planned defi- cits drive the fiscal adjustments in the mid- to late 1970s, on the one hand, and the early 1980s, on the other? As figure 7.1 indicates, the rising fiscal deficits that followed the first oil price shock were led by unplanned deficits, which began to rise after 1974 as a conse- quence of slowed income growth and accelerated inflation. Planned deficits also rose from 1975 to 1977, reinforcing the unplanned component. As income growth recovered from 1977 to 1979, planned deficits remained high, but unplanned def- icits declined, dominating the actual fiscal outcome, which was a decline in the ac- tual deficit from 1976 to 1979. Nevertheless, the intention of planning over this period, as reflected in the planned deficits, was consistent with the outcome: the public sector deficit was higher than it had been before the oil price shock. The period of the early 1980s is more controversial. The question is, when did Thailand begin to adjust to the growing extemal imbalance? From 1987 on, Thailand was enjoying an economic boom led at least in part-we would argue in the main- by external events. If adjustment was not demonstrably in place by then, some doubt must remain as to whether Thailand adjusted significantly at all. If not, then it might reasonably be thought that Thailand was simply lucky. It could then be said that Thai- land had borrowed its way out of the need to adjust to the terms of trade decline of the 1970s and early 1980s and that it was saved from the long-term consequences of this nonadjustment by an externally led boom beginning, fortuitously, about 1986 or 1987. From 1981 to 1984 planned public sector deficits were without doubt signif- icantly below their levels of the late 1970s (figure 7.3), but actual deficits were above their levels of the late 1970s. In the early 1980s the fiscal authorities were attempting a steady fiscal adjustment, but their efforts were clearly defeated by the operation of automatic fiscal stabilizers, responding to the slowed income growth of the first half of the 1980s, and the inflation of the early 1980s. It would be cor- rect to say that in the early 1980s the intention of fiscal policy was to adjust to the growing external imbalance, but here as elsewhere it is essential to distinguish be- tween the intention of policy and its actual outcome. The evidence provides little support for the claim that fiscal adjustment, defined to mean fiscal outcomes rather than fiscal intentions, began significantly before 1986 or 1987. Thus it seems that the fiscal expansion of the early to mid-1970s stimulated growth but that the short-term and intermediate-term fiscal outcomes observed thereafter were primarily a consequence of growth, caused by other more power- ful economic forces, rather than a cause of it.4 Sources of Stabilization If the actual fiscal adjustments have been stabilizing, and automatic stabilizers were the principal mechanism through which this occurred, what, then, is the Thul Role oq Fiscal PoIcvl 155 source of these stabilizers? Do they originate on the revenue or the expenditure side of the budget, and from which of their components? THE DETERMINATION OF PUJBLIC EXPENDITURE AND REVF NUE. The Thai budgetary process requires expenditures to be planned a year in advance of actual outlays. As already mentioned, budgetary regulations restrict the size of planned expendi- tures as a proportion of planned revenues. This system seems likely to produce a relationship between actual expenditures in the current period and previous period revenues. The latter are known at the time expenditures must be planned and ac- tual revenues in year t - I are the principal basis for determining planned revenues in year t. We used a log-linear regression analysis based on data from 1970 to 1990 to investigate this relationship. The estimated elasticities of real expenditures with respect to real previous period revenues are summarized in table 7.8. The positive elasticity of response indicates that as planned (previous period) revenues rise in real terms, expenditures also rise: but the elasticity of less than unity suggests that expenditures rise in a smaller proportion than the increase in revenues. That is, as incomes rise. the deficit contracts. We wish to go beyond this analysis, however, because the estimated relationship only partly explains the hehavior of expendi- tures and, of course, it does not explain the behavior of revenues. To study the responsiveness of government expenditure and revenue to changes in income and the price level, we regressed real expenditures and real rev- enues on real GDP and the nominal GDP deflator (table 7.9). Our estimate of the elasticity of total tax revenue with respect to real GDP is 1.4, which talls within the range found in earlier studies, but at its upper end (table 7.1(0). This estimate is also significantly greater (at the 5 percent level of significance) than our estimate of the income elasticity of total real expenditures (0.5). The implication is that as in- comes rise, the gap between aggregate expenditure and aggregate revenue nar- rows. The real fiscal stimulus declines. Table 7.8 Public Expenditure Elasticities and Tax Buoyancy, 1970-90 Ela. ticitv Expenditure Previous period revenue GDP deflator Current 0.82 0.29 (3.68) (0.94) Capital 0.36 -2.53 (0.59) (-2.37) Total public 0.75 -0.86 (5.79) (2.25) Note: Numbers in parentheses are t-statistics. Source: Calculated from Bank of Thailand, Monthly Bulletin. various issues, tables 23 and 24. 156 Thailand's Macroeconomic Miracle: Stable Adjustmnent and Sustained Growth Table 7.9 Elasticities of Real Public Expenditures and Revenues with Respect to Real GDP and GDP Deflator, 1970-90 Elasticity Real aggregate Real G'DP GDP deflator Revenue I1.40 -0.16 (12.5) (1.8) Expenditure 0.48 -0.86 (1.02) (2.25) Note: Numbers in parentheses are t-statistics. Source: Calculated from Bank of Thailand, MonthlA Bulletin. various issues, tables 23 and 24. A similar point applies to the effects of inflation. Inflation seems to have no ef- fect on public revenues in real terms, but it has an important negative impact on pub- lic expenditures, as indicated by the significance of the GDP deflator. The estimated magnitude of the elasticity (-0.86) indicates that as the price level rises nominal ex- penditure remains almost constant.5 These findings suggest that the behavior of Thailand's fiscal aggregates contributes to price stability. When the price level rises, the government does not insist on acquiring the intended level of expenditures. in real terms. As a result, during an economic boom that places upward pressure on the price level, there is less inflationary pressure from the public sector. Table 7.10 Summary of Previous Tax Elasticity Estimates Tax elasticities Period of Personal Corporate Researcher estimation Total income income Business World Bank (1974) 1963-73 1.4 1.2 1.1 1.2 World Bank (1978) 1970-77 1.3 - - - World Bank (1983) 1963-81 1.0 1.4 1.3 1.0 IMF (n.d) 1975-78 1.1 1.3 1.2 1.2 IMF (n.d) 1978-81 0.9 1.4 0.8 0.8 IMF (n.d) 1975-81 0.9 1.3 1.0 0.9 Chanchai (1983) 1972-82 1.0 1.1 1.3 1.1 Chompleon and others (1982) 1961-79 1.0 1.3 1.7 1.2 Ministry of Finance (1984) 1974-83 1.0 1.9 ().9 0.9 Yuttaphon (1984) 1974-83 1.0 1.8 1.0 0.9 Source: Rungsan Thanapoman (1985), cited in Medhi and others (1988). table 2,10. The Role of Fiscal Policy 157 In summary, our results confirm that the Thai fiscal system has a stabilizing feature with respect to both real income and the price level. First, although expen- ditures are insensitive to collected revenues, the latter are sensitive to fluctuations in aggregate economic activity. Thus. during boom years revenues rise faster than expenditures, and in recessions the reverse occurs. Second, inflation reduces real expenditures, but it leaves aggregate real tax revenues virtually unchanged. Thus when inflation accelerates, the real magnitude of the budgetary deficit declines (the budgetary surplus increases) and further inflationary pressure subsides. The sources of these mechanisms can be found among the components of aggregate expenditure and revenues. THE COMPOSITION OF EXPENDITURE. From 1970 on, current expenditures-on sal- aries, the use of consumable materials, and other recurrent costs associated with government activity-accounted for an average of about 78 percent of total gov- ernment expenditure (see table 4.2). Capital expenditures. those associated with increasing the public sector capital stock, averaged about 22 percent of total ex- penditures. But these proportions varied considerably, as did the share of total ex- penditure in GDP, which fluctuated between 15 and 2(0 percent of GDP. There was some variation in both current and capital expenditure, but capital expenditure was the more volatile of the two (see table 7.4). The central government's spending increased most rapidly during the expan- sionary period between 1975 and 1980, when it grew at the rate of 23 percent per year in real terms (table 7.11). The smallest increase occurred during the deceler- ation period of 1986 to 1990, when it expanded at an average rate of 7 percent. The share of capital expenditure in total expenditures decreased over time, from 26 percent during 1971-74 to only 14 percent during 1986-90. Thus the role of the central government in providing infrastructure diminished substantially during the 1980s. The budget share allocated to defense averaged 19 percent, but the rate of change of defense expenditure fluctuated widely. During years of fiscal expansion, it grew at an average rate of 22 percent, but this rate was cut to only 2.7 percent during the tight fiscal years of 1986 to 1989. The top priority of the budget is social services-such as hospitals, schools, and police. Although the share of adminis- tration gradually declined over time, the importance of the expenditure on eco- nomic services fell more significantly: from 24 percent in the early 1970s to only 14.6 percent in the late 1980s. The slowdown in public expenditure on infrastruc- ture was the result of the attempt to restrain the rise in public spending. Its effect was to reduce the size of public capital stock in relation to national output. Apparently, when an expenditure cut is desired, it is more expedient to cut capital expenditure than current expenditure. It has been easier to postpone large public investment projects than to delay increases in civil servants' salaries during inflation. Although the government has been able to freeze the wages of public employees from time to time-as it did from 1982 to 1989-the controllability of 158 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 7.11 Changing Structure of Government Expenditure, 1971-90 (percentage c hange, period average) Expenditure 1971-74 1975-80 1981-85 1986-90 Total 8.9 23.1 10.6 7.4 Economic classification Current 13.9 21.9 12.2 8.0 Average share (74) (76) (82) (86) Capital -5.0 28.9 4.9 4.1 Average share (26) (24) (18) (14) Majorfunctional classification Economic services -2.0 24.7 6.1 7.7 Average share (24) (22) (17) (15) Social services 12.4 22.7 11.0 8.1 Average share (28) (30) (30) (30) Defense 13.6 22.8 12.4 2.7 Average share (19) (19) (20) (19) General administration 11.4 22.4 8.7 8.7 Average share (14) (13) (13) (13) Unallocatable items 15.7 26.4 15.5 10.0 Average share (14) (14) (20) (23) Source: Comptroller-General's Department; Bank of Thailand. public consumption spending is ordinarily subject to political influences and infla- tionary expectations. The above arguments suggest that the fiscal spending structure, or the alloca- tion of government expenditure between consumption and investment, depends on the overall fiscal policy stance. If the budget was contractionary, public investment spending would rise proportionately less than public consumption, and vice versa. In figure 7.7, it is clear that the ratio of public investment to public consumption is correlated with the ratio of the public deficit to GDP. From 1970 to 1974, when fiscal policy was tight, the ratio of investment to consumption within public ex- penditure declined sharply. When easier fiscal policies were implemented, be- tween 1975 and 1985, the investment-to-consumption ratio rose rapidly. The ratio of public and private investment to GDP exhibits a striking inverse relationship (figure 7.8), which suggests that public investment may exercise a sta- bilizing role with respect to aggregate investment by maintaining a high and stable proportion of total capital formation in relation to income.6 Table 7.12 summarizes Thte Role of Fiscal Policy 159 Figure 7.7. Fiscal Structure and Public Deficit, 1970-91 Percent 6 -2 -4 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 - Capital expenditure/current expenditure (Deficit/iDP)* 10 Source: Bank of Thailand. Figure 7.8. Public and Private Capital Accumulation, 1970-91 Ratio to GDP 0.32 0.28 0.24 Private 0.20 0.16 0.12 0.08 ,,-- __ _ _sx ,, ~ ~ ~~~~~ Public" 0.04 , 0.00 I -------I I I -- I I - I - 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Source: NESDB, National Income of Thailand, various issues. 160 Thailand's Macroeconomic Miracle: Stable Adjustmnent and Sustained Growth Table 7.12 Growth, Inflation and Capital Formation, 1971-90 (percent) Rate of Xchange of Real GDP Inflation real invesment Year growth (CPI) Private Public 1971 4.96 0.25 1.45 -2.09 1972 4.07 5.05 -0.43 4.46 1974 4.35 24.37 6.41 -30.32 1975 4.85 5.36 -8.85 42.90 1980 4.78 19.76 -2.59 18.88 1982 4.06 5.24 1.38 -8.54 1985 3.51 2.42 -11.19 7.57 1986 4.92 1.81 0.51 -11.58 Low growth 4.44 8.03 -1.41 2.66 1979 5.31 9.87 5.13 1.13 1981 6.33 12.70 3.71 6.35 1983 7.25 3.71 13.45 11.50 1984 7.13 0.89 10.62 11.65 Medium growth 6.51 6.79 8.24 7.66 1973 9.86 15.38 21.52 -17.49 1976 9.88 4.13 3.87 45.14 1977 9.90 7.48 26.30 18.21 1978 10.44 7.95 7.38 21.10 1987 9.47 2.55 29.22 -13.07 1988 13.22 3.84 32.01 -9.39 1989 12.05 5.36 32.07 13.17 1990 10.00 5.98 24.24 35.59 High growth 10.60 6.58 22.08 11.66 Source: National Economic and Social Develpment Board. National Income oJ Thailand. various issues. the data on this point, dividing years into low, medium, and high growth groups. When a private investment boom occurred in 1974, public investment declined. Although private investment remained sluggish from 1979 to 1985 after the sec- ond oil price shock, public investment expanded. Again, during the private invest- ment boom of 1987 to 1990, the ratio of public investment to GDP declined significantly. Tih Role of Fiscal Polui 161 From table 7.8 it is apparent that public investment expenditure is highly sen- sitive to inflation. The stabilizing effect of public expenditure behavior with re- spect to the price level clearly derives from the behavior of public investment spending rather than public consumption. The large negative value of our estimate of the elasticity of real public investment expenditure with respect to the GDP de- flator (-2.53, which is significantly different from zero at the 5 percent level of sig- nificance) implies that increased inflation causes nominal, and not merely real, public expenditure on capital goods to decline when inflation rises. But the elas- ticity of current expenditure was not significantly different from zero, as indicated by the low value of its t-statistic. The economic process underlying this result is that public investment projects can he delayed in real terms if the nominal cost of capital is rising. Capital expen- ditures are not fixed in real terms. Actual expenditure will be lower than the planned level, even in nominal terms, if the allocated budget cannot cover the ris- ing cost of capital. If the budgeted funds are insufficient to purchase a particular expenditure item, they will not be spent at all in that fiscal year. It might he hypothesized that the inverse relationship between public and pri- vate investment was the result of a crowding out ot public investment, which raised interest rates and discouraged private investment. As will be shown later, however, borrowing from the nonbank private sector constituted a small portion of deficit financing. A large part of the deficit was money financed and thereby cre- ated a favorable condition for private investors since there was no upward pressure on the interest rate. That relationship appears to arise i'rom the sensitivity of public investment spending to inflation in the prices of capital goods, principally those used in the construction industry. Private investment is highly volatile. When it in- creases, pressure is placed on the prices of capital goods and materials used in con- struction. This has a powerful negative effect on public investment. THE EFFECTIVENESS OF LEGAL EXPENDITURE LIMITS. The 1959 Budgetary Law stipulated that the excess of planned spending over planned revenue should not ex- ceed 20 percent of the level of planned spending. The law thus implied that the planned government budget deficit was not to exceed 25 percent of the planned revenue.7 In 1973 the limit on spending was relaxed somewhat when it was an- nounced that the deficit was not to exceed 20 percent of planned expenditure plus 8 percent of the principal repayments of the public debt. The effectiveness of the budgetary law in restraining the growth of public ex- penditure can be seen in figure 7.9. The maximum amount of public spending is calculated by multiplying the level of planned revenue by 1.25 and adding 8 per- cent of the level of expenditures allocated to principal repayments of public debt. The figure shows ratios of planned expenditure (CG,l and actual expenditure (Ga) to the maximum expenditure permitted by the hudgetary law (G,,). Since the ac- tual level of spending was always below the planned level, G,I/,n was less than G,IGm. Except during 1972-74, the actual level of expenditures was well below the maximum permitted level. 162 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 7.9. Effectiveness of Budgetary Law, 1973-88 Percent of maximum expenditure 120 100 i___/ ___- -- -- -------------__ _ _ -_ 80 60 40 - Planned/Maximum -- - -Actual/Maximum 20 0 1 1-4--4 1973 1975 1977 1979 1981 1983 1985 1987 Source: Table 7.3 and authors' calculations. ADJUSTMENT OF GOVERNMENT REVENUES. Since 1970 the government has been able to increase its taxing capacity gradually, as illustrated in table 7. 1, thereby raising the ratio of tax revenue to GDP. It was not only the fiscal stance that was changed in response to the long-term effect of external shocks. The structure of tax- ation was also altered. Direct taxation became increasingly important (see table 4. 1). The share of income taxes in total taxation rose from an average 13 percent in the period 1971-74 to 22 percent in the period 1986-90. Personal income and cor- poration taxes were equally important, at about 10 percent of total tax revenue. The share of sales taxes also rose from 19 to 28 percent over the same period, while the share of business tax in total taxation remained at roughly 20 to 22 percent. The share of import duties in total tax revenue declined from 29 percent in the early 1970s to 22 percent in the late 1980s. Since the share of export taxes in total rev- enue was insignificant, the importance of foreign trade taxes declined. This implies that the feedback effect of external shocks to the government's fiscal position has di- minished over time. In some developing countries whose main sources of revenue are taxes on foreign trade, an abrupt change in the external terms of trade has strong re- percussions on the government's fiscal position. This is less true in Thailand. An examination of the statistical relationship between the components of total real tax revenues, on one hand, and aggregate real income and the GDP deflator, on the other, indicate that the buoyancy of corporate income taxes, import duties, and 7 /,it Role of Fiscal PolicY / 63 business taxes increased significantly in the 1980s (table 7.13). Although the buoyancy of personal income tax revenues declined slightly, possibly because of the reduction in the income tax rate, it remained large. The increase in tax buoyancy, as measured by time-series data, was partly the re- sult of long-tern structural changes within the Thai economy. The shift in the com- position of output away from agriculture and towar-d the manufacturing and formal services sectors brought a rising proportion of econl)mic activity within the tax base. The rise in the share of imports in total demand also led th l rising tax revenues because imports are taxed on average more heavily than domestically produced goods. More- over, during an economic boom, imports rise proportionately more than GIDP because the income elasticity of demand for imports exceedis unity. The rise in imports gener- ates import duties. The expansion of the tax base to cover activities that are sensitive to business cycles, together with more efficient tax collection, also played a role. THE CHANGING M4EANS OF DEFICIT FINANCE. The methods of deficit financing changed significantly between 1970 and 1990. The government increasingly re- lied on borrowing from domestic rather than foreign sources (table 7.14). Financ- ing by the use of treasury cash balance also (leclined during the same period. The share of foreign borrowing in deficit financing was again exceptionally high. at 36 percent of the total deficit during the economic slimp of 1985, a year in which Table 7.13 Tax Buoyancy, 1970-90 (perc ent) 1970-79 1980-90 1970-9() Tar anld GDP 0)1' GDiP duties Real GDP deflator Real GDP deflator Real GDP deflator Personal income 2.7 -1.1 2.2 -2.6 2.0) -0)39 (6.0) (3.2) (2.0) (()89) (5.1) (0.82) Corporate income 0.51 1.6 1.78 2.9 1.4 1.3 (0.95) (3.8) (2.03) (1.6) (2.7) (2.3) Import 1.64 -0.73 2.26 -1.32 2.1 -1.1 (4.8) (2.7) (3.5) ((0.96) (14.8) (7.3) Business 0.81 0.37 2.86 (0.02 1 .6 -0.24 (1.65) (0.96) (4.7) (0.24) (5.8) (0.8) Sales 0.18 0.26 ().81 ().98 1.45 -0.78 (0.15) (0.27) (0.98) i 0.56) (4.4) (2.2) Note: Numbers in parentheses are t-statistics. Sourme: Calculated from Bank of Thailand, Monthl\ Bulletin, various issues, tables 23 and 24. 164 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 7.14 Financing the Public Deficit, 1970-90 (percent of total deficit) Net other Use of Net domestic Net foreign liabilities Treasury cash Year borrowing borrowings of Treasury balances 1970 73.9 0.1 4.1 21.9 1971 74.3 0.6 12.0 13.1 1972 113.4 -1.0 6.1 -8.6 1973 85.3 2.4 36.0 -23.8 1974 -67.5 0.5 76.9 -109.9 1970-74 55.9 0.5 27.0 -21.5 1975 37.8 -1.5 31.1 32.5 1976 69.0 -0.2 19.9 11.4 1977 88.0 1.7 4.7 5.6 1978 109.3 -4.9 3.2 -7.6 1979 98.0 -7.1 1.6 7.5 1980 102.5 -4.2 4.8 -3.1 1975-80 84.1 -2.7 10.9 7.7 1981 104.5 -9.5 4.3 0.1 1982 105.9 1.9 -6.5 -1.3 1983 111.2 4.1 -17.1 1.8 1984 108.2 -2.2 -7.1 1.2 1985 81.1 36.4 -15.7 -1.8 1981-85 102.2 6.1 -8.4 0.0 1986 147.2 -18.9 -23.3 -4.9 1987 120.1 -36.9 -3.8 20.6 1988 -79.9 -12.6 1.3 8.9 1989 -17.1 -9.5 -7.1 -66.4 1990 -11.1 -36.6 -1.4 -50.8 1986-90 31.8 -22.9 -6.9 -18.5 Source: Computed from Bank of Thailand, Monthlv Bulletin, table 25. private investment declined by 11 percent in real terms. The large increase in foreign borrowing in 1985 enabled public investment to expand 8 percent and thus compen- sate for the shortfall in private investment. It is clear that the government has, in gen- eral, endeavored to avoid foreign borrowing except when it has been clearly The Role of Fis(al Policyv 165 necessary. Note, too, that foreign borrowing is subject to the legally imposed ceiling on the debt-service ratio (figure 7.10). Although the ceiling was adjusted in some years, its significance in restraining public sector foreign debt cannot be denied. Not only did the size of domestic borrowing to finance the deficit increase, its composition also changed. Reliance on inflationary means of deficit financing, such as borrowing from the Bank of Thailand, was reduced substantially from its high levels of the early l 970s (figure 7.1 1). In the period 1970-74, borrowing from the Bank of Thailand was 78 percent of total net domestic borrowing. It fell to only 25 percent during the period 1986-88. From 1970-74 to 1986-88, the govern- ment's borrowing from commercial banks increased from 16 to 29 percent of its total borrowing, and its borrowing from government savings banks increased from 3 to 40 percent of total borrowing. Through the 1 980s, not only the size of the budget deficit declined, but the de- pendence on inflationary means of financing it, that is, borrowing from the Bank of Thailand, was also reduced. The changing pattern of deficit financing clearly contributed to long-term price stability. Conclusions The short-run behavior of the public sector's fiscal system leads to a significant stabilization effect, which operates with respect to fluctuations in both income and Figure 7.10. Ceiling on Debt Service Ratio, 1970-88 Percent 25 T 20 j ,' ..... .. 0) -4 -I- d 197() 1972 1974 1976 1978 198( 1982 1984 1986 1988 -- --- Pri,ate ---- Public ---- Total Ceiling Rite Sourc e: Bank ol Thailand. 166 Thailand us Macroeconomic Miracle: Stable Adju tinent aind Sustained Growth Figure 7.11. Financing the Public Deficit, Domestic Sources, 1970-88 Proportion of total deficit 1.01 _ _ _ . 0.9 0.2 0.71 _ _ _ _ _ _ _ _ _ _ _~~~~~~~~~~~~~~~~~..... . ... 1 0. - - - -- -_ _ 0: 1 1970-74 1975-80 1981-85 1986-88 Bank of Thailand Commercial banks n71 Government savings banks * Others the price level. That effect does not arise from discretionary policy changes, how- ever, but from the unplanned component of the observed fiscal outcomes. The economic literature on automatic stabilization refers to the stabilizing role of income taxes and welfare payments through the social security system. When incomes rise, direct income taxes rise more than proportionately, and wel- fare payments fall. Thus, the size of the budget deficit falls. When incomes fall, the deficit rises. But these are characteristics of developed economies, not of de- veloping economies like Thailand's. The public finance literature would not lead one to look for automatic stabili- zation in Thailand's fiscal system. In 1990 personal income taxes accounted for only 10 percent of Thailand's total revenue. Company income taxes made up an- other 14 percent. Indirect taxes-import duties, business taxes, and sale taxes- were each far more important than these income taxes. Together, they accounted for 69 percent of 1990 revenues. But they are a source of the stabilization on the revenue side. On the expenditure side, Thailand has no social security system at all, but its rigid, bureaucratically driven budgetary process is a source of stabiliza- tion. Clearly, automatic stabilizers can be important for developing as well as de- The Role of Fiscal Policy 167 veloped economies, but the mechanisms through which they operate can be very different. Short-run discretionary changes in fiscal policy do not seem to have been sta- bilizing. If stabilizing fiscal policy is interpreted in this way, then it would be wrong to attribute Thailand's record of economic stability to this source. But pol- icy clearly had a crucial effect in a deeper and longer-ternm sense. Most obviously, short-term discretionary behavior was not significantly destabilizing. It could have been otherwise, and the instability experienced by so many other developing coun- tries may well reflect a significant difference in this respect. Equally important, the structural features of the Thai revenue and expenditure systems that produce the automatic stabilizing outcomes are themselves the prod- ucts of past policies. On the revenue side, this point is so obvious that it does not require elaboration, but on the expenditure side two features of the Thai public ex- penditure system should be emphasized. First, the determination of public investment expenditure has a built-in anti- inflationary feature. When rapid inflation occurs in the prices of capital goods re- quired for public sector capital formnation, public sector purchases of these mate- rials contract markedly in real terms. Construction materials are a prime example. The result is a significant negative correlation between private sector investment and public sector investment spending. The expenditure regulations that produce this outcome are of course the products of past policy decisions and reflect the anti-inflation bias which pervades the Thai policy enN ironment. Second, legal limits on planned expenditures date back to laws introduced during the Sarit government of the late 1 950s and early 1 960s. The important point is not that this caused short-run discretionary fiscal policy to be stabilizing-ap- parently, it did not. But it did contribute to fiscal discipline by limiting the overall magnitude of planned fiscal deficits and so, in effect, constrained the capacity of fiscal policy to be destabilizing, both in the short and the long term. From 1970 to 1990 the structure of Thailand's deficit financing shifted in- creasingly away from foreign borrowing and toward domestic borrowing. More- over, the composition of domestic borrowing shilted away from inflationary sources, such as borrowing from the Bank of Thailand, and toward domestic sources that do not imply monetary creation. Nevertheless, the share of borrowing from the Bank of Thailand in total deficit financing remained substantial over the entire twenty-year period. It follows that fiscal discipline was an indispensable component of long-term inflation control. The monetary policy components are discussed in chapter 8. Chapter Eight The Role of Monetary Policy The record of Thailand's monetary institutions has been impressive. Controlling inflation has been the principal mandate of the central bank, the Bank of Thailand, and it has accomplished this to an extent matched by few others.' Inflation has been subdued largely through long-term monetary restraint, which has made it possible to achieve a stable exchange rate combined with adequate levels of inter- national reserves and only moderate growth of the external debt. Another of the bank's mandates is to promote the short-term stabilization of income growth and external balance. According to the Bank itself (Bank of Thai- land 1992), a discretionary, countercyclical monetary policy directed to this end has been successfully implemented. Whether this interpretation of Thailand's monetary experience is supported by the evidence is the central question explored in this chapter. The discussion focuses on four main topics: the autonomy or otherwise of Thailand's monetary policies, in the presence of its fixed exchange rate policy; the components of Thailand's monetary base and the degree to which the Bank of Thailand can be said to control them; the manner in which the major instruments of monetary policy have been manipulated by the Thai monetary authorities in re- sponse to external shocks; and the effectiveness of these interventions in stabiliz- ing income growth and inflation in the short run. Thailand has followed a fixed exchange rate policy for several decades. The baht has been pegged in relation to the U.S. dollar since the 1 950s except for a few, moderately sized devaluations in the early 1980s (discussed in chapter 9). But the fixed exchange rate implies that the monetary authorities cannot fully control Thailand's money supply. Under a fixed exchange rate, Thailand's monetary base would be affected by the balance of payments as a result of the Bank of Thailand's intervention in the foreign exchange market. To maintain a fixed exchange rate, the bank must stand ready to buy or sell unlimited quantities of foreign exchange at that rate, and these transactions will affect the domestic money supply in a way that the bank cannot control. /68 Thie Role of Monetarn Policy 169 Conventional wisdom therefore dictates that, provided the capital account is open, attempts to combine a fixed exchange rate with a discretionary monetary pol- icy will fail. The claim that Thailand has successfully combined these two therefore appears dubious at best. Yet the evidence supports the view that weakly countercy- clical discretionary monetary policies have indeed been pursued successfully in Thailand. Regulations that restrict free capital movements have enabled the mone- tary authorities to pursue some degree of monetary independence in the short run, as is evident from significant divergences between domestic and foreign interest rates in the short run. Moreover, these divergences can be related to changes in monetary policy. To some extent, therefore, the Thai monetary authorities can be said to pos- sess a degree of short-run autonomy. The effectiveness of their monetary policies is aided by the stable relationship between money supply and the monetary base. but limited by offsetting capital flows. The main tools of monetary policy appear to be changes in the central bank lending rate, direct regulation of interest rates, direct regulation of volumes of credit, interventions in the bond repurchase mar- ket, and direct controls on capital flows. These monetary instruments appear to have been applied in a manner that has, on balance, been countercyclical. Economic stabilization is but one of the Bank of Thailand's objectives. It is also concerned with such development objectives as rural development, export promotion, and maintaining the stability of financial institutions. Sometimes, those development goals are inconsistent with price stability, since they require the Bank to provide credits for regional and sectoral development, for export pro- motion, and for maintaining the liquidity of ailing financial institutions. The Bank must compromise between these conflicting goals in determining the appropriate level of credit growth. The analysis of monetary policy in this chapter concentrates on the period 1970-90. But changes in Thailand's regulatory environment since 1990 will great- ly change the future operation of Thai monetary policy. This issue is also given some attention in the chapter. Autonomy of Mionetary Policy With a fixed exchange rate and an open capital account, Thailand should, in theo- ry, not be able to operate an independent monetary policy. According to the Mun- dell-Fleming hypothesis, under a fixed exchange rate policy, a monetary contraction, by raising domestic interest rates, would induce a capital inflow and force interest rates downward again. Monetary expansion would do the reverse. If applicable, this hypothesis would imply that domestic interest rates would move together with international rates and be outside the control of the Thai monetary authorities. As this chapter explains, the key to the apparent contradiction is that the Thai capital account has not been fully open. 170 Thailand s Macroeconomic Miracle: Stable Adjustmenr aind Sustained Growth Regulations Restricting Free Capital Movements There are four reasons for a lack of openness in the period 1970-90: interest rate ceilings, direct capital controls, controls on the foreign exchange position of com- mercial banks, and the withholding tax on foreign borrowing. INTEREST RATE CEILINGS. Regulatory ceilings on domestic interest rates, both lending and borrowing rates, prevented the domestic interest rate increases that would otherwise have induced capital inflows when the domestic money supply was contracted. These interest rate ceilings were adjusted only slightly during shocks. From 1970 to 1981, the rates on commercial bank deposits were equal to the ceiling rates, and from 1982 to 1985 the actual time deposit rates of interest paid by commercial banks remained 0.5 percent below the ceiling. DIRECT CAPITAL CONTROLS. Bank of Thailand permission has been required to move capital out of Thailand, and this policy has been policed vigorously. During periods of monetary expansion, this policy enabled the monetary authorities to prevent the outward flow of capital that would otherwise deprive them of the ca- pacity to expand the domestic money supply when desired. All outgoing payments were subject to approval. Until 1990, exporters were required to submit foreign exchange currency to banks within seven days alter receiving payments from abroad.2 Until 1990 Thai citizens were not permitted to hold foreign exchange deposits or to purchase foreign currencies for investment overseas. Thus they were unable to take much advantage of differentials between domestic and foreign rates of in- terest. Individuals were not permitted to take out of Thailand domestic currency exceeding 500 baht (equivalent to about $20)) or foreign currency exceeding $1,000. In 1993 these limits were 50,000 baht (equivalent to about $2,000) and $ 10,000 of foreign currency. The degree of substitution between domestic and for- eign assets was far from perfect; similarly, foreign liabilities were imperfect sub- stitutes for loans obtained from domestic banks, since domestic deficit units could not gain access to foreign capital markets. The government also imposed a limit on the volume of the foreign debt of public enterprises. CONTROLS ON FOREIGN EXCHANGE POSITIONS 01( COMMERCIAL BANKS. Since 1984, the net foreign exchange positions of commercial banks have also been sub- ject to regulation. Following the 1984 devaluation, the net future and current po- sition of each commercial bank-whether positive or negative-could not exceed $5 million in either direction or 20 percent of the net worth of the bank, whichever was smaller. In April 1990 the ceiling on the net position of commercial banks was raised to 25 percent of capital funds. WITHHOLDING TAX ON FOREIGN BORROWING. A withholding tax is applied to for- eign borrowing at rates that are varied by the Minister of Finance to encourage or The Role of Monetarv Polic v 171 discourage foreign capital inflows. This instrument creates a tax wedge between the domestic and foreign costs of capital. When the government wanted to reduce capital inflows, the withholding tax rate was usually set at 10 percent of the inter- est payments. It was exempted when the government considered the domestic money market too tight. Exemption was sometimes granted for loans with long maturity periods, such as one to three years, to attract long-term capital funds. As table 8.1 shows, the withholding tax rate has been varied significantly from time to time. The adjustments have been directed at influencing capital flows for stabi- lization purposes. The above regulations have restricted free capital movements. Thus even if the exchange rate is fixed, scope may remain for using monetary policy in a dis- cretionary manner. If the restrictions on capital mobility are sufficient to make this outcome possible, domestic and foreign interest rates should be imperfectly cor- related. More important, divergent movements between these two sets of interest rates should be traceable to changes in Thailand's discretionary domestic mone- tary policies. Interest Rate Divergence Figures 8.1 and 8.2 provide some background on the long-term relationships be- tween Thai and foreign interest rates. Figure 8.1 shows the annual average money market rates of interest in Thailand. the United States, and the three-month Lon- don interbank offer rate (LIBOR). The U.S. and LIBOR rates move closely together. The Thai rate clearly follows them over the long term, but short-term movements do not always do so. Table 8.1 Withholding Tax Rate on Foreign Borrowing Maturitv Duration conditions Period (monthis) Tax rate for exemption May 1979-September 1980 16 - October 1980-September 1981 9 10 12 months August 1981-December 1982 16 - January 1983-June 1983 6 10 12 months July 1983-June 1984 11 10 24 months July 1984-February 1989 55 10 March 89-February 90 1 1 10 36 months March 90-December 90 10 10 Note: Withholding tax was exempted for loans with stipulatedl maturity. Source: Bank of Thailand. 172 Thailand N Macroeconomic Miracle: Stable Adjus.rmenr and Sustained Growth Figure 8.1. Thailand and International Money Market Interest Rates, 1970-90 Percent 12 V~~~~~~~~~~~~~~~~~~~~~ 8 01 - 1- f- ___ f -'1980H-ff +-f 1970 972 1974 1976 1978 19X( 1982 1984 1986 1988 199() -----USA IIBOR -…- --i- -hild Sour e: Bank atf Thailand and IMF. vrou Six x, s issues. Figure 8.2 shows annual average domestic deposit and lending rates in Thai- land and the United States. Four points are notable here: * Broad similarities are again evident in trends over the long term, but short- term deviations between the movement of Thai and U.S. rates clearly occur. * Thai interest rates tend to exceed U.S. rates. This is especially true for lending rates. Thai lending rates exceeded U.S. lending rates throughout the two decades and exceeded them by four to five percentage points for all years except 1981, when the two converged. This difference seems to be greater than expectations of baht devaluation could explain. Thai depos- it rates were about two percentage point,, above U.S. rates except in the 1978-82 period, when U.S. rates were one to two percentage points higher than the corresponding Thai rates. * The average difference between lending and deposit rates in Thailand is about six percentage points. In the United States it is about two percentage points. The difference may reflect a lack of competition in the Thai finan- cial system, a consequence of the barriers to entry in the banking system discussed in chapter 3 and the well-known collusive behavior of the large Thai commercial banks. The wider interest spreads in Thailand can be at- The Role of Monetary Policy 173 Figure 8.2. Thailand and the United States, Lending and Deposit Interest Rates, 1970-90 Percent 20.00 Thailand 18.00 - ,A - ' 16.00 14.00 U States *' / _5x v ~~~~Thailand 12.00 - United States 10.00 -, s,//' 8.00- 6.00 - ' ' United states 4.00 2.00 0 .0 0 ,II,,,,,,,-f 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 ..... *Lending rates Deposit rates Source: Bank of Thailand and IMF, International Financial Statistics, various issues. tributed in part to the high intermediation costs implied by some regula- tions on commercial banks' portfolios, such as credit allocation controls. Thai rates are less volatile than their U.S. counterparts. The margin be- tween Thai and U.S. lending rates declines when U.S. rates rise and de- clines when they fall. This feature is observed even more strongly with deposit rates. These facts suggest that Thai interest rates are more flexible downward than upward, as a result of the interest rate ceilings. In the long run, then, Thailand's domestic interest rate does not deviate sig- nificantly from the world rate of interest; but in the short run it can deviate signif- icantly from the world rate. These results suggest that although capital mobility may be close to perfect in the long run, this may be may be far from the truth in the short run. Deviationsfrom Interest Rate Parity The hypothesis of perfect capital mobility can be tested by examining the interest rate parity relationship. In a world of perfect capital mobility with risk-neutral in- vestor-speculators, interest rate parity should hold. The difference between do- 174 Thailands Macroeconomic AMiracle: Stable Adjustment and Sustained Growth mestic and foreign interest rates would simply reflect the expected devaluation. In this case, since the foreign interest rate is exogenously given, and if the govern- ment cannot control the devaluation expectations of the public, the monetary au- thority would be unable to influence the domestic interest rate. If the expected devaluation can be measured by the forward premium price of foreign exchange (F), and this rate adjusts to the difference between the domestic (R(d) and foreign (RI) interest rates, then (8.1) F= Rd-Rf which implies that in equilibrium there is no profit to be made from arbitrage. The relationship described in equation (8.1) is known as nominal covered interest par- ity. If the hypothesis of perfect capital mobility was correct, we would expect the difference between F and Rd - Rf to remain close to zero. In our empirical analysis we use quarterly data from the beginning of 1980 to the end of 1989. The one-month forward premium of the U.S. dollar, measured in baht, is used as a proxy for the expected exchange rate devaluation. The London Eurodollar rate of interest is used as the foreign interest rate, and the interbank rate is used as the domestic interest rate. Figure 8.3, which shows the difference (Rd - Rf) - F, expressed as a percentage of F, reveals systematic deviations from covered interest parity in the 1980s. The negative deviations from 1980 to 1985 indicate that the interest rate differential was less than the forward premium. The positive Figure 8.3. Deviations from Covered Interest Rate Parity, 1980-90 Percent (quarterly) 4 7. 0 -2 -4t 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 Source: Bank of Thailand and IMF, International Financial Statistics, vafious issues. The Role ofMonetarv Pohihv 175 deviations after 1986, however, suggest that devaluation expectation was well be- low the interest rate differentials. Transaction costs and risk-averse behavior may be the reasons for deviations from interest parity. Political risks arising from different tax structures or capital controls may also contribute to the deviations (Aliber 1973). Evidence on the dis- tortionary effects that capital controls can have on interest rate parity comes from both Japan (Otani and Tiwari 1981) and Germany (Dooley and Isard 1980). Even with the possibility of offsetting capital inflows and even under a fixed ex- change rate, governments can in practice have a degree of monetary independence, because of various insulation mechanisms and their ability to sterilize externally in- duced changes in money supply (Llewellyn 1980X). These insulating factors will cause a deviation of the domestic interest rate from the foreign interest rate. If the central bank induces a deceleration of monetary growth by slowing down the expan- sion of the monetary base, the domestic interest rate w ill tend to rise above the for- eign interest rate. However, a withholding tax on foreign borrowing can be imposed to prevent capital inflows, in effect reducing the interest rate differential. One such insulating factor is the forward premium exchange rate, which nor- mally adjusts with the interest rate differential. It does so because as foreign bor- rowing rises to exploit the interest differential and this foreign capital is converted into baht. speculators purchase foreign exchange lorward to cover themselves against the possibility of a baht devaluation. The danger of a devaluation is that it would raise the future cost (in baht) of purchasing the foreign exchange required to repay the foreign loans. But as the demand for these forward contracts rises, their price rises as well. The concern here is how quickly the forward premium ris- es in response to this increased demand. The profitability of covered arbitrage will be reduced if the forward rate adjusts instantaneoius]\ with the interest differential. Forward covered positions are used for short-term borrowing and trade credit. It is possible that short-term capital flows will be inelastic with respect to changes in in- terest differentials, since the forward premium becomes an insulating mechanism. Fig- ure 8.4 shows the relationship between the Jorward premium exchange rate and the interest rate differentials, but this time these two v ariables are graphed against one an- other. Although the relationship is not one to one, as indicated by some observations that deviate from the 45-degree line in the figure. the forward rate clearly adjusted pos- itively to the interest rate differentials. This implies that the profitability of internation- a] arbitrage is reduced substantially when covered positions are taken into account.: We can explore this issue further by estimating the relationship between the forward premium and the nominal interest rate differential discussed earlier in re- lation to equation (8.1). The Ordinary Least Squares (OLS) result, corrected for first-order autocorrelation. is (8.2) F = 0.31 + 0.88 (Re, - R,) (0.59) (6.01) T2.= 0.6 l,DW = 1.99, p = 0.42 (2.67), n = 39 176 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 8.4. Interest Rate Differential and Forward Premium Rate, 1980-90 Forward premium rate (quarterly) 10T 8 t 64 4 2 *- () 4 * E. U { -2 -4 -- - - - -4 -2 0 2 4 6 8 10 Interest rate differential (percent) Source: Bank of Thailand, Monthlv Bulletin, various issues, and authors' calculations. where figures in parentheses are t-statistics, R2 is the coefficient of determination adjusted for degrees of freedom, DW is the Durbin-Watson statistic, r is the first- order autoregressive coefficient, and n is the number of observations. Regression (8.2) essentially tests for nominal covered interest rate parity. It indicates that the forward rate adjusts in the same direction as the interest rate dif- ferentials but does not adjust fully to the change in the interest rate differential. Capital mobility is still far from perfect, since the coefficient of the explanatory variable is less than unity. The linear relationship in regression (8.2) imposes the assumption that the coefficients of Rd and Rf are equal but with opposite signs. We can further relax this assumption, with the result (8.3) F = -3.13 + 0.95 R, - 0.62 R, (3.99) (8.97) (5.19) R2 = 0.70; DW = 1.82; n = 40 (l980QI to 1989QIV) (t-statistics in parentheses). The Role of Monetarv Policy 177 The forward premium rate adjusts more to changes in the domestic interest rate than to changes in foreign rates. If monetary contraction leads to an increase in domestic lending rate, the forward premium would rise by the same proportion, thus mitigating the effects of offsetting capital inflows, since the cost of covered interest arbitrage also increases. If one assumes that there is no long-run money illusion, one could argue that real covered interest rate parity should hold. In testing the hypothesis of real inter- est rate equality, we introduce the rate of change of the domestic price level (Pd*) and the rate of change of the foreign price level (pj*). If real interest rate parity holds, the real return on money. taking into account the foreign exchange risk, should be equalized between countries. (8.4) F = Rd - Pd*-( Rf -p ). This implies that in terms of the nominal interest rate differential, (8.5) Rd - Rf = F + (Pd* - P.*). This equation can be used for testing the hypothesis of perfect capital mobility by examining the estimated parameters in the relationship: (8.6) Rd - Rf = a + b [F + (P9d* - pf*)] If capital mobility is perfect, a = 0 and b = 1. In other words, perfect capital mobility will eliminate the covered interest rate differentials measured in real terms. Data on the forward rate between the baht and the U.S. dollar are not avail- able before 1980. We thus employed quarterly data from 1980 to 1989, inclusive, in regressing the interest rate differential on a constant term and the difference be- tween the forward rate and the difference between the inflation rates in Thailand and the rest of the world.4 We obtained the following result from OLS estimation, corrected for first-order autocorrelation: (8.7) Rd - Rf = 1. 19 + 0.]3 IF + (Pd* - Pf )] (2.16) (2.74) T2 = 0.34; DW = 1.97; p = 0.45 (2.98); n = 38. The constant term is significantly different from zero, while the slope coeffi- cient of the forward rate and inflation differentials is significantly different from unity. Thus the hypothesis of perfect capital mobility is rejected. It could be argued that the forward premium rate may not represent the true value of the expected change in the exchange rate. According to a test of the joint hypothesis that spec- ulators are risk-neutral and their expectations are rational, the forward premium 178 Thailand :v Macroeconomic Miracle: Stable Adjustment and Sustained Growth exchange rate may be a poor predictor of future change in the exchange rate (Cum- by and Obstfeld 1984). Thus one may wish to assume perfect foresight in the sense that the public can correctly anticipate a future change in the exchange rate. The forward premium rate (F) can then be replaced by the actual percentage change in the exchange rate (E). We obtained the following result, again corrected for first-order autocorrela- tion: (8.8) Rd - R-= 1.40 + 0.05 IL + (Pds* - Pf*)) (2.24) (1.19) - 0.23; DW = 2.03; p = 0.48 (3.23); n = 38. It is clear that the constant term is significantly different from zero and the es- timated slope parameter (0.05) is close to zero, rather than unity. The real interest rate parity hypothesis does not hold, even if perfect foresight is assumed. The above evidence clearly indicates that in the case of Thailand the degree of capital mobility is less than perfect. Comparable results were obtained by Mon- adjemi (1990): using a similar approach, he found that the degree of capital mo- bility of five OECD countries was far from perfect. Using an indirect testing method. Feldstein and Horioka (1980) similarly found that the ratio of domestic investment to income can be explained mainly by the ratio of domestic saving to income. Therefore, capital mobility was not high enough to destroy the relation- ship between the two ratios. Evidently capital inflows do not adjust rapidly and massively to interest rate differentials. When the central bank attempts to raise domestic interest rates by monetary contraction, the interest rate differential will widen, thus inducing capi- tal inflows to exploit the profit opportunity. The responsiveness of the forward pre- mium will, however, partly choke off the exploitation of these interest differentials. The explanatory power of regressions (8.2) to (8.8) is not particularly high, suggesting that there may be omitted variables that could help explain the system- atic interest rate differential. The next question to consider is to what extent these interest rate differentials respond to changes in monetary policies. Interest Differentials and Monetary Growth Just because the domestic rate of interest can differ from the foreign rate, does this mean that the central bank is able to manipulate the domestic rate of interest for stabilization purposes? Figure 8.5 shows the relationship between (a) the deviation between the domestic interest rate and the foreign interest rate, expressed as the difference between the domestic interbank rate and the Eurodollar interest rate in London, and (b) monetary growth, expressed as a three-year moving average of Ihe Role of Monetanr Policv /79 Figure 8.5. Autonomy of Monetary Policy, 1973-90 Percent 25T 20)9 0 1973 1975 1977 1979 1981 1983 1985 1987 1989 -R Rd-Rf - - - - gM 1-3 yr moving average Source: Bank ot Thailand, Monthlv Bulterin. various issues. the growth rate of MI. The money market rate is employed since, unlike the lend- ing deposits rates, it was not subject to the interest rate ceilings. From figure 8.5, the domestic rate of interest was always higher than the for- eign rate, except in 1981. The differential (Rd - Rf. was more than 5 percent be- tween 1975 and 1977, and again in 1985. To what extent were the differentials due to domestic conditions? Since the domestic rate of interest adjusts with some lags when money supply changes, a three-year moving average of monetary growth is used in figure 8.5, which reveals a striking inverse relationship between the inter- est rate differential and monetary growth. An acceleration in monetary growth pro- duces a lower domestic rate of interest in relation to the foreign rate, and vice versa. The largest interest spreads-between 1975 and 1977 and in 1985-oc- curred when monetary growth declined substantially. After 1985, the interest spread declined, when monetary growth rose rapidly. To confirm this finding statistically, we used quarterly data from 1970(QIII) and 1989 (QIV) to explain the interest rate differentials. We obtain the following OLS result, again corrected for first-order autocorrelation: (8.9) Rd - Rf = 5.69 - 0.099 M* + 0.535 R, + seasonal duinmies (1.52) (2.31) (2.3) 2 =0.44;DW=2.18;n=77 180 Thailand : Ma ,vroeconomi Miracle: Stable Adjustment and Sustained Growth where M* is the three-quarter moving average growth rate of money supply (MI) and R, is the ceiling on the lending rate. The results suggest that an increase in the growth of money supply reduces the interest rate differential. Thus, provided the money supply can be controlled, monetary policy could in principle be employed to induce a short-run deviation of the domestic interest rate from the level of the foreign interest rate. In addition, regression (8.9) sug- gests that the ceiling rate, a policy instrument of the Bank of Thailand, can be used to influence the adjustment of the domestic rate in response to changing conditions. Liberalizatiorn and the Determination of Domestic Interest Rates Despite the impediments to full capital mobility, the effect that foreign interest rates have on domestic Thai interest rates cannot be denied. This effect is obvious from figures 8.1 and 8.2. Because of imperfect capital mobility during the study period, the monetary authorities still possessed some control over domestic interest rates. But with the relaxation of controls over the capital account after the second quarter of 1990, the domestic interest rate is expected to become more sensitive to changes in foreign interest rates. This proposition can be tested by exploring the relationship between the interbank and the Eurodollar interest rates. Monetary policy would be- come less effective in controlling the domestic interest rate if the interbank rate (R,n) became more responsive to the Eurodollar rate of interest (Rf). (8.10) Rb = -39.6 + 0.25 R1 + 0.35 Ld + 0.87 Rc -14.5 D + 1.78 (D x R1) (5.4) (1.8) (4.1) (4.6) (3.7) (3.8) R2 = 0.82; DW = 2.3. D = 0 ( I 980QII - I 990QII) = I (1990QIII-1991QIV) The interbank rate reflects the interaction between the short-term demand and supply of bank funds. Its movement is determined by the domestic money market conditions proxied by the loan-deposit ratio (Ld ) as well as the foreign interest rate. The ceiling on the end of period interest rate (R.,) is included in the explana- tion of the interbank rate, since it provides information on the stance of the mon- etary authorities. This information is important in an industry that has an oligopolistic structure and where price competition is consequently lacking. The product of the dummy variable and the Elurodollar rate is highly signifi- cant and positive, as expected, which indicates that after the relaxation of capital controls, the domestic interest rate, represented by the interbank rate, became more sensitive to changes in the foreign interest rate. Nevertheless, although for- eign interest rates have become more powerful in determining the course of the domestic interest rate, domestic conditions still matter. The Role of Monetarv Policy 181 Monetarv Growth and GDP Growth The autonomy of monetary policy requires the existence of efficient monetary pol- icy instruments. We have seen that the Thai monetary authority can create a short- run deviation from the world rate of interest. However, the domestic interest rate is only an intermediate target. We wish to know whether monetary policy can achieve stabilization with respect to its ultimate targets, such as real output and in- flation, at least in the short run. Real business cycle theorists would argue that changes in the money supply have no impact on real output, either in the short run or the long run, while monetarists claim that monetary policy can affect real GDP only in the short run. The new classical econo- mists assert that monetary policy can affect output if the policy is unanticipated. If mar- kets clear and economic agents are rational, anticipated policies would have no effect, even in the short run. Neo-Keynesians would contend that money still matters because of wage and price rigidities caused by contractual arrangements in the labor market. To shed light on whether money matters in Thailand, given the Thai institu- tional mechanisms and the underlying mode of expectations formation, we exam- ine the relationship between growth rates of the monetary base and real output. The effectiveness of monetary policy would be enhanced by the existence of stable relationships between nominal GDP and the money stupply, and between money supply and the monetary base. Figure 8.6. Monetary Base and Real Output, 1972-90 Deviation from trend growth rate (percent) 6- 4 1972 1974 1976 1978 198() 1982 1984 1986 1988 1990 Monetary base (lagged one period) --. G)P growth rate Source: Bank of Thailand, Mltonthlv Bulletin, * arious isSues. 182 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth We first calculate the trend growth paths of output and the monetary base from a simple regression with a linear time trend. We obtain the detrended series of growth rates of these variables by taking the difference between the actual growth rates and the trend growth rates derived from the estimated equations. The rela- tionship between output growth and monetary base growth, after removing the trend components, is shown in figure 8.6. The deviation of base money from its growth path is lagged one year to ensure that the base money variable is not a co- incident indicator for real output growth. This ensures that the direction of causa- tion, if any, is from the monetary base to income growth and not the reverse. It seems that monetary base growth precedes real output growth. A period of monetary expansion is followed by a surge in real output growth. When monetary growth decelerates, however, the growth rate of real output also declines. The out- standing exceptions to this relationship have occurred during periods of negative external shocks. These are evident in the years 1974-76 and 1979-81. At such times, even if the growth rate of the monetary base accelerates, real output does not. That is, the relationship between monetary growth and real output growth is perturbed by negative real shocks. Clearly, after 1981, in the absence of such ex- ternal shocks, the monetary base and real output became more closely related. Table 8.2 Monetary Volatility, 1973-90 (percent) Average Coefficient of Item Period growth rate variationa Money Supply (Ml) 1973-78 14.07 0.39 1979--85 6.82 0.86 1986-90 17.97 0.29 Monetary base 1973-78 13.62 0.23 1979-85 10.53 0.35 1986-90 16.56 0.22 Coefficient of Item Period Average value variation Money multiplier 1973-78 1.24 0.02 1979-85 1.15 0.08 1986-90 1.10 0.02 Credit multiplier 1973-76 2.91 0.17 1979-85 4.85 (.18 1986-90 6.65 0.09 Velocity of money 1973-78 8.38 0.06 1979-85 10.34 0.09 1986-90 10.33 0.04 a. The coefficient of variation is the ratio of the standard deviation to the mean value. Source: Calculated from Bank of Thailand, MonthA Bulletin. various issues. The Role of Monetary Policy 183 The stability of the relationship between money supply and the monetary base is dictated by the stability of the money multiplier. As table 8.2 indicates, the coeffi- cients of variation of the money multiplier, as well as the credit multiplier, declined from the period 1979-85 to the period 1986-90. Both the money supply and bank credit volume exhibit a more stable relationship with the monetary base after 1985. A stable relationship between monetary base and money supply is not suf- ficient to guarantee the effectiveness of monetary policy. The final link be- tween the ultimate target (output) and the intermediate target (money supply) must also exist. The velocity of money (MI. shown in figure 8.7, remained stable between 1970 and 1990. Narrow money changed hands, on average, be- tween eight and ten times within a year. The coefficient of variation of velocity also indicates that it fluctuated on average 4 percent per year from its mean value during the period 1986-89, compared with 9 percent during the period 1979-85. The velocity of money calculated from M, exhibits a greater varia- tion than the velocity of Ml, because M2, consists of both Ml and quasi money. The demand for the latter is dictated mainly by wealth accumulation motivated by real interest rates and real income, rather than by transactions motives. We conclude that changes in the monetary base led to deviations from the trend growth path of the monetary growth rate, which in turn produced an effect on output. Figure 8.7. The Velocity of Money, 1970-90 Percent 12 21 6T 2 - ---- - - -- - - -- - - - 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Velocity 1 (VI=GNP/MI) ------ Velocity 2 (V2=GNP!M2) Source: Table 3.4 and Bank of Thailand, Monthl/ Bulletin, various issues. 184 Thailand:s Macroeconomic Miracle: Stable Adju.%tmeni aind Sustained Growth Control of the Monetary Base We have argued that a stable and predictable relationship exists between the size of the monetary base and the intermediate and ultimate targets-interest rates and GDP growth. But does the Bank of Thailand really control the level of monetary base? This issue is important because it is clear that one component of the mone- tary base is outside the control of the Bank of Thailand-net foreign assets. It is possible that the Bank of Thailand can raise the domestic interest rate but only for a short time. The widened spread between domestic and foreign interest rates will induce capital inflows, thereby increasing the net foreign assets. If the Bank of Thailand must defend the existing level of the exchange rate, the increased mone- tary base would thwart the Bank of Thailand's attempt to raise domestic interest rates. Whether the impact of offsetting capital flows is strong depends on the struc- ture of capital flows as well as the degree to which capital flows respond to chang- es in interest rates. Composition of the Monetary Base Table 8.3 shows the decomposition of the change in the monetary base into net for- eign assets (NFA) and net domestic credit (NDC). The latter consists of net claims on government (NCG), net claims on financial institutions (NCF), and net other lia- bilities (NOL) of the Bank of Thailand. Thus (8.11) B = NFA + NDC and (8.12) NDC = NCG + NCF + NOL. The claims of the central bank on financial institutions is a controllable item in the sources of the monetary base. To the extent that the central bank is able to act independently of the government, claims on the government is also a con- trollable item. These two components are the domestic assets of the central bank; they can be thought of as the part of monetary base over which the central bank has some control. Under the system of a (relatively) fixed exchange rate, in which the central bank has to intervene in the foreign exchange market to maintain the stability of the exchange rate, the net foreign assets component of the monetary base is an uncontrollable itemn. It the foreign assets constitute a high proportion of the monetary base, the ability of the central bank to control the money supply will be further limited. In other words, the effectiveness of monetary policy will be reduced if the central bank wishes to maintain exchange rate stability. The Role of Monetary Policy 185 Table 8.3 Foreign and Domestic Components of Monetary Base, 1971-90 (percent) Year dB/B dNFA/B dNCG/B dNCF/B dNOLIB dNDC/B 1971 10.97 -3.73 24.34 2.44 -12.09 14.70 1972 17.76 20.60 3.18 1.53 -7.55 -2.84 1973 18.15 24.02 -1.71 8.67 -12.82 -5.87 1974 13.04 47.49 -23.40 5.41 -16.46 -34.45 Average 1972-74 16.32 30.70 -7.31 5.20) -12.27 -14.38 1975 10.71 -6.21 11.51 12.85 -7.11 16.92 1976 13.56 0.07 59.88 -4.59 147.07 13.49 1977 9.32 2.37 16.98 0.65 15.49 6.95 1978 16.96 28.26 28.62 8.82 41.36 -11.30 1979 16.63 22.44 20.40 21.74 47.52 -5.82 Average 1976-79 14.12 13.29 31.47 6.66 62.86 0.83 1980 14.03 -2.36 33.28 4.85 18.45 16.38 1981 6.57 -21.38 18.61 4.62 -2.08 27.96 1982 11.95 -2.38 27.95 2.20 13.72 14.33 1983 10.47 -9.77 17.84 2.83 2.47 20.24 Average 1980-83 10.75 -8.97 24.42 3.63 8.14 19.73 1984 5.58 17.97 -6.37 5.63 12.05 -12.39 1985 8.49 2.12 16.67 7.73 12.11 6.38 1986 11.31 23.76 -5.47 12.18 13.86 -12.45 1987 21.10 36.84 -9.07 7.0)4 16.45 -15.74 1988 14.87 48.21 -33.70 14.78 4.13 -33.34 1989 16.92 74.76 24.02 -11.34 6.08 -57.84 1990 15.67 27.17 -61.77 -22.32 13.84 -11.50 Average 1984-90 13.42 32.98 -10.81 1.96 11.22 -19.55 Note: d = First difference operator, dB = B - B(-l I B = Monetary base. NFA = Net Foreign Assets. NCG = Net Claims on Govemment. NCF = Net Claims on Financial Institutions. NOL = Net Other Liabilities. NEC = Net Domestic Credit (B-NFA). Source: Computed from Bank of Thailand, Monthly Bulletin, various issues, table 3. 186 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth The Structure of Capital Flows Over the past two decades, Thailand's current account was usually in deficit, while, correspondingly, the capital movements account was in surplus.5 Thailand was able to sustain rapid economic growth by maintaining a high ratio of invest- ment to GDP. The investment-savings gap is financed by net capital inflows. Long- term capital flows are usually larger than the short-term flows. In some years, the value of the long-term net flows alone overwhelmed the current account deficit (figure 8.8). Long-term capital inflows are more stable than short-term inflows, and the investment-savings gap is financed primarily by the former. Long-term private flows accounted for 40 percent of total capital flows. The largest component is the long-term borrowing from the government enterprises. The long-term borrowing of private enterprises is 10 percent of total capital flows. Direct investment and long-term private flows represent an average of 9 and 21 per- cent of the monetary base, respectively. Direct foreign investment was responsible for a large surplus in the capital account in 1974 and from 1987 to 1990. Its average share in total capital flows between 1970 and 1990 was 26 percent. Table 8.4 indi- cates that the share of direct investment in total capital flows is relatively stable. Portfolio investment from abroad is included as part of long-term private flows. During the development stage of the capital market in Thailand, portfolio invest- Figure 8.8. Net Capital Inflows, 1970-90 Ratio to current account deficit 5 7 4t A 1 97() 1 972 1 974 1 976 1 978 1 981) 1 982 1 984 1 986 1 988 1 99() Short-term ----- Long-term Source: Bank of Thailand, Monihly Bulletin, various issues. I'he Role ofl A'fontettrrv Polic !' 187 Table 8.4 Components of Capital Movements, 1970-90 Average shlare A.s percent of mionetary base Coniponenit Mean SD/Mean Mealn X)/Mean Direct investment 25.6 0.64 S.49 0.71 Long-term private flows 39.42 0.57 20.63 0.92 Long-term government entrepreneurship 20.37 0.94 10.10 1.1 0 Long-term private entrepreneurship 9.73 2.66 7.19 1 .4 0 Portfolio investment 8.71 1.59 3.12 1.74 Short-term private flows 22.88 (.8 8.84 0.85 Short-term loan 0.56 23.77 1.40 3.32 Trade credit 8.14 1.91 1.95 2.56 Other 14.18 1.43 5.49 1.0) Central government 12.1 1.83 5.91 1.28 NVore: SD denotes standard deviation. Sourece: Computed from Bank of Thailand, Montil/il Bulletin. tahle I and 39. ment was relatively stable and remained only 3 percent of the monetary base.6 it was only in 1986 that the Thai capital market was able to attract a considerable amount of foreign portfolio investment. The large surplus in the balance of pay- ments in 1987 and 1989 stemmed partly from a surge in portfolio investment. The most variable component of total capital movements is short-term loans of private enterprises, which accounted for an average of on]y 1.4 percent of the monetary base. Its share in total capital movement was small in relation to trade credit and other short-term capital movements. The latter two components have a larger share than short-term loans, but their fluctuations were much smaller. Total short-term private capital flows accounted for 23 percent of total movements. They also showed a small degree of variation and represented 8.8 percent of the mone- tary base. If short-term capital flows are large in relation to the monetary base, it will become very difficult for the central bank to sterilize the impact of offsetting capital flows. The Offsetting Capital Flows Hypothesis an(d the Feasibilitv of Sterilization As table 8.3 makes clear, the net foreign assets component of the monetary base fluctuated more than other components. The offsetting capital flows hypothesis 188 Thailands Macroeconomic Miracle: Stable Adjutment 1 and Sustained Growth implies that the change in the policy-induced domestic component of the mone- tary base will be offset by (uncontrollable) capital outflows. If discretionary mon- etary policy induces offsetting capital flows by causing the domestic interest rate to deviate from the world market interest rate, this response must stem mainly from the interest-responsive components of capital flows. Thus to assess the valid- ity of the offsetting capital flows hypothesis it is necessary to examine the sensi- tivity of different components of capital flows to interest rates. From table 8.3, a negative correlation between changes in NFA and NCG is ap- parent (the correlation coefficient is -0.48). This relationship could be explained by the central bank's sterilization of capital flows, which implies that the increase (reduction) in NFA was the cause, via the sterilization response, of the reduction (increase) in NCG. The mechanism is that when a capital inflow occurs, and the central bank is thus obliged to purchase foreign currency (NFA rises), the increase in the monetary base is offset by the sale of government bonds (NCG falls). But sterilization may not be feasible. It massive capital flows result from small interest rate differentials, the central bank may not possess the resources re- quired to sterilize their effects on1 the domestic monetary base. In a world of in- perfect capital mobility, it might be feasible for the central bank to prevent the domestic interest rate from converging rapidly to the foreign interest rate by ster- ilizing the effect of offsetting capital inflows or by using capital controls. If do- mestic and foreign assets are imperfect substitutes, the elasticity of capital flows with respect to interest rate differentials may be small. The elasticity of capital flows with respect to domestic and foreign interest rates thus determines the fea- sibility of sterilization measures. We employed quarterly data from 1970 to 1989 to calculate the interest elas- ticity of short-term and long-term capital flows. This was done by regressing the values of capital flows on the interest rate differentials. For short-term flows (STF), the total values of imports and exports (X + M) as well as the forward premium rate (F) are included as regressors. Since the government's regulations did not per- mit the maturity of forward contracts to exceed six months, the variable F was not included in the equation for long-term flows (LTF). but the level of GDP is included. In both cases, previous quarter capital flows are included to capture lagged adjust- ment effects. The results are (8.13) STF = -940.2 + 207.8(Rd-Rdt + 0.199 (X+M) - 294.5F + 0.449STF (3.02) (2.08) (5.93) (3.18) (4.58) -2 = 0.78, h = 0.087 (8.14) LTF = -248.6(Rd-RI) + 0.021 GDP + 0.49LTF_ (0.84) (4.59) (4.61) R2 = 0.66, h = -1.83 where h is Durbin's h-statistic and other diagnostics are as before. The Role of Monetarv Policv 189 The results reveal that although short-term flows respond to interest rate dif- ferentials, long-term flows do not. An increase in the level of economic activity, as represented by international trade and GDP, stimulates capital inflows. A rise in the forward premium rate reduces the flows of short-term capital. Both short-run and long-run flows exhibit lagged adjustment patterns. as indicated by the significance of the lagged dependent variables. The estimated coefficients from (8.13) and (8.14) were used to calculate the short-run and long-run sensitivity of capital flows with r espect to their correspond- ing determinants, and the results are reported in table 8.5. The elasticities of both short-term and long-term net capital inflows with respect to interest rate differen- tial are small, which indicates that international capital mobility is relatively low. Short-term capital flows respond more rapidly to changing trade values. Thus, as noted earlier, short-term capital inflows are able to offset the current account def- icit. Although in the short run the growth rate of GD)P is higher than the growth rate of net long-term capital inflows. in the long run the elasticity of long-term flows with respect to GDP is greater than unity. Control of the monetary base involves fiscal as w ell as purely monetary is- sues. A clear example occurred in the late 1980s. Figure 8.9 shows changes in the level of the monetary base (DBASE) from 1971 to 1991 and its components: net foreign assets (DNFA), claims on financial institutions (DCF), and net claims on government (DNCG). While net foreign assets increased substantially from 1986 to 1991, the monetary base did not rise correspondingly. The reduction in the net claims on government and the slowdown in the claims on financial insti- tutions offset the rapid increase in net foreign assets. Because of the surplus in the government budget, the increased capital inflows did not result in excessive growth in the monetary base. These issues thus have bearing on the discussion of fiscal policy in chapter 7. Had the government run a budget deficit, the level of monetary base would have risen to levels that would have threatened price stability. Table 8.5 Interest Rate Elasticity of Net Capital Flow, Quarterly Data, 1970 II to 1989 IV Short-term nflow Long-termnflow Short ruin Long run Shiort runl Long run Interest rate differential 0.177 0.322 -0).069a -0. 135a Trade v'olu,ne Domnestic output Slhort run Long run zShort run Lonig rut Economic activity 1.084 1.970 0.673 1.323 a. The estimated coefficient Insignificant at 5 percent level. Source: Authors calculations from data in Bank of Thailand. 41/ omh/v Bul/etie. various issues. 190 Thailand s Macroeconomic Miracle: Staible Adjustment and Sustained Growth Figure 8.9. Change in Monetary Base Components, 1971-91 Percent 120 100 80 60 40 \ 20 0 -20 -40i -60 \ -80 1 -+-~- -+-+ -+ - - 4_ - _+- +i 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 -- DBASE -{-} DNFA E13 DCF ---DNC( Source. Bank of Thailand, Monthly Bulletin, various issues, and authors' calculations. Instruments of Monetary Policy It appears that there has been some scope for exercising discretionary monetary policy in Thailand, although that scope has been present in the short run only and has been limited by the existence of offsetting capital flows. Consider now the manner in which monetary policy has been employed over the two decades to 1990, particularly the way it conditioned Thailand's responses to external shocks. Two important instruments of monetary policy are adjustments to the bank rate and adjustments to interest rate controls. Adjustments of the Bank Rate The main instrument of monetary policy is the central bank lending rate. Various monetary aggregates, including the monetary base, are used as intermediate targets. Table 8.6 provides a historical description of the changing nature of Thailand's monetary policy, based on movements of the bank rate and changes in the monetary growth rate. Between 1970 and 1990 monetary policy changed its direction six times. The longest expansions lasted about three years (1970-72 and 1975-77), while the longest contraction lasted four years, (1978-81). The periods of expansion and contraction distinguished in table 8.6 were identified by changes in the Bank of Thailand lending rate. Comparison with table 3.4 reveals that these changes coin- The Role of Monetary Polity 19i Table 8.6 Direction of Monetary Policy and Average Key Economic Indicators, 1971-89 (percent) Deviation Real of actual Current Monetan, GCP income account base Government growth from trend le.fic it/ growth budget Year Direction rate Inflation income GDP rate deficit/GDP 1971-72 + 4.52 2.63 -0.86 .16 14.37 4.65 1973-74 - 7.11 19.93 -0.55 2.09 15.60 0.63 1975-77 + 8.04 5.56 -1.06 4.36 11.20 2.98 1978-81 - 6.72 12.55 2.80 6.95 13.55 2.79 1982-83 + 5.66 4.49 -1.34 5.66 11.21 3.80 1984 - 7.13 0.85 -0.87 5.51 5.58 3.49 1985-87 + 5.97 2.25 -4.31 1.91 13.63 2.60 1988-90 - 12.38 5.06 5.58 4.94 16.79 -3.77 a. A plus sign (+ indicates expansion, a minus sign (-) indicates contraction. Directions of monetary policy are evaluated from the direction of change in the bank rate and the growth rate of monetary base. Source: (As for table 7.1). cided closely with changes in the real quantity of narrow money (MI). Table 8.6 also reveals that periods of monetary expansion (contraction) coincided broadly with periods of fiscal expansion (contraction).7 Net domestic credit consists of two main components: the claims on govern- ment and the claims on financial institutions. The former is dictated by the bor- rowing requirements of the government, which depends on the size of the deficit. The latter depends on the borrowing requirements of commercial banks, which de- pends on the cost of borrowing from the central bank relative to other sources. The bank rate, or the rate that the central bank charges commercial banks, can be used to affect the claims on the financial institutions (NCF) component of the monetary base and to influence capital flows. The Bank of Thailand influences the short-term market rate of interest by ad- justing the bank rate and by intervening in the repurchase market. The latter con- sists of buying and selling government bonds, but it is a less important instrument than adjustments to the bank rate. Bank rate adjustments occurred most frequently during the period of foreign interest volatility from 1980 to 1982. Table 8.7 provides a more detailed summary of the adjustments in the bank rate between 1979 and 1990. Adjustments were especially frequent between 1979 and 1982, during which time monetary policy reflected the desire to increase net domestic credit in an effort to revive the recessive economy. The Bank of Thailand did not adjust the bank rate between 1987 and 1988, since the other component of net domestic credit (the claims on government) was already declining owing to the 192 Thailand 's Macroeconomic Miracle: Stable Adju tmemit tandi Sustaiined Growth Table 8.7 Adjustment of the Bank of Thailand Lending Rate, 1979-90 Number of Direction Net rate of Year adjustments Positive Negative changea 1979 2 2 0 2.5 1980 6 3 3 0.0 1981 5 3 2 1.0 1982 4 1 3 -2.0 1983 2 1 1 0.5 1984 1 0 1 -1.0 1985 1 ( 1 -1.5 1986 3 0 3 -3.0 1987 0 0 0 0.0 1988 0 0 0 0.0 1989 0 0 0 0.0 1990 2 2 0 3.0 a. Percent. Source: Calculated from Bank of Thailand, Monthl1 BUlletiti. various issues. budgetary surplus. In 1990 the bank rate was adjusted upward twice, by a total of three percentage points (table 8.8), to curb the overheated economy. As tables 8.7 and 8.8 suggest, a positive difference between the bank rate and the Eurodollar rate induces capital inflow, since it is cheaper to borrow from abroad than from the domestic money market. At such times, the ratio of Thai commercial bank's borrowings from the Bank of Thailand in relation to foreign banks is likely to decline. Given the size of the central bank's lending, a negative sign on this ratio indicates capital inflow, while a positive sign indicates capital outflow. In 1979 the Bank of Thailand provided considerable credit for commercial banks for liquidity purposes. In 1985 commercial banks tried to reduce the out- standing amount of foreign loans due to excess liquidity. Except for the above two years, there was a negative correlation between the differential between domestic and foreign rates and the rate of change of the commercial banks' borrowing ratio. Table 8.8 indicates that adjustments in the Bank's lending rate can be employed to affect the balance of payments position. Adjustments of Interest Rate Controls Aside from intervening in the short-term money market, the Bank of Thailand has set maximum interest rates on both lending and borrowing. As with the exchange rate between the U.S. dollar and the baht, which remained unchanged in the 1960s The Role ofl Moneturv Poliyv 193 Table 8.8 Adjustments in the Bank Rate and Capital Movements, 1970-90 Commercial bain&k Rate of' Eurodollar Differettial borrowingsc/ change of BanDk rate' rtteb [(I) - (2)1 Joreig'n borroting1 d column 4 Ye ar (1) (2) (3) (4) (5) 1970 9.0( 8.52 0.48 0.25 - 1971 9.10 6.58 2.42 0.39 60.87 1972 8.(( 5.46 2.54 0.36 -8.20 1973 10.00 9.24 0.76 0.42 15.26 1974 11.00 11.01 -0.(1 0.53 25.95 1975 1(.00 6.99 3.01 0.91 72.26 1976 9.00 5.58 3.42 0.60 -33.88 1977 9.(0 6.05 2.95 0.40 -32.45 1978 12.50 8.78 3.72 0.33 -18.93 1979 12.50 12.01 0.49 0.47 43.94 1980 13.50 14.06 -0.56 0.65 37.93 1981 14.50 16.82 -2.32 0.67 2.89 1982 12.50 13.16 -0.66 (.81 21.24 1983 13.00 9.60 3.40 0.61 -25.41 1984 12.00 10.78 1.22 0.52 -14.19 1985 11.00 8.34 2.66 0.66 26.28 1986 8.00 6.77 1.23 1.29 96.99 1987 8.00 7.11 0.89 1.44 10.96 1988 8.00 7.91 0.09 1.09 -24.12 1989 8.0( 9.10 -1.1( 0.59 -45.80 1990 12.00 8.28 3.72 0.44 -25.42 a. Bank rate or loan rate. b. Eurodollar rate in London. c. Commercial bank borrowings from Bank of Thailand. d. Borrowings from banks abroad. Source: Calculated from Bank of Thailand. Monthlv Bulletin, and International Monetary Fund, In- tertetionil F inancial Stiwitjc.s, various issues. and 1970s, the Bank of Thailand did not change the maximum interest rates on lending and borrowing over the same period until the first oil price shock. The maximum rates were effective; commercial banks usually paid their depositors at the maximum rate prescribed by law. Despite the ceilings on bank deposit and 194 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth lending rates, real interest rates usually remained positive (table 8.8) and high enough to attract financial savings. The ratio of bank deposits to GDP has increased over time, reflecting financial deepening in Thailand from 1970 to 1990 (figure 8.10). The increased availability of financial resources enhanced the quantity and quality of investment. Bank cred- its, as a percentage of GDP, also grew considerably, following the rapid rise in bank deposits. In an economy where financial intermediaries efficiently perform the role of savings mobilization and credit allocation, the ratio of investment to GDP can be high, since investors are not limited by self-finance. The two oil shocks significantly disrupted this regulatory system by raising the rate of inflation and driving domestic interest rates to negative levels. During the first oil shock, inflation rose from 4.9 percent in 1972 to 15.4 and 24.3 percent in 1973 and 1974, respectively. In 1975, it dropped to 5.3 percent. Similarly, dur- ing the second oil shock in 1979, inflation rose sharply from 9.9 percent to 19 per- cent in 1980 before falling to 12 percent in the following year. We may thus summarize the record of inflation in Thailand over the past two decades as being a nornal rate in the single-digit range, interrupted by two brief periods of between one and two years where the rate surged to about 20 percent in response to external price shocks. Figure 8.10. Financial Deepening, 1970-90 Ratio to GDP 0.8 0.7 0.6 0.5 - .., 0.4/, 0.3 < 0.2 _...- 0.1 0 I -+--i- - -----A-- - + + -- 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 Bank deposits - Bank credits Source: Bank of Thailand, Monthly Bulletin, various issues, and authors' calculations. The Role of Monetnr' Poli cc 195 Table 8.9 shows that real interest rates were significantly negative in 1973. 1974, and 1980. As a result, the growth of the volume of bank deposits slowed. External financial shocks, in the form of rising foreign interest rates, discouraged commercial banks' foreign borrowing and the credit shortage in the world money markets was thus transmitted to domestic credit markets. The illiquidity of the money market can be seen from the high growth rates of bank credit in relation to Table 8.9 Maximum Interest Rates and Real Deposit Rates, 1970-90 (percent) Net Real net Loan Interest deposit Inflation deposit Year rate Time deposit rate tax raite trate raite' rate 1970 14.00 7.00 O.(0 7.00 -0.09 7.09 1971 14.00 7.00 0.00 7.(X) 0.44 6.56 1972 14.00 7.00 0.)0 7.00 4.91 2.09 1973 14.00 7.00 (.00 7.00 15.47 -8.47 1974 15.00 8.00 (.00 8.0() 24.3(0 -16.30 1975 15.00 8.00 (.00 8.00 5.30 2.70 1976 15.00 8.00 (.00 8.00 4.20 3.80 1977 15.00 8.00 1(.00 7.20 7.60 -0.4() 1978 15.00 8.00 10.00 7.20 7.90 -0.70 1979 15.00 9.00 1(.00( 8.10 9.90 -1.80 1980 18.00 12.00 10.00 10.8( 19.70 -8.90 1981 19.00 13.00 10.0( 11.70 12.70 -1.00 1982 19.00 13.00 (12.50)b 12.50 10.94 5.20 5.74 1983 17.50 13.00 (12.50) 12.50 10.94 3.80 7.14 1984 19.00 13.00 (12.50) 12.50 10.94 (1.9( 10.04 1985 19.00 13.00 (11.00) 12.50 9.63 2.40 7.23 1986 15.00 9.50 (7.25) 15.0() 6.16 1.90 4.26 1987 15.00 9.50 (7.25) 15.(0( 6.16 2.5(0 3.66 1988 15.00 9.50 (8.63) 15.0() 7.34 3.85 3.49 1989 15.00 9.50 (9.88) 15.()( 8.40 6.31 2.09 1990 19.00 - (13.75) 15.()() 11.69 5.35 5.65 a. The real net deposit rate is the difference between the nominal time deposit rate and current infla- tion rate. b. The figures in parentheses are the actual rate of interes,t paid by commercial bank>. Before 19X2, the banks deposit rates were equal to the maximum rates. Source: Bank of Thailand, Monthly Bulletin, varioLus issues. 196 Thailand s Macroeconomic Miracle: Stable A4dju.stmeni and Sustained Growth bank deposits. In 1974 inflation reached 24 percent. and The Bank of Thailand re- sponded by raising the maximum lending rate from 14 to 15 percent. To keep the interest spread intact, the ceiling on time deposits was also raised by one percent- age point, from 7 to 8 percent. In 1980 inflation reached 20 percent and the max- imum lending rate was raised by 3 and I percentage points in 1980 and 1981, respectively, while the maximum deposit rate was raised by 3 percentage points, to 12 percent from the 9 percent ceiling established in 1979. These regulatory responses can be considered late and inadequate. The ex- post real rates of deposit were still substantially negative in both periods. More rapid restructuring of interest rates to establish a realistic positive rate was clearly necessary to alleviate the burden of external shocks and to restore private savings. As can be inferred from figure 8.7 the ratio of quasi money to GDP actually de- clined from its rising trend after the real interest rate became negative. The intro- duction of interest income tax on time deposits in 1978 further delayed this return to the preshock path. The restructuring of interest rates to reestablish positive real rates resulted in an escalating real rate of interest between 1981 and 1984. The out- come was a sharp increase in the ratio of time and saving deposits to GDP after 1982. Demand conditions permitting, a surge in the growth rate of quasi money is likely to be followed by a sharp increase in credit expansion. The slowdown in the growth rate of deposits between 1977 and 1979 led to a sharp decline in the credit expansion between 1978 and 1980 but a sharp rise in credit growth between 1981 and 1983.The rapid credit expansion is said to have contributed to a deterioration in the current account balance (Chaiyawat 1984). The Thai monetary authorities applied a credit restraint policy by requiring commercial banks to restrict the growth rate of credit in the first half of 1984 to 9 percent, and for the whole year to 18 percent of 1983's credit outstanding. From 1980 to 1983, the ceiling on the deposit rate remained at 13 percent. Since 1982, and for the first time in decades, conlinercial banks paid their depos- itors below the ceiling. After 1982 the real rate of return from deposits at banks was relatively high. The escalating real rate of interest between 1981 and 1982 re- sulted in a sharp increase in bank deposits. The monetary authorities hesitated to reduce the interest rate ceiling, despite the well-known downward rigidity of in- terest rates caused by the oligopolistic structure of the Thai banking industry. As a result, commercial banks were not willing to reduce deposit rates even though the real rate of interest went up to 11.6 percent in 1984. With a credit restraint pro- gram being implemented, commercial banks were not able to rid themselves of ex- cess liquidity. In 1985 the time deposit rate was reduced from 12.5 percent to II percent. It took three years before the deposit rate finally came down. The implementation lag of interest rate policy thus retarded economic growth. The growth rate of real investment in 1985 was negative 5.2 percent. Finally, in 1986 the Bank of Thailand came to the important realization that it had underestimated the downward rigidity of interest rates.8 The bank decided to reduce the maximum interest rate on loans The Role of Monetarv Policy 197 by 4 percent and on deposits by 3.5 percent. Commercial banks followed suit by reducing the deposit rate from 11 to 7.25 percent. The experience of the two oil price shocks brings home two important points. First, at a time of a negative external shock, the combination of "sticky" interest rate controls and above-normal inflation discourages private savings, stimulating rather than retarding aggregate absorption. Unless the shock is viewed as strictly temporary, this is the opposite of the adjustment that is required. Second, in such an environment, inflation control becomes doubly important. High inflation will produce negative real interest rates and undermine the supply of private domestic and foreign savings. In the Thai case, further damage was averted by the rapidity with which inflation was contained, following each of the two oil price shocks. The Performance of Monetary Policy Consider now the performance of Thailand's monetary policy adjustments in sta- bilizing output growth and the price level. Two internal balance targets are obvi- ously involved, with the possible additional importance. from time to time, of issues of external balance. This raises the possibility of conflict between objec- tives. Precisely this conflict can be observed in the responses to the two oil price shocks. No single pattern emerges from the Thai experience as to how these con- flicts have been resolved, but inflation typically became the dominant concern of monetary policy whenever the rate of inflation rose above roughly six percent. Be- low that level of inflation, stability of income growth seems to have been a more important objective of monetary policy. To facilitate discussion of these issues we standardize actual growth and in- flation rates, along with the values of other macroeconomic variables. A subscript Z will subsequently be used to indicate the standardized value of a variable: the value of that variable minus its mean, all divided by its standard deviation. For ex- ample, Yz is defined as the standardized value of the variable Y and is given by (8.15) Yz= (Y-YT)Vy where Y is the mean of Y and ay is its standard deviation. In table 8.10. the years 1971 to 1990 are classified according to whether the standardized values of their growth rates (G,) and inflation rates (P2) were above av- erage, (corresponding to G; and P, positive), or below average (G, and P2 negative). The change in the level of the domestic credit component of the monetary base as a percentage of GDP will be used as a measure of the direction of monetary policy. Changes in domestic credit, expressed as a proportion of GDP, have been standard- ized into the variable denoted M, If Mz is positive (negative), the monetary policy stance is considered expansionary (contractionary). The standardized values of growth rates and inflation are summarized in the scatter diagram in figure 8.1 1. 198 Thailands Macroeconomic Miracle: Stable Adjustment and Sustained Growth Table 8.10 Macroeconomic Performance and Policy Stance Fiscal Monetary Real growth Inflation CAD/GDP stance stance Year (G7) (PI) (C) (F ) (M4 High growth and high inflation 1973 0.943 1.380 -1.244 -0.791 0.207 1977 0.957 0.088 0.638 0.407 0.512 1978 1.142 0.150 0.362 0.533 1.184 Low growth and high inflation 1974 -0.997 2.689 -1.173 -1.035 -1.255 1979 -0.652 0.467 1.402 0.047 1.712 1980 -0.841 2.020 0.967 1.761 1.367 1981 -0.288 0.905 1.310 0.772 0.649 Low growth and low inflation 1971 -0.778 -1.107 -0.535 -0.826 0.574 1972 -1.099 -0.380 -1.181 -0.482 0.247 1975 -0.819 -0.292 0.096 0.434 0.072 1982 -1.102 -0.295 -0.396 -1.179 0.892 1984 -0.007 -1.075 0.465 0.255 -0.808 1985 -1.302 -0.787 0.114 0.389 0.522 1986 -0.794 -0.902 -1.627 -1.743 -0.529 High growth and low inflation 1976 0.777 -0.511 -0.454 0.860 0.185 1983 0.036 -0.561 1.277 0.170 0.479 1987 0.809 -0.770 -1.118 -1.426 -0.814 1988 2.080 -0.548 -0.417 -0.768 -2.208 1989 1.741 -0.287 -0.439 0.425 -1.909 1990 0.194 -0.184 1.930 2.198 -0.354 Note: Subscript z indicates that the variable is standardized by subtracting its mean value and divid- ing the resulting difference by its standard deviation. CG = real GDP growth rate. P. = inflation rate (cpm). C. = current account deficit( ratio to GDP). Fz = changes in public investment (ratio to GDP); Fz is used as an indicator of the direction of fiscal policy . M. = changes in domestic credit of the Bank of Thailand (ratio to GDP), M. is used as an indicator of monetary policy stance. The Role ofMonetarv Policy 199 Figure 8.11. Growth and Inflation, 1971-90 Inflation rate (standardi/ed F,,) 1974 2.5 19800 2 15 * 197A 1981 1 1979 0 0.5 1978 1977 I~~~~~~~~ ~ ~~~~~~~~ ~~ ~~~~~~~~~~~~~~~~~~~~~~~ I I I * -1.5 19X2 * 0 * 015 2 1.5 2 2.5 9823 o ~~~1991) 1972 1975 -U5 1983 * 1976 1989 * 198s 1985* *19860i *1987 19850 @1986 o18 *1971 1984 -1.5- GDDP growth rate, standardized (Gz) Source: Table 8.10. Growth and Monetary Policy Stance We summarize the relationship between monetary growth and income growth in figure 8.12, where Mz is shown together with G. Countercyclical observations will lie in the northwest and southeast quadrants-a negative relationship between growth and monetary stance. Procyclical (accommodative) observations will lie in the other two quadrants-a positive relationship. Inspection indicates that mone- tary policy can be characterized as both accommodative and countercyclical for different periods, but that countercyclical observations predominate. Procyclical observations in the northeast quadrant of the diagram are ob- served for 1973. and during the period of 1976-78. In 1973, the positive external shocks in the price of export commodities such as rice, rubber, and maize pro- duced a surplus in the current account (corresponding to Cz negative in table 8.10). The boom in export demand created strong growth and a rising price level (in the northeastern quadrant of figure 8.1 1). The Bank of Thailand accommodated the demand for credits from commercial banks as a result of the rapid growth in ex- ports. During the period 1976-78, the growth rate in credits extended by the cen- 200 Thailand's Macroeconomic Miracle: Stable Adjustnent and Sustained Growth Figure 8.12. Monetary Policy Stance and Growth, 1971-90 2 t Monetary policy stance (MA) 1979. 1.5 1980 0 * 1978 19820 1981 191 1985 6 19710 0.5 _ 1983 * 1977 19720 1975 1976 0 1973 -1.5 -1 -OS e OS lS 2.5 1986 0 5° 1990 41 1984 0 1987 _1 19740 -1.5- -2 1989 * 19880 -2.5 GDP growth rate, standardized (Gz) Source: Table 8. 10. tral bank to finance the budget deficit was enormous. While inflation in 1976 was below its average, it was slightly above average in 1977 and 1978. The southwestern region of figure 8.12 includes two years, 1974 and 1986. In 1974, Thailand suffered a cost- and push-led recession as a result of the first oil price shock. Inflation, rather than unemployment, was considered the primary pol- icy concern, however, and the monetary policy response was restrictive rather than expansive. Although 1986 was a year of both low inflation and low economic growth, monetary policy was restrictive, according to our measure. As discussed in chapter 7, in 1986 the Bank of Thailand's credits to the government declined substantially because of a significant shift in the method of budget deficit financ- ing from inflationary to noninflationary methods. This fact, more closely related to long-term fiscal issues than to stabilization policy, explains why the growth rate of domestic credit was unusually low in 1986. It should be noted that the response of monetary policy during the second oil price shock was different from the first. The monetary response was restrictive during the first oil price shock in 1974, when inflationary effects on price and out- put lost were more severe than after the second shock. But monetary policy was expansionary between 1979 and 1981, a period of low growth and high inflation. Thte R/le of Monetarv Polici 20] During the second oil price shock, the central bank gave priority to stabilizing in- come rather than maintaining price stability, and this decision led to a deteriora- tion in the current account deficit (table 8.10). The growth rate of (GDP could have been even lower had the Bank of Thailand adopted a similar strategy to that em- ployed during the first shock. The northwestern region of figure 8.12 contains the period 1987-90. when Thailand enjoyed a period of high growth and relatively low inflation. Although inflation during this period was below its mean, it was creeping upward. The mon- etary authorities anticipated a rising price level and adopted an anticyclica1 policy response, thereby reducing the growth rate of domestic credit. The rapid income growth was driven by exports and private investment, implying increased demand for money, but the Bank of Thailand did not exacerbate inflation by fully accom- modating this growth with its own credit supply. In some cases, monetary policy has been procyclical with respect to income growth: the commodity boom year of 1973 and the period between 1976 and 1978 were periods of high growth and monetary' expansion, while the years 1974 and 1986 recorded low growth and monetary contractioni. Aside from these episodes, the negative overall relationship between normalized GDP growth and monetary stance suggests that monetary policy was on balance countercyclical with respect to income; expansionary measures were associated with a slowdown in economic growth, and contractionary measures with strong growth. Inflation and Monetary Policy Stance The relationship between monetary stance and inflation is summarized in figure 8.13. Again, the two variables are expected to have a negative relationship. but the situation is more mixed than that seen for income stabilization. The number of countercyclical observations (those in the northwest and southeast quadrants) is exceeded by the number of procyclical observations (northeast and southwest quadrants). The northeastern region of the diagram contains the years 1977 and 1978. In- flation rates were marginally higher than average, and the Bank of Thailand could afford to expand its credit to accommodate the increase in the demand for money. More important, this was a period of highly authoritarian military government. Having seized power through a coup. following the democratic period of 1973- 76, this government was not about to have its authority challenged by bureaucrats. The independence of the Bank of Thailand was significantly weakened during this period and was not restored until the Prem government of 1981. Price stability was not high on the monetary policy agenda in the late 1970s and credits extended to financial institutions and the central government increased substantially in those years. In 1980, when inflation remained high as the result of the second oil price shock, the monetary policy stance was still expansionary. This monetary policy re- sponse in 1980 was in sharp contrast to the response in 1974, the sole point in the 202 Thailand's Macroeconomic Miracle. Stable Adjustment atid Sustained Growth Figure 8.13. Monetary Policy Stance and Inflation, 1971-90 Morietary policy stance (Afz, 2 + 1.5 0 1979 * 1978 * 1980 1982 S 197'. *1983 1977 @1981 41985" 1972 1976 0 - 1975 * 1973 -1.5 -1 -0.5 09 1 1.5 2 2.5 3 19860 ° - 1984 * 0 1987 0 1974 -1.5 19890-2 - 19880 -2.5 Inflation rate, standardized (Pz) Source: Table 8.10. southeastern region, when the monetary policy was contractionary. In 1980 the ex- pansion of the monetary base was due mainly to the large increase in the money- financed deficit through borrowing from the Bank of Thailand. When the new gov- emnment came to power in 1981, inflationary means of deficit financing were dis- couraged strongly. Figure 8.13 also indicates the pattern of the monetary policy response in the high inflation years of 1973, 1974, 1981, and 1979. Here, except for 1980, mone- tary contractions were larger when inflation was most severe. Inflation in 1979 was not as serious as in 1981, and the monetary response was not as strong. In the northeastern region of figure 8.13, inflation was not considered a major problem and the central bank gave priority to achieving a higher growth rate by increasing the level of domestic lending. In the southwestern region, the price level increased less rapidly than on av- erage, but the policy stance was still restrictive. As noted earlier, inflation gradu- ally gained momentum during the late 1980s, with the rapid expansion in the Thai economy. The expectation of an overheated economy prompted the central bank to slow the increase in bank credit. The years 1984 and 1990 also saw a huge cur- fhu Role of Monetary Policv 203 rent account deficit, and priority shifted to some extent from internal stability to external stability. The central bank attempted to control its lending to avoid an ex- cessive current account deficit. The reduction in money deficit financing in 1986 and 1987 also prompted the central bank to slow down lending. On balance, the contribution of monetary policy to short-run stabilization of the price level is significantly less impressive than its contribution to the stabiliza- tion of income growth. This is somewhat surprising, because of Thailand's obvi- ous success in using monetary policy to control inflation over the long run. The evidence indicates that monetary policy has been much less effective in control- ling inflation in the short run. Conclusions To a limited extent, discretionary monetary policy in Thailand operated effectively in the short run in a stabilizing. countercyclical manner over the period 1970-90. This occurred even though Thailand simultaneously pursued a fixed exchange rate. The key to the apparent paradox is the limited degree of capital mobility that existed in Thailand over most of these two decades. The operation of discretionary monetary policy was limited by offsetting capital flows, and the implementation of monetary policy was complicated by lags in the response of the key target variables-inflation, income growth, and external balance-to changes in monetary policy. Short-term deviations from interest rate parity occurred because of the limited substitutability between domestic and foreign assets and liabilities, capital, and foreign exchange controls, as well as uncertainty regarding the exchange rate and the possibility of changes to capital controls. The forward premium exchange rate adjusted to the interest rate gap, thus providing an insulating mechanism that dis- couraged offsetting capital flows. The withholding tax rate was another insulating instrument available for regulating capital flows. The outcome was that long-term capital flows, which have represented a large proportion of total capital flows, were insensitive to interest rate differentials. Although the interest rate elasticity of short-term flows was statistically significant. it was still small. Evidence on the behavior of the components of the monetary base suggests that the Thai monetary authorities were able to exert some short-term control over domestic interest rates. Although capital movements did occur in response to dif- ferences between domestic and foreign interest rates, the degree of international capital mobility in Thailand has been far from perfect. The Bank of Thailand has thus possessed some degree of autonomy in conducting monetary policy, in spite of its fixed exchange rate policy. Short-run discretionary monetary policy has apparently contributed to Thai- land's macroeconomic stability. Surprisingly, the success of monetary policy in achieving short-run economic stabilization is clearer in the case of income stabi- lization than price-level stabilization. Nevertheless, Thai monetary policy is not a 204 Thailand s Macroeaonomic Miracle: Stable Adjusmrnen tand Sustaiined Growth story of short-run discretionary adjustment at all, but one of long-run monetary discipline in pursuit of long-term price stability. In this, monetary policy operated jointly with the long-run fiscal discipline described in chapter 7. The coordination of monetary policy with fiscal policy was particularly im- portant during the economic boom of the late 1 980s. The reason for the moderate increase in the monetary base during this period, despite the large surplus in the balance of payments, was that the government also ran a budget surplus. Instead of spending the increased revenue resulting from the boom, the government used it to retire external government debt, thus mitigating the expansion of the mone- tary base that would otherwise have been induced by the rising level of interna- tional reserves. Thailand's interest rate controls caused serious problems during periods of rapid inflation. Following each of the oil price shocks, domestic inflation rose to about 20 percent. In combination with interest rate controls, which were adjusted belatedly and insufficiently, this pushed real interest rates to negative levels and undermined private savings. Long-term damage was averted only because infla- tion was quickly brought under control. This must be considered an impressive achievement, aided as it was by the credibility of monetary policy in Thailand. Financial liberalization has been undertaken gradually in Thailand. All kinds of interest rate ceilings had been abolished by 1992. But even with controls, as long as inflation remained low, real interest rates remained positive and full liber- alization could reasonably proceed slowly. In the meantime, it was thought that the Bank of Thailand should strengthen its supervision of commercial banks to ensure that financial liberalization did not create instability in the banking system. The Thai experience must be considered a reasonably successful example of the grad- ualist approach to financial liberalization. With the abolition of the ceiling interest rate on lending, the central bank lost another policy instrument for controlling the domestic interest rate. Less direct means of affecting the domestic interest rate still exist, by altering the liquidity conditions in the money markets through loan windows and the markets for repur- chasing government bonds. The effectiveness of such sterilization is limited, how- ever, by the avai]ability of government bonds. Furthermore, in the long run the central bank cannot sterilize the effects of capital inflows as long as capital mobil- ity is near-perfect and the exchange rate remains essentially fixed. The liberalization of capital market controls in the early 1990s has removed much of the basis for Thailand's monetary autonomy. Financial liberalization can be expected to have important economic benefits and will advance Thailand's attrac- tiveness as a regional financial center. But by allowing international capital to have greater mobility, it will also limit the capacity of the Bank of Thailand to pursue an independent monetary policy. The commitment to a fixed exchange rate has not abated; indeed, some observers have suggested that the determination to defend a stable exchange rate in relation to the U.S. dollar is stronger than ever. The monetary strategy of the bank must adjust to the policy-induced changes in capital mobility unless Thailand decides instead to pursue a more flexible exchange rate system. Chapter Nine The Role of Exchange Rate Policy Thailand's record of monetary conservatism has been accompanied by a great re- luctance to use the exchange rate as an instrument of discretionary macroeconom- ic management. From the 1950s until 1984 the baht was pegged to the U.S. dollar. Exchange rate policy since 1984 has been officially described as a managed float, but approximate fixity to the U.S. dollar still continiues. Devaluation, which in Thailand means devaluation in relation to the U.S. dollar, has been seen as a capit- ulation to inflation. It has occurred infrequently, in response to serious problems of external balance, and then usually only when other corrective measures have al- ready been tried and failed. It is essential to recognize that exchange rate policy and monetary policy are intimately linked: Thailand'\ record of exchange rate sta- bility has been possible only because of the sustained monetary restraint described in chapter 8. This chapter examines exchange rate management in Thailand and its rela- tionship to the movement of relative prices. This analysis is motivated by the so- called Australian model of the balance of payments, which stresses the distinction between traded and nontraded goods. The results strongly confirm that monetary and exchange rate policy in Thailand have the power to influence relative domestic prices. They also confirm the relevance of the traded- versus nontraded-goods an- alytical model as a tool for understanding issues onl macroeconomic adjustment. The Record of Exchange Rate Policy From 1961 to 1980 Thailand's exchange rate remained at roughly 20 baht per U.S. dollar. The long history of stable exchange between the baht and the U.S. dollar had been accepted as a normnal feature of Thailand's international trade. Succes- sive devaluations in the early 1980s raised this rate to 27 baht to the dollar at the end of 1984 (table 9.1). 20 J 206 Thailand s Macroec onomic Miracle: Stable Adjustment and Sustained Growth Table 9.1 Dates and Magnitudes of Devaluations (baht/U.S. dollar rate) Date Percentage increase May 1981 1.1 July 1981 8.7 November 1984 14.9 December 1985 1.9 Source: Bank of Thailand. Timing of Devaluations For summary purposes, these events may be collapsed into two significant deval- uations: one in mid-1981 of roughly 10 percent; and one in late 1984 of roughly 15 percent. The objectives of the 1981 and 1984 devaluations were to reduce the existing balance of payments deficits. Economists at the Bank of Thailand had concluded that the cause of the deficit lay in the overvalued baht. Although the rate of exchange between the baht and U.S. dollar had scarcely changed, the dollar it- self appreciated in relation to other currencies. Thus the baht, pulled by a strong dollar, appreciated against other currencies as well. Figure 9.1 shows the move- ment of the effective exchange rate between the baht and the currencies of Thai- land's seven major trading partners: Japan, the United States, the United Kingdom, the former Federal Republic of Germany, Hong Kong, Malaysia, and Singapore. The series shown uses quarterly data, and the country weights are total trade shares, imports plus exports, over the fifteen-year period 1976-90. Note that the devaluations of 1981 and 1984 were small in comparison with the exogenous movements in Thailand's effective exchange rate induced by movements in other countries' exchange rates in relation to the U.S. dollar. Second, the two devalua- tions were each immediately preceded by appreciations of the effective exchange rate of the baht, induced by appreciations of the U.S. dollar in relation to other currencies. In the year before the 1981 devaluation, the baht depreciated marginally in re- lation to the U.S. dollar, Japanese yen, and Singapore dollar, but it appreciated substantially when compared with the British pound and the German mark. For the same reasons, one year before the 1984 devaluation the appreciating dollar caused the baht to appreciate significantly against the pound and the mark. The de- valuations of the baht in 1981 and 1984 can thus be seen as corrections for the overvalued baht. They were policy responses intended to restore Thailand's inter- national competitiveness, but they were also long-term responses to the deteriorat- ing terms of trade that began in 1974 with the first oil shock. A further decline in the terms of trade occurred in 1985. In early December 1985 the baht was devalued further by 1.9 percent. The objective of this minideval- uation was again to recapture Thailand's competitiveness. One year before the 1985 The Role of Exchange Rate Policy 207 Figure 9.1. Nominal and Real Effective Exchange Rates, 1970-90 Baht/U.S. dollar 30 25 15 10{ 50 0 I I -r I I|i 1970 1972 1974 1976 1978 198( 1982 1984 1986 1988 199(1 Nominal rate - - -…- -Real effective rate t 1970 = 20.8) Source: Bank of Thailand, Monthlv Bulletin, various issues, and authors' calculations. devaluation the baht depreciated against the pound. mark, and yen and marginally appreciated against the U.S. dollar, Hong Kong dollar, and the Malaysian ringgit. The 1I.1 percent devaluation of May 1981 reflected a decision on the part of the monetary authorities to devalue the baht by small amounts, but this was to be done repeatedly, rather than by devaluing by a single large percentage. It was believed that the latter course would generate greater public protest. Although this minidevalua- tion caused little such protest, it did result in devaluation speculation. As was widely recognized, the 1I.1 percent devaluation was not commensurate with the appreciation of the dollar and the baht against other currencies (figure 9. 1). The exchange equal- ization fund was required to continue selling dollars for fourteen working days until the baht was devalued once again, this time by 8.7 percent. During these fourteen days, there were seldom any sellers of dollars to the exchange equalization fund. The deterioration in the terms of trade and the overvalued baht led to a persis- tent balance of trade deficit before the 1981 and 1984 devaluations. In particular. the 1983 current account deficit was alarmingly high (see table 3.4), and the do- mestic rate of interest was higher than the world rate by a full five percentage points. This contrasted with the situation in 1981, when the domestic rate of inter- est was about 2.5 percent below the world money market rate. In early 1981 the domestic money market rate was about 2.5 percentage points below the LIBOR rate 208 Thailands Macroeconomic Miracle: Stable Adjustment anid Sustained Growth and capital began to flow out in large amounts. The rise in the foreign rate of in- terest can be seen as a form of external shock, adversely affecting the balance of payments position. Since a ceiling rate of interest was in place, establishing a pos- itive rate of interest-and thereby preventing capital outflow-was seen as a nec- essary policy response. The appreciation of the U.S. dollar can be seen as another type of external shock, in addition to those discussed in chapter 5. Since the baht was pegged to the U.S. dollar, its value also appreciated. In addition to the expectation of further devaluation following the 1.1 percent devaluation of May 1981, the rising foreign rate of interest caused commercial banks to rush to pay foreign debts in an effort to accumulate foreign assets. Between the May and July devaluations of 1981, the value of foreign assets held by commercial banks increased by 20 percent. The level of Thailand's international reserves declined from 1979 to 1985 (figure 9.2), although if the adequacy of reserves is taken into account, as mea- sured by the number of months of imports covered, the declining trend started long before 1979. Although seven to eight months of reserve-covered imports were quite high, the precipitous drop in early 1981 contributed to the expectation of fur- ther devaluation. The months of import coverage series shown in figure 9.2 pro- vides a potential explanation for the timing of the mid-1981 and late 1984 Figure 9.2. Value of Reserves and Months of Import Coverage, 1975-90 Billions of U.S. dollars Months 16 16 14 billions of dollars 14 12 12 10 10 6 6 4 ~~~~~~~~~~~~~~~~~~~~4 2 2 O L - l l - o 1975 1977 1979 1981 1983 1985 1987 1989 1991 Value of reserves ---------- Months of import coverage (Billions of U.S. dollars) Sourec: Bank of Thailand, MonthlY Blletil, various iasues; IMF, Internationa/ IFinancial Statistics., various issues: and authors' calculations. The Role of Exchange Rate Policy 209 devaluations. In both periods, the number of months of coverage represented by reserves were historically low and had been declining for several quarters. The ab- solute value of reserves (shown in billions of U.S. dollars) had been declining in the quarters before the 1981 devaluations and had been rising only slightly prior to the 1984 devaluation. Moreover, the premium for the forward rate of the dollar was exceptionally high in the first half of 1981 and in early 1984. In the cases of both the 1981 and 1984 devaluations, the policy response was applied belatedly and reluctantly. This was true not only of exchange rate policy, but also of interest rate policy, as described in chapter 8. During the first oil shock in 1974. the deposit and loan interest rate ceilings were raised by one percentage point, notwithstanding a 15 percent inflation rate in 1973. As a result, real interest rates were substantially negative. During the second oil price shock, the rate of in- flation increased to 19 percent in 1980, an alarming rate by Thai standards, al- though smaller than the 24 percent inflation of 1974, following the first oil shock. The ceilings on interest rates were adjusted upward by 3 percentage points, but this still produced a negative 6.7 percent real interest rate. As a consequence of Thailand's long-term exchange rate stability in relation to the U.S. dollar, the Thai price level has been closely tied to the U.S. price level. This relationship is closest in the case of the wholesale price indices for the two countries, because traded goods are more dominant in these price indices than is the case for consumer prices (figure 9.3). Thai devaluations had only a small effect on the close relationship between the two price indices. Figure 9.3. Thai and United States' Wholesale Price Indices, 1960-92 Index (1985 100) 140 120 - 100 _ 80 _ 60 - 40 - 20 - 20 1960 1965 1970 1975 1980 1985 1990 Thailand ------ United States Source: Bank of Thailand, Monthly Bulletin. various issues; IMF, International Financial Statistics, various issues; and authors' calculations. 210 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth The Policy Environment of Devaluation Devaluations inevitably have unpopular political consequences. They create hard- ships for those depending on imported raw materials as well those holding debts denominated in foreign currencies. In Thailand, devaluations have usually been employed only after other methods of coping with trading imbalances have al- ready been attempted. These other means have included an increase in import tar- iffs, contractionary fiscal and monetary policies, and credit controls. This response reflects the strong aversion of Thai monetary authorities to the inflation that deval- uation causes. But its cost may well be that when devaluation finally becomes an inevitable means of correcting the balance of payments problem it must then occur at a higher rate than would have been required had action occurred earlier. An example of such a pattern of policy response was the limit on the credit growth of commercial banks that began in 1984. This was a considerable shift in monetary policy since it was the first time that a quantity control was imposed to- gether with a price control (that is, the interest ceiling). The shift in the policy stance was attributed to the conditions set by the IMEN for the stand-by loans agreed in that year (Robinson, Yangho, and Ranjit 1991). Alarmed by the huge current ac- count deficits of 1983, the Bank of Thailand decided to take this drastic measure against the expansion of imports. The commercial banks were subject to the max- imum credit expansion at 18 percent of the 1983 level of credit. The Bank of Thailand's credit ceiling was blamed for causing extensive fail- ures among small and medium-size businesses. FErom January to June 1984 the conservative Thai Farmers Bank reported a 70 percent increase in the number of dishonored checks, compared with the previous year, and an 86 percent increase in their value. In August 1984, Prime Minister Prem Tinsulanonda canceled the credit ceiling after seven months of operation. Before the decision was made, the government had to face mounting criticism from the press, opposition parties, and some parties in the ruling coalition. A special working group, appointed by the Council of Economic Ministers to study the issue, recommended scrapping the credit ceiling on the grounds that the negative political effects outweighed the sup- posed positive economic benefits. The working group included the governor of the Bank of Thailand, Nukul Prachuamoh; secretary of the National Economic and Social Development Board, Snoh Unakul; and the prime minister's chief econom- ic adviser, Virabongsa Ramangkura. The credit ceiling was considered unnecessary since the two largest banks, Bangkok Bank, and Krung Thai Bank wanted to reduce their lending growth ow- ing to the excessive lending in 1982. In the first half of 1984 their lending growth rate fell to only 3.0 and 0.5 percent, respectively. Before implementing the credit ceiling, the Bank of Thailand had contemplated other measures, such as raising banks' reserve requirements and controlling the issuance of letters of credit (LCS). These measures were in fact undertaken at the end of 1983. The banks had to maintain the value of LCs issued between December 1983 and November 1984 at the same level as in the preceding year. When these measures failed to reduce the The Role of E(change Rate Policv 211 balance of trade deficit, a ceiling was imposed on the volume of credit in general. Devaluation was postponed to avoid its adverse political consequences. By apply- ing other indirect measures (such as credit controls) to correct the external deficit, the Bank of Thailand had complicated the problem. Large and unanticipated devaluations bring more resistance and dissatisfac- tion than small and anticipated ones. Nevertheless, small-step devaluation cannot be undertaken frequently since speculation on further devaluation will develop. The November 1984 devaluation was not anticipated. The forward premium for U.S. dollars was very low. Thus 1984 contrasted sharply with the 1981 devaluation in terms of timing and magnitude. There was no massive run on international re- serves. Government officials repeatedly stated that there would be no devaluation since other policies such as the 18 percent credit ceiling were working out. The ceiling was abandoned in August 1984 since credit monetary growth was in line with the target. On October 24, 1984. the government announced a six-month ex- tension of the control of LCS. The combined effect of these measures was an 18 percent reduction in the trade deficit from January to September 1984. That is why the November devaluation surprised most people. The timing of policy implementation was crucial. Inflation was high in 1980 and 1981, at 19.7 and 12.7 percent, respectively. As a result, the authorities were reluctant to devalue at a high rate. The 1.1 percent devaluation in May 1981 can be viewed as a preliminary corrective measure to avoid fueling inflation. In con- trast to the inflation rates of 1980 and 1981, those in 1983 and 1984 were only 3.8 and 0.9 percent, respectively. The decision to devalue by 14.9 percent, the highest in modern Thai history, was thus considered acceptable. At the time of the 8.7 percent devaluation in July 1981, the opposition parties were strong and Parliament was in session. Thus it was easy to form a coalition and use Parliament as a forum to attack the government in protest at the devalua- tion. In contrast, in November 1984. when the 14.9 percent devaluation was an- nounced, Parliament was in recess. It was difficult for political parties to oppose the government. In any case, the opposition parties were weak and unpopular. In 1984, the government had sufficient time to make its case clear to the public. The redistributive effects of the devaluation were also involved in the timing of its announcement. At the time of the July 1981 devaluation, Thailand had accu- mulated unusually large stocks of unsold agricultural products such as rice, cassa- va, maize, and rubber. Exporters were waiting for an upturn in the world prices, while importers tried to pay the bills that high interest rates had imposed on them. Devaluation was seen as a means of helping exporters dispose of their unsold stocks. Finance Minister Sommai Hoontrakul announced that the November 1984 devaluation would help the farmers, since their products were still in their posses- sion. The annual export crop season was from December to March. If devaluation took place after November, the windfall would accrue primarily to middlemen. The devaluation in 1981 taught economic policy planners several lessons. The political cost of the adjustment to a realistic exchange rate led the government to postpone devaluation. In addition, opposition parties as well as some members of the 212 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth coalition attacked the devaluation. Fear of inflation was one of their great concerns, although some members were opposed to devaluation because they failed to under- stand devaluation per se. The record of Parliament in July 1981 also indicated that some members were afraid that the budget would be invalid since the value of the baht had eroded by 8.7 percent. Others suggested alternative policies such as totally banning imports, increasing tariffs, or asking the Bank of Thailand to assume the monopoly role of foreign exchange dealer. Other members regarded devaluation as equivalent to lese majeste since the value of the bank notes that bear the king's pic- ture had been reduced. There was also concerted opposition from the press. The dep- uty finance minister, Paichitr Uathavikul, pictured as the main culprit, resigned. The 1981 devaluation was opposed mainly by politicians. The 1984 devaluation brought a confrontation between the prime minister and the commander in chief of the army, General Athit Kamlangek. In a public broadcast, the army commander called for the cancellation of the devaluation and the reshuffling of the cabinet, which implied the removal of the finance minister responsible for the devaluation. The finance minister had earlier strongly opposed the military's plan to purchase six- teen F16-A tactical fighters, arguing that it would substantially increase Thailand's external public debt. He had thus incurred the hostility of the military leadership. Devaluation was seen as a further threat to the plan to modernize the Thai military since the jet fighters would now cost 14.9 percent more in nominal baht terms. The real possibility of a coup against the government made the overt opposi- tion of the military particularly ominous. Prime Minister Prem had acknowledged his involvement in the decisionmaking process leading to the devaluation. He also strongly defended the devaluation on the grounds that it would maintain the coun- try's financial stability. Prem's position was therefore also under attack, but he lat- er gained the public's support, as well as that of other generals. Prem satisfied the military by promising to find the 3 billion baht needed to cover the impact of the devaluation on the purchasing of military equipment. In addition, he promised that the government would control the price of oil and some other necessities. These latter measures were designed primarily to placate the nonmilitary opposition to the devaluation. Five days later, the army comnmander backed away from his op- position, announcing that the dispute had all been "a misunderstanding." Despite some strikes of railworkers, coup rumors, and attempts by the oppo- sition to call an extraordinary session of Parliament to debate the devaluation, the government survived. Devaluations are serious matters in Thailand; they are not undertaken lightly. Effects of Devaluation: Wages, Balance of Payments, Inflation Wage earners tend to be worse off as a result of devaluation if their money wage rate does not rise by the same proportion as the inflation rate induced by devalua- tion. The effectiveness of devaluation in improving the trade balance, however, de- The Role of Exchange Rate Policy 213 pends in part on the extent to which the real wage is reduced. Such a reduction is required, but for this to occur nominal wages must not rise by the full extent of the price rise caused by the devaluation. Labor unions in Thailand have not been a powerful pressure group as far as wage determination is concerned, but minimum wage regulations do have an impact on rates of inflation. The Prem government increased the minimum wage rate by 12 and 6 percent in the years following the 1981 and 1984 devaluations, respectively. Since the rate of inflation was high in 1981, the real minimum wage rate decreased marginally; however, the adjustment in the minimum wage rate in January 1985 represented a real increase of 4.3 per- cent over the previous year. As table 9.2 suggests, increases in the minimum wage were usually small, es- pecially the adjustments following devaluation. Consequently, the inflationary ef- fects of devaluation were contained, in that the government was able to hold inflationary expectations below the percentage change in the exchange rate. From January 1985 to December 1989, the minimum wage was raised only once, by 4.3 percent in 1987. Inflationary expectations can also be controlled by other methods, accompanying devaluations, such as freezing electricity tariffs, reducing import duties and income taxes, and setting maximum prices for vital commodities. After the devaluation of July 1981, devaluation speculation and the run on dollar inter- national reserves came to a halt. The forward premium of the dollar declined to its normal level. More important, the devaluation induced capital inflows and im- proved the current account imbalance considerably. From figure 9.2, it can be seen that the level of international reserves, measured in months of import coverage, re- versed its declining trend. Table 9.2 Minimum Wage Rate in Bangkok Area, 1977-87 Wage rate Percentage Real wage Percentage Effectivea (baht/dav) increase rateb change 1977 28 12.0 39.6 6.0 1978 35 25.0 45.6 15.2 1979 45 28.6 53.7 17.6 1980 54 20.0 54.0 0.6 1981 61 12.3 53.3 -1.3 1982 64 4.9 51.9 -2.6 1983 66 3.1 51.8 -0.2 1985 70 6.1 54.0 4.3 1987 73 4.3 54.9 1.7 a. The effective date usually was on I October, except in 1985 and 1987. the effective dates were set on January I and April I, respectively. b. The real wage rate at 1980 price level calculated from the cpi. Source: Data from the Interior Ministry. 214 Thailand s Ma(croeconomi MiraL le: Stable AdjuNstnen! and Sustained Gnosth The short-run effect of the July 1981 devaluation is shown in table 9.3, which measures the percentage change of the components of the balance of payments six months before and after the devaluation. Whereas the value of exports rose sub- stantially, the value of imports declined, compared with the corresponding period a year before, and the result was an improvement in the current account. The au- thorities have tended to overestimate the inflationary impact of devaluation. As ta- ble 9.4 indicates, the actual inflation rates in 1981 and 1985 were 2 to 3 percentage points below the predicted rates. It is not certain whether the authorities would have postponed devaluation had they correctly estimated the impact of this mea- sure on inflation. Table 9.3 Short-Run Effects of the July 1981 Devaluation Item affected January-June 1981 July-December 1981 Percentage change Exports 8. 1 22.0 Imports 20.2 9.8 Imported oil 14.3 7.5 Change in millions oj/bnht Trade balance -12.3 +5.( Balance of payments -19.7 +17.1 Note: Change from the corresponding period in the previotu year. Source: Bank of Thailand, Monthh/v Buletin, Februars 1982. Table 9.4 Government Expectations of Devaluation Consequences (billions o/ ha/it) Devaluation Predicted conscque!tce Actual consequence July1981 (8.7 perctent) Inflation (percent) 14-15 12.7a Balance of payments -It) +2.5 Balance of trade -68 -65.8 November 1985 (14.9 percent) Inflation (percent) 5-6 2.4b Balance of payments 0 +1(0.6 Balance of trade 70 -68.8 a. 198( = 19.7. b. 1984 = ().9. Sourc e: The forecast for 1981 was made publicly h\ the Minister of Finaice in the Parliament. The forecast tor 1985 was made by the Bank of Thailand. The R(Ie (f Erch)nge Rate, PoliCv 215 Post-1984 Exchange Rate Setting Since the November 1984 devaluation, the Thai exchange rate system has official- ly been described as a managed float. The baht has been pegged to a basket of cur- rencies, the composition of which is secret. From the stability of the baht/U.S. dollar rate, in spite of the volatility of the U.S. dollar in relation to most other cur- rencies, it is obvious that the dollar represents a large share of the basket. It has been conjectured among Thai economists that the share of the U.S. dollar within the basket may have increased since the basket system was introduced. The anal- ysis that follows attempts to determine the implied composition of the basket and its change over time by studying the behavior of exchange rate data since 1985. Exchange rate data for ten countries were used in the regression analysis. These are the ten major trading partners of Thailand: Australia, Belgium, France, Germany, Hong Kong, Japan, Malaysia, Singapore, United Kingdoin, and the United States., It is hypothesized that the baht is pegged to a basket of currencies whose composition by shares changes linearly over time. The share coefficient for each currency in the basket therefore has two components: an intercept component for that currency and a compo- nent that summarizes the rate of change in that currency's share through time. The share coefficient for each currency is thus detined as (9.1) a= a() + t1t where a' is the overall coefficient for currency i at time t, a is the country inter- cept, and ait is the time-dependent component of the coefficient for country i. In the empirical application of this analysis, the sum of the share coefficients across countries, at . is constrained to be equal to unity at each time period. The results of the regression analysis are summarized in table 9.5. The inde- pendent variables used in explaining the baht/SDR rate are the SDR rates of the Jap- anese yen; Singapore dollar; U.S. dollar; an aggregated "others" variable obtained by using a geometric mean of the SDR rates of the remaining currencies; a dummy variable (to take into account the small devaluation of the baht in November 1985), denoted D below; and time, denoted t. The three main explanatory currencies were manipulated by dividing them by the "others" variable. The coefficient for the oth- ers variable can be recovered by subtracting the sum of the other three explanatory variables from unity. The formula for the regression was thus: (9.2) In( Baht +(-2++ ljD+U,2+(l,(tt)ln (_Yen Others ) Others s Others +(a,) + a] t)ln ( Singapore The regression results confirm that the U.S. dollar is a major component in the basket of currencies and that its share is increasing over time. The estimated implicit weights at the midpoint of the series (January 1989) were U.S. dollar, 66 percent; Japanese yen, 9 percent; Singapore dollar, 16 percent; and others, 9 per- 2/6 Thailands Macroeconomic Miracle: Stable Adju.stment and Su.tained Growth Table 9.5 Estimated Composition of Exchange Rate Setting Basket Variables Coefficient valuea Currencv intertceptb Currency slope Constant 2.45 (29.36) Dummy 0.02 (4.70) Time 0.24 (0.13) U.S. dollar 0.66 0.52 0.28 (26.31) (13.81) (6.90) Japanese yen 0.09 0.07 0.04 (10.02) (2.92) (0.90) Singapore dollar 0.16 0.20 -0.06 (4.91) (4.86) (-1.56) a. Coefficient value is the value of a, in equation 9.1 and is equal to the currency share evaluated at the sample mean. b. Currency intercept is the value of ao in equation 9. 1. Note: R2 = 0.99; Durbin-Watson statistic = 1.83; and log of likelihood function = 386.25. Source: Authors' calculations with data drawn from IMw, Intrnational Financial Statistics, various issues. cent. The estimated shares for the U.S. dollar, Japanese yen, and Singapore dollar were all significantly different from zero at the 5 percent level of significance. Fig- ure 9.4 simulates the composition of the exchange rate setting basket implied by the results shown in table 9.5. The figure clearly demonstrates the growing impor- tance of the U.S. dollar in the currency basket to which the Thai baht is pegged. The estimated share for the U.S. dollar rose from 52 percent at the beginning of the data period (January 1985) to 79 percent at its end point (December 1992). Parallel to the growth of the U.S. dollar in the basket is the decline in the impor- tance of other currencies. At the same time, the Japanese yen and the Singapore dollar have maintained steady shares in the basket. Exchange Rate Policy and Relative Prices What role did exchange rate and demand management policy have in directing the macroeconomic adjustments Thailand made through the 1970s and 1980s? The discussion now turns to the role of relative price movements within Thailand in regulating the adjustment process. A simple theoretical model is used to illustrate the respective roles of external shocks, exchange rate policy, and domestic relative prices in balance of payments adjustment. Attention is also given to the actual be- The Role of Erchange Rate Policv 2/7 Figure 9.4. Simulated Composition of Exchange Rate Setting Basket, 1985-93 OLS prediction with time trend (percent) 100 80 40) 207_ 1985 1986 1987 1988 1989 199() 1991 1992 1993 O U.S. dollars O Others M Singapore dollars O Japanese yen Source: Authors' calculations. havior of key relative prices within Thailand and to the degree to which their movements explain the course of Thai macroeconomic adjustment. A Model of the Balance of Payments and External Shocks The simple two-commodity model of the balance of payments used in this analysis is known as the Australian model. The fundamental distinction in this theory is be- tween goods that are exchanged internationally, termed tradables, and those that are not, termed nontradables. Many (but not all) primary commodities and manu- factured goods are examples of tradables, while many (but not all) services are ex- amples of nontradables. The point of distinguishing between them is that the nominal domestic prices of tradables are determined by their international prices, the exchange rate, and any subsidies, tariffs, or other taxes that may be applied to them. In contrast, the nominal prices of nontradables are determined by domestic supply and demand conditions. According to the Australian model, a shock to the balance of payments such as a resource boom, a change in the terms of trade, or a transfer of foreign aid, will lead to equilibrating adjustments in the relative prices of traded and nontraded goods. For example, a manufacturing boom that adds to foreign exchange earnings will lead to a reduction in the price ratio of domestic traded goods versus nontrad- ed goods. This price movement induces the domestic adjustments in demand and resource allocation that cause the increased foreign exchange earnings to be ab- sorbed domestically. Furthermore, the Australian model analyzes the mechanism by which policy instruments such as exchange rate adjustments affect the balance 218 Thailand's Macroeconomic Miracle: Stable Adju.smeni and Sustained Growth of payments by focusing on their effects on the domestic relative prices of traded and nontraded goods. The role of the domestic prices of tradables in relation to nontradables in in- fluencing the balance of payments is depicted in figure 9.5. Value-weighted aggre- gated quantities of tradables are shown on the vertical axis and a similar aggregate of nontradables is shown on the horizontal axis. The schedule PP shows produc- tion possibilities for these two categories of goods, and social preferences regard- ing aggregate "absorption" of them are shown by indifference curves such as U1 and UO. They are assumed to have the usual regularity properties. The distinction between investment and consumption is not important for this analysis. Thus for simplicity both are treated as absorption. The crucial difference between tradables and nontradables in this analysis is that the absorption of nontradables must be matched by domestic production, but the same does not apply to tradables. If more of a particular tradable good is ab- Figure 9.5. Tradables, Nontradables, and Balance of Payments Tradables Q T', I P, N N, Nontradables Solurce: Authors' compilation. ThLt Role of Lwhatnge Rttw Policv 219 sorbed than is produced, then this good is a net import. The reverse is a net export. If in total more tradables (by value) are absorbed than are produced, this corre- sponds to a current account deficit, since the difference between absorption and production of tradables must correspond to an excess of net imports over net ex- ports. A current account surplus is the reverse. By definition, nontradables cannot be either imported or exported, so their domestic absorption cannot exceed their production. Current account balance is indicated by point A in figure 9.5. At this point, consumption and production are equal for both tradables and nontradables. The slope of the price line Po gives the prices of nontradables in relation to tradables. Now consider a current account imbalance. We assumte production possibilities to remain at PP. disregarding the schedule QQ tfor the moioenit. Suppose that the price of nontradables in relation to tradables rises abovc P,) . say, to P1. This new price ratio is indicated by the slopes of lines PI and P/. which are drawn parallel to each other. We will discuss the possible causes of this event presently. Production of nontradables now rises and that of tradables ifalls. Pioduction will now occur at point B. At this higher price ratio, however. tradahles will no longer be consumed in the same quantities in which they are produced. The lower price of tradables en- courages their absorption in relation to nontradables, hut has the reverse effect on production. As already explained, a divergence between absorption and production can only occur in the tradables market; absorption and prodluction of nontradables can- not differ. The possible absorption outcotnes that tnight occur under price PI thus lie along a vertical line drawn through the production point. B. These are the points at which the production and absorption of nontradables are equal. The actual so- lution will be that point along this line at which an indifference curve has a slope equal to that of PI, say, point C. At point C, absorption of tradables exceeds their production. The difference is a current account deficit, equal to the vertical dis- tance between point B and C. How might a current account deficit like that occurring at point C arise'? The key lies in the way the prices of tradables and nontradables are determined. Sup- pose that domestic monetary expansion occurs with a fixed exchange rate. The monetary expansion will raise domestic demand and thus increase domestic pric- es. But these price increases will be concentrated in nontradables rather than trad- ables because the nominal prices of tradables are dieternlined by international prices expressed in foreign exchange. the exchange rate, and any subsidies, tariffs, or other taxes that may be present. Thus the monetary expansion bids up the prices of nontradables in relation to tradables and thereby produces a current account deficit. Now consider an external shock such as a reduction in petroleum prices. Be- cause of the importance of petroleum as an interme(liate good, we represent this shock as a vertical shift in the production possibilities frontier. to QQ. That is, ftor a petroleum importer like Thailand we represent a reduction in international pe- troleum prices as a quantum increase in the economy's net capacity to produce 220 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth traded goods. For convenience in the diagram, we choose an external shock that has the same foreign exchange value as the current account deficit arising in our previous analysis. There will now be no current account deficit if production and absorption both occur at point C. Point C is now an equilibrium with zero current account deficit and a price ratio given by the slope of Pl. Several conclusions can be drawn from this analysis. First, the theory is clear on what an increase in the price of an export good in relation to the prices of other traded goods will mean for the relative prices of traded versus nontraded goods. There will be an income effect and a substitution effect, and both will lead to an increased demand for nontradables. The prices of nontradables should rise in re- lation to all tradables prices other than the export good whose price rose. Second, the theory is more equivocal on the effect of an increase in the inter- national price of an importable. When the price of an import good rises, there will be a similar substitution effect to the one discussed above, but an income effect that operates in the reverse direction. Depending on the relative strengths of these two effects, nontradables prices could rise or fall in relation to the prices of trad- ables other than the import good whose price rose. The greater the share of the val- ue of net imports of the good in national income and the greater the income elasticity of demand for nontradables, the greater the likelihood that the relative price of nontradables will fall. Third, assuming that an increase in an import price requires a reduction in the domestic price of nontradables, then under a fixed exchange rate policy inflation at home must be lower than inflation abroad, at least temporarily. As long as the exchange rate remains fixed, inflation abroad determines the rate of increase of tradables prices. For nontradable prices to decline in relation to them, the average price level at home must rise more slowly than the average level abroad. Under a fixed exchange rate policy, this requires lower inflation domestically than the rate abroad, thus putting severe pressure on domestic monetary and fiscal policies. Fourth, this lower-than-normal rate of inflation, consistent with adjustment to the worsening in the terms of trade, is temporary. Once the required domestic price ratio has been attained, inflation at the world rate will sustain the required nontrad- ables/tradables price ratio. Fifth, the Dutch disease phenomenon, under which tradables sectors such as agriculture and manufacturing are harmed by absorption of the benefits of the im- provement in the terms of trade, has a monetary component as well. Suppose there is an increase in export prices. The rate of absorption of the benefits of this im- provement in the terms of trade depends on the rate at which the nontradables/trad- ables price ratio rises, which in turn depends on domestic aggregate demand. The likely relative price effects of a devaluation can be inferred from figure 9.5. A devaluation will raise the nominal prices of tradables and will do so rela- tively quickly. The subsequent behavior of nontradables prices will be strongly af- fected by monetary and fiscal policy. If the initial position was one of an unsustainable current account deficit and the views of those government officials and advisers arguing for monetary and fiscal restraint prevail, then the initial in- The Role o1fExchange Rate Policy 221 crease in the prices of tradables in relation to nontradables may be sustained for some time. But if domestic aggregate demand grows at a higher rate and thus leads to continued rapid inflation, the price ratio will steadily decline from its post-de- valuation peak. Relative Price Movements within Thailand Suppose for a moment that "traded goods" refer to an index of all traded goods, including the ones whose prices changed. This distinction will be important in some cases. In the Thai case, for example, the temporary commodity boorn of 1973 would have caused the domestic prices of those traded goods whose prices increased in international markets to rise in the short run in relation to nontraded goods, and then to decline as the boom passed. But the prices of nonbooming com- modities should fall in relation to the prices of nontraded goods through the Dutch disease mechanism. The longer-term implications of the positive income effects caused by this boom would be to accentuate the decline in the prices of traded goods in relation to those of nontraded goods. A rise in international petroleum prices would ordinarily produce an increase in local petroleum product prices, but then the negative income effects resulting from this deterioration in the terms of trade should produce an increase in traded goods prices in relation to those of nontraded goods. Similar results to this latter affect would be expected from 1979-80 oil price increases and the interest rate in- creases of the early 1980s. To investigate the actual behavior of traded and nontraded goods prices, indi- ces of the two have been constructed as follows. The index of traded goods prices was constructed from thirty-three individual commodities included in the whole- sale price index. Monthly data were obtained from January 1968 to January 1988 for the nominal prices of each of these commodities. This gave 241 observations for each commodity. The particular commodities chosen for inclusion in the index were those that seemed to satisfy the standard economic definition of traded goods-they are traded internationally and their domestic prices are determined by international prices and the exchange rate. The index then took the form: 33 (9.3) Pt= X aiP,I T where PTis the value of the traded goods price index in period t, ai is the weight applying to commodity i in index form in the construction of the wholesale price index, Pi TiS the price of traded commodity i in period t, and Pi Tis the same price in period T. Dividing by the price in period T normalizes the price series for com- modity i and this is necessary because the commodity weights apply to indexed prices, not to nominal prices. June 1976 was chosen fort because this is the month in which the commodity weights were estimated. 222 Thailand s Macroeconomic Miracle. StableAdjustment aind Sustained Growth The index of nontraded goods prices was constructed from forty-two individ- ual components of the consumer price index. They were selected on the criteria that the commodities or services concerned are not traded internationally and that their domestic prices are not controlled. The period of coverage was the same as for the traded commodities discussed above. The index was thus 43 (9.4) pN= - /p j=1 where pN is the value of the nontraded goods price index in period t, and all other terms are analogous to those appearing in (9.3). June 1976 was again chosen for t. Tables 9.6 and 9.7 list the commodities used in the construction of the two indi- ces and their respective weights. The results are shown in figures 9.6 and 9.7. The series are each indexed so that the value in August 1973 is 100. Figure 9.8 shows the behavior of the traded and nontraded goods price ratio P,=P lpN, which is formed from the separate PTand pN series and is thus automatically indexed such that the value in August 1973 is unity. Before discussing figures 9.6 to 9.8, it is important to qualify the methodolo- gy used. Some arbitrariness enters the specification of the set of commodities in- Table 9.6 Composition of Index of Prices of Traded Goods Commodity Index Commodity Index number Commodity name weight number Commodity name weight I Ammonium sulphate 0.061 18 Gunny bag 0.017 2 Tanned leather 0.005 19 Coffee powder 0.005 3 Tetron fiber 0.083 20 Tea 0.0)04 4 Cotton with synthetic 0.099 21 Refrigerator ).005 5 Silk yarn 0.004 22 Cassava (0.125 6 Cotton yarn 0.050 23 Super kenaf 0.002 7 Broken white rice 0.049 24 Kenaf 0.003 8 Farm tractor 0.023 25 Maize 0.016 9 Air conditioner 0.003 26 Coffee beans 0.005 10 Television set 0.007 27 Rice (100%) 0.031 1 1 Shoes 0.003 28 Rice (5% broken) 0.039 12 Briefcase 0.002 29 Steel ).040 13 Kerosene 0.010 30 Paint 0.006 14 Super gasoline 0.055 31 Cement 0.052 15 Regular gasoline 0.060 32 Diesel 0.136 16 Penicillin 0.003 33 Plastic blocks 0.032 17 Shirt 0.002 Source: Department of Business Economics, Ministry of Commerce. Bangkok. The Role of Exchange Rate PolicY 223 Table 9.7 Composition of Index of Prices of Nontraded Goods Commodity Index CommoditY index number Commodith name weight number Commoditv nwane weight I Vermicelli 0.003 22 Duck egg 0.015 2 Pork 0.059 23 Lard oil 0.(1)1 3 Cured meat 0.005 24 School fee 0.054 4 Chicken 0.025 25 Shrimp paste 0.0)03 5 Dressed duck 0.001 26 Salt 0.0()1 6 Serpent-head fish 0.008 27 Soda water (.(()1 7 Catfish 0.016 28 Ice 0.()04 8 Mackerel 0.004 29 Rice curry 0.343 9 Steamed mackerel 0.011 30 Women's patoong ).006 10 Fresh shrimp 0.011 31 Laundry 0.006 l l Dried fish 0.004 32 Mattress 0.()(0 12 Kale 0.007 33 Mosquito net 0.003 13 Chinese mustard 0.005 34 Electricity 0.(88 14 Cucumber 0.003 35 Charcoal 0.021 15 Sweet potato 0.001 36 Water supply 0.0)22 16 Garlic 0.006 37 Hospital fee ).044 17 Fresh chili 0.007 38 Man's haircut (.() 1 18 Fresh peanuts 0.001 39 Woman's haircut 0.M1( 19 Coconut 0.009 40 Bus fare 0.080 20 Pineapple 0.014 41 Cinemaadmission 0.036 21 Chicken egg 0.011 42 Newspaper 0.021 Source: Department of Business Economics, Ministry of Commerce. Bangkok. cluded in the two indices, and the results would presumably change somewhat if these decisions were made differently. Also, the results shown do not exclude commodities whose international prices have been changing. For example, petro- leum-based products are included in the set of traded goods, as are rice. cassava, and other agricultural products whose international prices have been highly vola- tile over the period discussed. Indices could be constructed that excluded these commodities. Figure 9.8 shows the broad features of peaks in the traded/nontraded goods price ratio in 1974 and 1978. The 1974 peak quickly vanished and within two years the price ratio had returned to roughly its value before the peak. The 1978 peak also vanished, but only after it was sustained for almost five years. The two peaks fit the theoretical predictions reasonably well. The 1974 peak corresponds to the first oil shock while the 1978 to 1982 peak corresponds to the second oil shock followed 224 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 9.6. Traded Goods Price Index Monthly, 1968-88, (August 1973 = 100) Index 250 - 200 - 150 _ 100_ January January January January January 1968 1973 1978 1983 1988 Source: Table 9.6 and Department of Business Economics, Ministry of Commerce. by the debt crisis induced by high interest rates. The puzzle is, what caused the rel- ative price declines from 1975 to 1977 and from 1983 to the present? Before that question can be explored, a statistical worry concerning the traded goods price series that we have used needs to be cleared up. An index of traded goods prices excluding petroleum products has been constructed by assigning zero weights to commodities 13 to 15, inclusive, in table 9.6. Otherwise, the rela- tive weights are unchanged. The behavior of this traded goods price index both considered separately and in relation to the nontradables index described previ- ously is similar to those shown in figures 9.6 and 9.8, respectively. Excluding pe- troleum products from the tradables price series does not make any substantive difference to our qualitative results. It is helpful now to turn to two separate nominal price series of traded and nontraded goods prices shown in figures 9.6 and 9.7, respectively. Taking traded goods prices first, we shall employ the usual assumption of the "law of one price": namely, the domestic prices of traded goods are determined by international prices and the exchange rate. The first oil shock followed a primary commodity price boom that began in 1972 (see the terms-of-trade series in table 3.4). Tradables prices rose, and this was reinforced by the subsequent rise in petroleum prices. The commodity boom had vanished by 1976 and the average price of tradables had fallen. By 1976, roughly half the increase in tradables prices occurring between The Role of Exchange Rate Policy 225 Figure 9.7. Nontraded Goods Price Index, 1968-88 (August 1973 = 100) Index 300 - 250- 200 - 150 - 100 - January January January January January 1968 1973 1978 1983 1988 Source: Table 9.7 and Department of Business Economics, Ministry of Commerce. 1972 and the peak in 1974 had vanished. The second oil shock sent the average price of tradables upward again, beginning in 1978. Significant declines in commodity prices occurred in the early 1980s, but their effects on the average prices of tradables were partly counteracted by the four deval- uations we have discussed: 1. 1 percent in May 1981, 8.7 percent in July 1981, 14.9 percent in November 1984, and 1.9 percent in December 1985. It must be stressed that these devaluations are measured in terms of the U.S. dollar only, not in terms of a trade-weighted index of currencies. As figure 9.1 revealed, from early 1978 to the end of 1984 the trade-weighted effective exchange rate of the baht scarcely changed. As for the nontradables prices in figure 9.7, a monetary expansion occurred from 1972 to 1973 (see chapter 8), but the price effects of this expansion were con- fined mainly to tradables, whose prices were rising dramatically in relation to non- tradables. Nontradables prices rose only moderately, even though inflation reached 16 percent in 1973. A severe monetary and fiscal contraction followed, from 1973 to 1974. Inflation peaked at 24 percent in 1974 but had declined dra- matically to only 5 percent by 1975. Nontradables prices continued their steady rate of increase from before the external price shocks. A monetary expansion began in 1975, and in the following year nontradables prices began to rise more rapidly. This was reflected in a rise in the rate of inflation, 226 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Figure 9.8. Traded/Nontraded Goods Relative Price Index, 1968-88 (August 1973 = 100) Index 1.6 1.4- 1.2- 1.0 0.8 0.6 1 1 1 January January January January January 1968 1973 1978 1983 1988 Source: Figures 9.6 and 9.7. from 4 to 7 percent from 1976 to 1977. A monetary, but not a fiscal contraction began in 1978, continuing to 1981. The rate of increase of nontradables prices de- clined accordingly. The second oil shock produced a surge in inflation, but this was concentrated in tradables prices. The recession of 1981 -82 provoked a mon- etary expansion from 1982 to 1983. The prices of nontradables again surged, be- ginning at the end of 1982. One other matter seems worth raising. As noted earlier, the 15 percent deval- uation of November 1984 caught most observers by surprise. From figure 9.8, it does not seem surprising. By late 1984 the price ratio of Thailand's tradables to nontradables had fallen to levels not seen since the early 1970s, except for a brief period in 1977. "Exchange rate protection," a devaluation intended to restore the domestic competitiveness of traded goods producers, may well have played a role. Nevertheless, balance of payments considerations were undoubtedly important in the timing of the devaluation, as was the decline in the rate of inflation to only about I percent in 1984, which made a large devaluation at that time politically more acceptable. 7The, Role o( Exchange Rate Policv 227 Conclusions Since the 1950s Thailand has followed essentially a fixed exchange rate policy with respect to the U.S. dollar. The reluctance to use the exchange rate as an in- strument of discretionary macroeconomic policy has coincided with a long-term commitment to low inflation. The government has been able to maintain a stable exchange rate with respect to the U.S. dollar, in combination with adequate levels of internationial reser-ves, only through monetary and fiscal discipline. The capac- ity to maintain a stable exchange rate has been seen as a barometer of macroeco- nomic stability in Thailand. Devaluations have been seen as a surrender to higher rates of inflation, and when they have occurred they have been politically costly to the government. Significant devaluations with respect to the U.S. dollar occurred in 1981 (roughly 10 percent) and 1984 (roughly 15 percent). These devaluations occurred only when (a) the level of reserves was declininlg, (b) the ratio of traded goods prices to nontraded goods prices within Thailand had declined, and (c) other pol- icy instruments, directed at these matters, had been tried and failed. These deval- uations were a consequence of the external shocks of the 1 970s, together with the failure of monetary policy to adjust to them, especially in the late 1 970s. After the devaluation of 1984. Thailand officially adopted what was described as a "nmanaged float" exchange rate policy. Since then, however, the baht has re- mained closely linked to the U.S. dollar, despite the dollar's volatility with respect to other major currencies. Thailand's exchange rate with respect to currencies oth- er than the U.S. dollar has thus been driven by the exchange rate policy of the Unit- ed States. Wholesale prices in Thailand have closely followed those in the United States. Econometric analysis of relative exchange rate behavior since 1984 shows that the share of the U.S. dollar in the Bank of Thailand's implicit exchange rate setting basket has averaged more than two thirds and has been rising over time. Chapter Ten Conclusions: Thailand's Miracle? A feature of traditional Thai culture, especially apparent in its classical music and dance, is its emphasis on slow, controlled lightness of movement. Sudden, jerky movements are abhorred. Thai macroeconomic policy seems to display a similar characteristic. The emphasis is on steadiness, predictability, and stability. This does not mean laissez-faire. Control is used, but it is kept to a minimum. What is more important is to avoid sudden changes of policy whenever possible. The re- luctance to shift course at short notice has often been described as a weakness of Thai economic policy. In the area of macroeconomic policy, this weakness has been apparent in several instances, especially where interest rate control and ex- change rate adjustment have been involved. But on the whole there can be little doubt that the cautiousness of Thai macroeconomic management has served the nation well. The pattern of Thailand's macroeconomic responses to the major external shocks of the past two decades is of particular interest because of the way in which discretionary changes in government policy and the cautious and conservative economic policy stance of successive administrations of seemingly different ori- entation contributed to the country's stable adjustment. The central question ex- plored in this concluding chapter is how well did Thailand really adjust to the external shocks it experienced, and how miraculous was Thailand's performance, after all? Pattern of Response to Shocks For convenience, Thailand's pattern of response to the shocks of 1970-90 can be divided into three episodes: 228 Conclusions: Thailands Miracle? 229 Episode A. The 1972-74 temporary commodity price boom (shock 1), followed by the 1973-75 oil price increase (shock 2). Episode B. The 1978-79 oil price increase (shock 3). followed by the 1980-82 interest rate increases. Episode C. The 1987-90 boom in the volume of labor-intensive man- ufactured exports (shock 4). The policy responses to the above episodes have reflected three primary goals: the countercyclical stabilization of national income; avoidance of inflation; and protection of Thailand's international reserves under its fixed exchange rate system. The first of these is a short-term goal, while the other two are medium- term goals; but they are potentially in conflict. The pattern of events in response to episodes A and B was similar in several respects. Both shocks reduced income from trend, and both increased inflation in at least one year to about 20 percent. This inflation v as concentrated almost ex- clusively in the prices of traded goods, determined internationally. There was an increase in money supply sufficient to accommodate these price rises without re- quiring nominal declines in nontradable prices, but not sufficient to increase non- tradable prices. In response to this inflation, the authorities contracted monetary policy, and this action produced a recession that lasted for about two years. These were the years 1974-75 (shock 2) and 1980-82 (shock 3). The rate of inflation dropped quickly. The price ratio of tradables to nontradables had almost doubled since the year before the shock. Following that recession, there was a monetary expansion that lasted for two or three years. This expansion caused the prices of nontradables to rise in relation to tradables to levels seen before the respective shocks. At that time-late 1977 to early 1978 in the case of episode A and 1983 to 1984 in episode B-a devaluation was a possibility. The trade deficit had become high and international reserves, measured in months of imports, were falling. At this point, the pattern of events differs in an important way. A devaluation occurred in 1984 (following Episode B) but not in 1978 (the corresponding date following Episode A). Why the difference'? In 1978 Thailand had a relatively new and very conservative military government that was especially averse to devalua- tion. Inflation was already high (8.4 percent), partly as a result of the monetary ex- pansion promoted by this very government. Reserves were falling but still corresponded to six months of import coverage. The government decided to de- fend the exchange rate, even when the second OPE( oil price shock intervened (shock 3). In 1981, when reserves fell to only 2.2 months of import coverage, there was a new, less doctrinaire government. Two devaluations occurred, amounting to roughly 10 percent. In 1984, inflation was low (I percent) and reserves fell to only 2.4 months of import coverage. After trying other policy instruments unsuccess- fully, the government chose a politically opportune moment to devalue by roughly 15 percent. This devaluation provoked a confrontation between the government 230 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth and the military leadership. Although the devaluation was not rescinded, as initial- ly demanded by the military, the government was obliged to compensate the mil- itary for the impact of the devaluation on the cost of its imported equipment. Episode C had a more important internal component than A or B, but it was still driven primarily by external events: the migration of labor-intensive manufac- turing away from the newly industrializing economies of East Asia (Korea, Taiwan (China), Hong Kong, and Singapore), in response to rising wages. Thailand's boom was fueled by unprecedented levels of foreign and domestic investment. Overheating was avoided by means of a fiscal contraction in which large fiscal sur- pluses offset the domestic monetary expansion that would otherwise have resulted from the conversion into baht of the large current account surpluses in the hands of the private sector. After 1990 this boom slowed, because of domestic political conflict and also the growing congestion of infrastructure facilities overburdened by the pressures resulting from the boom. Were Fiscal and Monetary Policy Stabilizing? To what extent has government policy been responsible for Thailand's economic stability? This is a central question of our study and we have attempted to explore it in depth. The ex-post fiscal deficit of the government has behaved countercyclically in the short run. This is also true of the main instrument of Thai monetary policy, the bank lending rate set by the Bank of Thailand. Both fiscal and monetary aggre- gates seem to have been expansionary during periods of low inflation and low growth, and contractionary in the reverse. Two competing interpretations of these facts are possible. On one hand, it may be that fiscal and monetary and policy have been implemented in a deliberate countercyclical fashion, in line with the declared objectives of Thailand's macro- economic managers; the observed stabilization would not have occurred in the ab- sence of these policy decisions. On the other hand, it may be that automatic stabilizing adjustments have operated to induce the observed adjustments to the budget deficit and the observed interest rate adjustments; these automatic stabiliz- ers, rather than deliberate policy decisions, resulted in the observed stabilization. In the case of fiscal policy, the argument for the first interpretation-policy- induced stabilization-would stress the importance of discretionary variations in public expenditure and their role in determining the budget deficit. Less emphasis would be placed on policy-induced variations in revenue receipts, but the possibil- ity that they could also play a role would not be excluded. In the case of monetary policy, this line of argument would emphasize impediments to capital mobility, at least in the short run. The existence of persistent differentials between Thai and foreign real interest rates for prolonged periods and their positive relation with the Bank's monetary stance lends further support to this view. It would point out that Conclusions: Thailand's Miracle? 231 domestic lending and borrowing rates have been regulated by ceilings imposed by the Bank of Thailand and that actual borrowing and lending rates have been equal to these ceiling rates during almost all of the period under discussion. It would also point to the rhetoric of Bank of Thailand officials, emphasizing a countercyclical role in their behavior. The argument in support of the second interpretation-automatic stabiliza- tion-would explain the observed negative correlation between the size of the budget deficit and the rate of income growth by pointing to structural features of the fiscal system driving changes in government revenue and expenditure. When incomes grow. government revenue grows, through commodity taxes and trade taxes, while expenditures may be less sensitive to income growth. The greatest emphasis would be placed upon the stabilizing behavior of government revenues, rather than expenditures. The observed negative correlation between the govern- ment budget deficit and the trade deficit would be interpreted as a reflection of this automatic stabilization because of the government's reliance on trade taxes and because the trade deficit increases with income growth. The main argument for automatic stabilization in the case of monetary policy is that in the presence of a fixed exchange rate the domestic money supply is at least partly endogenous, depending on the degree of international capital mobility. Despite the impediments to capital mobility, in this view international capital movements are still possible to a sufficient extent to enable movements in the un- controllable components of the monetary base to dominate the controllable ones. The behavior of the Bank of Thailand's lending rate would thus be interpreted as a response to endogenous changes in the supply of money, rather than as a reflec- tion of exogenous (policy-determined) changes in its supply. Somewhat different- ly, it may be argued that the reason for the positive correlation between income growth and changes in the bank's lending rate is that when incomes grow the de- mand for money rises, and interest rates rise accordingly. Under this argument, the bank's rate is thus again a reflection of market forces. In our view, the evidence supports the interpretation that a more important sta- bilizing force than the behavior of the government or the Bank of Thailand has been the behavior of the household sector, in particular, the responsiveness of household savings to changes in income. Beyond this, the evidence suggests that countercyclical monetary management plays a genuine role in the short run, but that automatic stabilizers, both in fiscal and monetary behavior, are a far more sig- nificant explanation for Thailand's observed short-run stability of both fiscal and monetary outcomes. In the case of monetary policy. discretionary stabilization on the pan of the Bank of Thailand appears to occur, but only in the short run, and only with respect to income growth.' With regard to the short-run stabilization of inflation, the record of monetary policy has been surprisingly mixed. The importance of automatic stabilizers is particularly clear in the case of fis- cal outcomes. Stabilization through fiscal outcomes is more attributable to the be- havior of revenues than expenditures. On the revenue side, automatic stabilizers are a consequence of the structure of the tax system, especially the reliance on 232 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth business taxes and import duties and the elastic response of imports to changes in national income. On the expenditure side, they are a consequence of the govern- ment's fixed rules for planning expenditure and the consequent slow response of expenditure to changes in income. A strong negative relationship can be observed between capital spending and inflation, and this relationship can be attributed to the government's expenditure rules. When inflation occurs in the prices of capital goods, the budgeted funds may be insufficient to purchase the planned capital items, with the result that the expenditures may have to be deferred for at least a year. The stabilizing behavior of Thailand's fiscal aggregates can be fully explained by the properties of its automatic stabilizers. Discretionary fiscal policy, as re- vealed by the behavior of planned adjustments to fiscal aggregates, did not con- tribute to fiscal stabilization; but the outcome of discretionary fiscal behavior was not significantly destabilizing either. This could well be the most significant fea- ture that distinguishes the Thai adjustment experience from the unstable outcomes experienced in so many other developing countries. It is important to note that although automatic stabilization through adjust- ments to public investment clearly contributed to macroeconomic stability in Thailand, this form of stabilization can be harmful for long-run growth. When a sustained boom in private investment and growth occurs, a contraction in public investment can produce infrastructure bottlenecks that can choke off the sustain- ability of the growth, as was the case in Thailand in the late 1 980s and early 1 990s. The sustainability of the export and investment-led boom of that period was seri- ously threatened by the increasingly congested public roads, ports, and telecom- munications. Public investment in these infrastructure areas was contracting at the very time that the demand for them was greatest. The structural features of the fiscal and monetary systems that contributed to automatic stabilization can always be described as the outcome of policy. But these are long-term policy outcomes, not to be confused with discretionary adjust- ments to policy in the short run. Indeed, the economic literature on Thailand's ad- justment miracle has placed undue emphasis on the importance of short-run policy adjustment and has thereby overlooked those long-term structural features of the policy environment that have facilitated smooth adjustment. With the partial ex- ception of monetary policy, short-term adjustments to discretionary economic pol- icy have been unimportant sources of stabilization. Thailand's outstanding record of macroeconomic stability can be understood properly only when the long-term sources of policy stability are emphasized. Why the Macroeconomic Conservatism? Aversion to inflation and a commitment to financial prudence has a long history in Thailand. The Thai kings of the nineteenth century believed financial stability was Conclusion s Thailand's Miracle? 233 essential to keeping foreign gunboats-especially those of the French and Brit- ish-from forcing colonial rule onto their people. Until 1950, foreign financial ad- visers-always British, and usually from the Bank of England-were employed to assist the kings in this respect. The policy recommendations of the financial ad- visers were highly conservative and were influential. Thailand's only experience of sustained high inflation occurred during and immediately after World War II. This experience coincided with the foundation of the Bank of Thailand, which was assigned the task of controlling inflation and was granted considerable independence from political interference. Interestingly, this independence has subsequently continued to be respected. in spite of frequent tur- moil in the political sphere. With minor exceptions. the bank has been allowed to pursue its inflation-control mandate single-mindedly. At the first sign of high inflation, the Thai monetary authorities endeavored to contract monetary policy and sustained this monetary stance until inflation fell be- low 6 percent. The monetary contractions of 1973-74, 1979-80, and 1980-82 are cases in point. These contractions were not especially prolonged because inflation responded quickly. It must be emphasized that the past record of Thailand's con- servative monetary management has been such that its monetary policy remains highly credible. When the Bank of Thailand starts raising its lending rate, infla- tionary expectations start to abate. Exchange rate policy has largely been driven by attitudes to inflation. Deval- uations have been avoided because they are seen as capitulations to inflation. They have occurred only when the following conditions have all been met: the effective exchange rate has suddenly appreciated as a result of currency realignments among Thailand's trading partners: the price ratio of tradables versus nontradables has been falling; the level of reserves, measured in months of import coverage, has been declining for several quarters: the current level of reserves is inadequate to defend the existing exchange rate; the current rate of domestic inflation is low: and other measures to correct the trade imbalance have already been tried (including tariff increases, credit controls, and domestic monetary contraction). Why has the Bank of Thailand been so successful in pursing its sound money policies? Individuals are part of the story. Two early governors of the Bank- Prince Wiwattanachai Chaiyan (Prince Wiwat) and Puey Ungphakorn-each played a crucial role in developing the bank's traditions and professionalism. But why were successive governments willing to grant them the political indepen- dence they needed to accomplish their anti-inflation mandate? Thailand's policy formation process has been dominated by its civil servants. This is especially true of the Bank of Thailand. The rapid inflation of the 1940s had devastating effects on the real incomes of Thailand's civil servants, causing genuine hardship and increased corruption. These civil servants were the very group who were subsequently to administer Thailand's financial policies and dominate the formation of policy as well. But the economic conservatism of the civil servants is only the first component of what must be explained. The greater 234 Thailand s Macroeconomic Miracle: Stable Adjustment ancd Sustained Growth puzzle is why successive governments consistently empowered the civil service to set the constraints on economic policy and not merely to implement them. Technocrats have no inherent political authority. If they have power, it is only because politicians grant it to them. Recent developments in political science offer a potential explanation as to why they do so. It centers on a game-theoretic analy- sis known as the politician's dilemma. Its basic assumption is that politicians are motivated primarily by the desire to remain in office. There is constant pressure on them to approve the spending of public money in ways that benefit this or that spe- cial interest group. The self-interest of the politicians, it is argued, tends to favor giving in to these pressures. But when all politicians behave in this way, the accu- mulated economic effect will be large budgetary deficits, and the problems of fi- nancing these deficits may threaten the survival of the regime as a whole. It is therefore in the interests of the politicians to find a way to tie their own hands so as to restrain them from this ultimately self-defeating form of behavior. If this analysis is valid, two manifestations of this mechanism can be identified in Thailand: the empowerment of technocrats and legislated fiscal rules. EMPOWERMENT OF TECHNOCRATS. By investing technocrats with the power to say "no" to politicians, a state can institutionalize long-term fiscal and monetary re- straint, despite the short-term incentive for politicians to act otherwise. The power of the Bank of Thailand to decline to finance government budgetary deficits by purchasing government bonds, thereby expanding the money supply and promot- ing inflation, is an example; the power of the Bureau of the Budget to resist expen- ditures demanded by politicians, or at least to delay them, is another. LEGISLATED FISCAL RULES. Laws introduced during the Sarit government of 1958-63 constrained subsequent government behavior regarding such matters as the magnitudes of planned deficits in relation to planned revenues and the magni- tudes of the government's foreign borrowing, in any one year.3 These laws were amended somewhat by subsequent governments, but the point is that they were not ignored or repealed, even though governments changed frequently, sometimes through violence, and sometimes replaced the constitution at the same time. Now and then these fixed rules have acted as binding constraints on government policy. They could have been changed radically or repealed, but it seems that once laws of this kind existed and were accepted, the political cost of disposing of them was prohibitive. Their existence and wide acceptance. especially within the bureaucra- cy, generated a political inertia that helped constrain subsequent fiscal behavior for at least three decades. The theory of the politician's dilemma is an abstract notion that may provide an explanation for the otherwise puzzling outcomes of the Thai experience. But if it is basically correct, then its manifestation in a particular setting will reflect the circumstances of the time, and particular individuals may play an important role. The theory does not provide a convincing account of the mechanism by which pol- itician-constraining institutions actually come into existence in one particular en- Conclusions: Thailand's Miracle? 235 vironment and not in another, nor could a general theory of this kind be expected to do so. What it may provide is an underlying rationale for the tendency for some pol- iticians voluntarily to introduce institutional constraints on their own behavior and then for subsequent politicians to permit the continued existence of these constraints. How Well Has Thailand Adjusted? The terms-of-trade changes represented by negative external shocks reduced Thai national income in the long term. How did Thailand adjust? Before 1985, Thailand's adjustment had been unspectacular. The reasons for this judgement are, first, that the income effects of the adverse terms-of-trade movement were financed entirely by in- creased external borrowing and by depletion of international reserves; second, that the ratio of tradables prices to nontradables, which should have risen if Thailand was to have restructured its economy in response to the price shocks, had not done so. By 1985 this price ratio had returned to its level before the first oil shock. Thailand had avoided the short-run crises that so many other oil-importing countries had suffered. Macroeconomic stability had been preserved. Despite the apparent success of short-term management, the fact remains that the long-term adjustment of the Thai economy necessitated by the change in its external envi- ronment had been deferred. The increased cost of Thailand's petroleum imports was matched by an increase in external debt and a depletion of reserves. In this respect, the record of post- 1970 economic management in Thailand is significant- ly different from the ultracautious pre- 1970 emphasis on building the country's re- serves and avoiding indebtedness. Was the change a mistake? The answer depends on how well the external savings absorbed by Thailand were used. The result could have been better, but on the whole the resources were used productively. The resulting economic growth led to a sustained improvement in economic welfare for the vast mass of the population. Our account of the Thai experience is in a sense negative. When compared with other developing countries, many severely destabilized by exogenous shocks similar to those experienced by Thailand, the Thai case resembles Sherlock Holmes's "dog that did not bark." The dog did not bark because the intruder was not really such an inherent threat. It was possible to adjust to international price shocks of the 1970s and 1980s without drama. The Thai experience proves that. It was also pos- sible to allow these events to cause macroeconomic havoc, resulting in widespread human suffering. The experience of Thailand's near-neighbor, the Philippines, which has a similar economic structure and was hit by almost identical shocks, proves that. In one important respect, Thailand was fortunate. Efforts at policy-led ad- justment were implemented in the early 1980s, but by the middle of the decade these efforts had produced little apparent effect. If the externally-driven export and investment boom of the late 1980s had not occurred, Thailand would have 236 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth faced a difficult adjustment from the mid- 1980s onward. The events that led to Thailand's boom in the late 1980s produced simultaneous booms in Indonesia, Malaysia, and the southeastern corner of China. This fact shows the importance of external forces in producing the boom. But there were no such booms in the Philippines or Sri Lanka. Investors-domestic and foreign-were not willing to risk their capital in such unstable and unpromising environments. Thus internal factors also played a role. It would be wrong, however, to attribute Thailand's macroeconomic success simply to fortuitous circumstances. The same good fortune that assisted Thailand in the late 1980s was available to others as well. There is more to be extracted from the Thai experience than the importance of good luck. The principal lesson here is that giving priority to long-term macroeconomic stability promotes economic growth. It does so because macroeconomic stability is a necessary condition for attracting the domestic and foreign investment needed to exploit economic opportunity. When governments postpone action on inflation control, the cost of controlling it will even- tually be higher, not least because high inflation means that the intellectual resources of the government must then constantly be devoted to thinking of ways to contain it. The Thai experience also illustrates the value of relying principally on market solu- tions to resolve problems of resource allocation, supplemented by cautious public investment in productive infrastructure and human capital. A miracle is something that is unusual, beneficial, and inexplicable. Thailand's experience of stable growth has been miraculous in the first two senses, but not the third. There was nothing particularly brilliant, farsighted, or mysterious about the way Thailand adjusted to its changing environment: it was nothing that other de- veloping nations could not emulate or even improve upon if they chose to do so. Notes Chapter Two 1. An important example has been the traditional role of senior military figures in controlling Thai- land's public enterprises. Critics have attributed much of the corruption and inefficiency associ- ated with these organizations to military domination of their controlling Boards. 2. The following account is based primarily on Bank of Thailand (1962) and Ingram (1971). 3. Prince Wiwattanachai Chaiyan, popularly known as Pnnce Wiwat, was born in 1899, the son of Prince Mahisra Raj Haruetai. Prince Wiwat was a graduate of Cambridge University and Ecole des Sciences Politique in Paris. In 1938 he was appointed the first adviser to the minister of fi- nance and in 1940 he became the director of the National Banking Bureau. 4. This recommendation was in his August 1944 report on postwar monetary reconstruction, as con- tained in Wiwatchaiyanusorn (Wiwat memorial) (see Bank of Thailand 1961). 5. This appeared in his lecture notes published in Thanakhan Haeng Prathet Thai (Bank of Thai- land), a memorial volume issued at the cremation of Gen. Phao Bariphanyutthakit (Bank of Thai- land 1970). Chapter Four I. The only exception to this pattem was the change in the structure of effective protection between 1984 and 1987, which showed a weak positive relationship to changes in export performance. 2. Note that the revenues of the public enterprises include the total proceeds from their sale of goods and services, which somewhat distorts the comparison with the rest of the public sector. Chapter Five I. In this autoregressive model, the usual DW statistic has reduced power and is biased toward 2. The relevant test for autocorrelation is Durbin's h-statistic. Under the null hypothesis of no auto- correlation, h is asymptotically normal with 0 mean and unit variance. Our estimate of Durbin's h-statistic is 0.74, which is less than the critical value of 2.756 (at the 99 percent significance lev- el) and indicates the absence of significant autocorrelation 237 238 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth Chapter Six I. This taxonomy has benefited greatly from discussions with W. M. Corden. Chapter Seven I . Thailand's fiscal year begins on October I and ends on September 30 of the following calendar year. We shall subsequently write, for example, fiscal 1990 to mean the fiscal year 1989-90, end- ing on September 30 1990. 2. By "inflation" we mean, as elsewhere in this book, the rate of change of the consumer price index. The mechanism by which inflation retards expenditure growth is discussed in a subsequent sec- tion of this chapter. 3. This interpretation differs from that of Robinson and others (1991: 16). This important study mentions automatic stabilizers in passing but emphasizes what is assumed, without supporting evidence, to be the stabilizing role of discretionary fiscal policy. 4. This interpretation is further supported by the evidence on the behavior of the current account, discussed above. 5. An estimate of zero would have meant that real expenditures remained constant as the price level rose and thus that nominal expenditures rose with the price level; an estimate of -1.0 would in- dicate that nominal expenditures remain exactly constant as the price level increases. 6. On the basis of data for 1970-89, the correlation coefficient between the private investment/ODP ratio and the public investment/GDP ratio was 0.66. 7. If (E - R) = 0.2E, then E = I .25R and (E - R)/R = 0.25. Chapter Eight ]. Only ten countnres reported lower average rates of inflation than Thailand over the twenty-five years from 1965 to 1990: Ethiopia, Chad, Burundi, Honduras, Panama, Hungary, Singapore, Ger- many, Canada, and Switzerland (World Bank 1992: table 1). 2. Relaxed in 1990 to fifteen days, following the acceptance by Thailand of iMF article XIII. 3. The negative effect of the premium forware exchange rate on short-term capital flows will be con- firmed by regression (8.1 3). 4. The latter variables were computed from the domestic consumer price index and the world import price index from IMF, International Financial Statistics, various issues. 5. It was only in 1986 that the annual current account was in surplus. 6. When market capitalization grew, portfolio investment also grew as a proportion of the monetary base and became more volatile. 7. Following the discussion in chapter 7, it is essential to distinguish between fiscal outcomes, as referred to here, and fiscal intentions. As already demonstrated, they are not necessarily the same. 8. See Naris (1993) for a discussion of this issue as it relates to the collusive determination of inter- est rates within the oligopolistic structure of Thailand's commercial banking sector, combined with the major banks' difficulty in agreeing on interest rate reductions. Notes 239 Chapter Nine 1. The data for the regression were obtained from the Intemational Monetary Fund, International Financial Statistics, various issues. The series used is denoted 'aa" and is the ratio of the end-of- period national currency value of the special drawing right (SDR). Chapter Ten I. To a lesser extent, the Ministry of Finance also plays a role in monetary stabilization, through adjustments to the withholding tax on foreign borrowing. 2. The name derives from the famous game-theoretic analysis known as the prisoner's dilemma. See Williamson ( 1994). 3. Examples include the 1959 Budgetary Law, which limited any excess of planned spending over planned revenue to no more than 25 percent of the planned revenue: and the 1960 External Public Debt Law, which set a 5 percent ceiling on the govemment's debt service ratio and limited the size of foreign debt service to less than 13 percent of the level of exports. Bibliography Aghevli, Bijan B., and S. Mohsin Khan. 1978. 'Govemment Deficit and Inflation- ary Process in Developing Counties." IMF Staff Papers 25 (September): 383- 416. Aliber, Robert Z. 1973. "The Interest Rate Parity Theorem: A Reinterpretation." Journal of Political Economy 81 (November/December): 1451-59. Ammar Siamwalla. 1975. "Stability, Growth and Distribution in the Thai Econo- my." In Puey Ungphakorn and others, eds.. Finance, Trade and Economic De- velopment in Thailand: Essays in Honotur oj Khunving Suparb Yossutldra. Bangkok: Sompong Press. . 1978. "Farmers and Middlemen: Aspects of Agricultural Marketing in Thailand." Economic Bulletin for Asia and Pat ific 29(1 ):38-50. Ammar Siamwalla, and Prayong Nettayarak. 1990. "Estimating the Cost of Sub- sidies for Agricultural Credit." Thailand Development Research Institute, Bangkok. Processed. Ammar Siamwalla, and Suthad Setboonsarng. 1989. Trade, Exchange Rate and Agricultural Pricing Policies in Thailand: Comparative Studies on the Agricultural Pricing Policy. World Bank, Washington, D.C. Ammar Siamwalla, Suthad Sethboonsarng, and Direk Patamasiriwat. 1993. "Ag- riculture." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Amnuey Saengnoree. 1984. Comparative Analyxis of Market Structure, Econo- mies of Scale and Profit Behavior of Thai Commercial Banks and Finance and Securith Companies. Bangkok: King Mongkut Institute of Technology. (In Thai.) Anat Arbhabhirama, Dhira Phantumvanit, J. Elkington, and Phaitoon Ingkasu- wan, eds. 1988. Thailand Natural Resources Profile. Bangkok: Thailand De- velopment Research Institute. Argy, Victor. 1971. "Monetary Policy and Internal and External Balance." IMF Staff Papers 18 (November): 508-27. Armatonoff, G. L. 1965. State Owned Enterprise s of Thailand. Bangkok: Agency for International Development. Asjana Watananukij. 1985. "Foreign Debt of Public Enterprises." Paper presented at the seminar Foreign Debt Crisis of the Thai Government, February 11-12, Faculty of Economics, Thammasat University. Bangkok. (In Thai.) 240 Bibliography 241 Bandid Nijathaworn, and Madee Weerakitpanich. 1987. "Economic Fluctuations and Stability of the Commercial Banking System." Symposium paper on the stability of financial institutions, Thammasat University, Bangkok. (In Thai.) Bank of Thailand. 1961. Viwatchaivanusorn (Wioi at Memorial). Bangkok: Sivaporn. (In Thai.) 1962. Twentieth Anniversary of the Establishment of the Bank of Thlailand, 10 December 1962. Bangkok. (In Thai). . 1970. Thanakhan haeng Prathet Thai (Bantk of Thailand). Memorial volume issued at the cremation of Gen. Phao Bariphanyutthakit. Bangkok: Sivaporn. (In Thai.) . 1986. Financial Institutions in Thailand. Bangkok. . 1992. 50 Years of the Bank of Thailand, 1942-1992. Bangkok. Various issues. Month1v Bulletin. Bangkok. . Various issues. Monthlv Economic Report. Bangkok. (In Thai.) Various issues. Quarterlv Bulletin. Bangkok. Barker, Randolph, Robert W. Herdt, and Beth Rose. 1985. The Rice Economy of Asia. Washington D.C.: Resources for the Future. Beltz, P. A., and B. P. McCormack. 1982. "The Origins of the Banking Industry in Thailand." Thai-American Business 14(6): 7-11. Bertrand. Trent J., and Lyn Squire. 1980. "The Relevance of the Dual Economy Model: A Case Study of Thailand." Oxford Economic Papers 32(3): 480-511. Bhagwati, Jagdish. 1984a. "Splintering and Disembodiment of Services and Developing Nations." The World Economy 7(2): 133-44. 1984b. "Why Are Services Cheaper in the Poor Countries?" Economic Journal 94 (June): 279-84. . 1989. Protectionism. Cambridge, Mass: MIT Press, Bhanupong Nidhiprabha. 1993. "Monetary Policy." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Bhanupong Nidhiprabha and Atchana Wattananukit. 1987. "Asset Holding Be- havior and the Stability of Thai Commercial Banks." Paper presented at the symposium The Stability of Financial Institutions, Thammasat University, Bangkok. (In Thai.) Blinder, A. S., and R. M. Solow. 1974. "Analytical Foundations of Fiscal Policy." Economtzics of Public Finance. Washington, D.C.: Brookings Institution. Block, Peter, Sirilaksana Chutikul, and Nipon Poapongsakorn. 1986. "Public Sector Employment in Thailand: Civil Service and State Enterprises." Background Paper for World Bank Economic Report on Thailand. Washington, D.C. Buiter, W. H. 1977. "Crowding Out and the Effectiveness of Fiscal Policy." Journal of Public Economics 7(3): 309-28. 1. 983. "Measurement of Public Sector Deficit and Its Implications for Policy Evaluation and Design." IMF Staff Papers 30(2): 306-49. Caldwell, J. C. 1967. "The Demographic Structure." In T. H. Silcock, ed., Thailand: Sociial and Economic Studies in Development. Canberra: Australian 242 Thailand s Macroeconomic Miracle: Stable Adjuistiment and Sujstained Growth National University Press. Central Statistical Office. 1935. Statistical Yearbook of Siam 1933-35. Bangkok. Chaipat Sahasakul. 1993. "Fiscal Policy." In Peter G. Warr, ed., The Thai Econo- my in Transition. Cambridge: Cambridge University Press. .1987. Features of the Tax System in Thailand. Bangkok: Thailand Devel- opment Research Institute. Chaipat Sahasakul, Thongpakda Nattapong, and Kraisoraphong Keokam. 1989. Lessons from the World Bank's Experience of Structural Adjustment Loans (SALs). A Case Study of Thailand. Bangkok: Thailand Development Research Institute. Chaiyawat Wibulswasdi. 1984. "Formulation and Implementation of Monetary Policy: A Study of Thailand Monetary Experience during 1983-1984." Paper for the 15th SEANZA Central Banking Course, Kathmandu, Nepal. Chaiyong Pativintranond. 1984. "Lending Behavior of Commercial Banks in Thailand." Master's thesis, Thammasat University, Bangkok. (In Thai.) Chalongphob Sussangkarn. 1993. "Labour Markets." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Chand, S. K. 1973. "Summary Measures of Fiscal Influence." IMF Staff Papers 20(3): 405-47. Chelliah, R. J. 1973. "Significance of Alternative Concepts of Budget Deficit." IMF Staff Papers 20(3): 741-77. Chesada Loohawenchit. n.d. "State Enterprises and Their Impact on the Govern- ment Budget." Thammasat University, Bangkok. (In Thai.) Processed. . 1984. "Public Enterprises in the Thai Economy." Paper presented at the seminar Thai Public Enterprises: Past, Present, and Future, January 26-27, Bangkok. (In Thai.) Chirmsak Pinthong. 1977. "A Price Analysis of the Thai Rice Marketing System." Ph.D. diss., Stanford University. . 1984. "Distribution of Benefit of Government Rice Procurement Policy in 1982/1983." Thatnmasat University Journtal 13(2): 166-87. (In Thai.). Corden, W. M. 1967. "The Exchange Rate System and the Taxation of Trade." In T. H. Silcock, ed., Thailand: Social and Economic Studies in Development. Canberra, Australian National University Press. . 1987. "Relevance for Developing Countries of Recent Developments in Macroeconomic Theory." World Bank Research Observer 2(2): 171-88. Corden, W. M., and H. V. Richter. 1967. "Trade and the Balance of Payments." In T. H. Silcock, ed., Thailand: Social and Economic Studies in Development. Canberra: Australian National University Press. Cumby, Robert E., and Maurice Obstfeld. 1984. "International Interest Rate and Price Level Linkages Under Flexible Exchange Rates: A Review of Recent Ev- idence." In John F. Bilson and Richard C. Marston, eds., Erchange Rate The- ory and Practice. Chicago: University of Chicago Press. Direk Patamasiriwat, and Suewattana Sakeddao. 1990. "Sources of Growth of Ag- ricultural Production 1965-1985: Analysis Based on TDRI Model." Warasarn Settasat Thammasat 8(1): 43-69. (In Thai.) Bibliographv 243 Direk Patamasiriwat, and Supaporn Prompongsri. 1987. "Financial Centralization and the Setting of Economic Policies." Paper presented at the symposium The Stability of Financial Institutions, Thammasat University. Bangkok. (In Thai.) Direk Patmasiriwat, and Pairoj Benjamanon. 1982. "Concentration and Mobility of Commercial Banking System in Thailand." Journal of National Institute of DevelopmentAdministration 22(2): 278-304. (In Thai.) Dooley, M. P., and Peter Isard. 1980. "Capital Controls, Political Risk, and Devia- tions From Interest Rate Parity." Journal of Political Economv 88: 370-84. Dornbusch, Rudiger, and Stanley Fischer. 1992. Macroeconomics. New York: McGraw-Hill. Driscoll, Michael J., and Asok K. Lahiri. 1983. "Income Velocity of Money in Agricultural Developing Economies." Review of Economics and Statistics 65(3): 393-401. Emery, R. F. 1970. The Financial Institutions of Sozutheast Asia: A Country-by- Country Study. New York: Praeger. Enrique, R. S., and others. 1985. "Thailand: Poverty Review." World Bank, East Asian and Pacific Division, Washington. D.C. Processed. Feder, Gerschon, Tongroj Onchan, Yongyuth Chalamwong, and Chira Hongladarom. 1988. Land Policies and Farm Productivity in Thailand. Baltimore. Md.: Johns Hopkins University Press. Feldstein. Martin. 1983. "Domestic Savings and International Capital Movements in the Long Run and the Short Run." European Economic Review 21(March): 129-51. Feldstein, Martin, and Charles Horioka. 1980. "Domestic Saving and lnternational Capital Flows." Economic Journal 90: 314-29. Frankel, Jacob A., and Michael L. Mussa. 1981. "Monetary and Fiscal Policies in an Open Economy." American Economic Review 71(May): 253-58. Friedman, Milton. 1968. "The Role of Monetary Policy." American Economic Review 58(1): 1-17. Fuller, T. D., Prathet Kamnuansilpa, P. Lightfoot, and Samakhom Rathanamongkolmas. 1983. Migration and Development in Thailand. Bangkok: Social Science Association of Thailand. Goldfeld, Stephen M., and Lester V. Chandler. 1986. The Economics of Money and Banking. New York: Harper and Row. Herring R. H., and R. M. Marston. 1977. National Monetary Policies and International Financial Markets. Amsterdam: North-Holland. Hewison. Kevin. 1989. Bankers and Bureaucrats: Capital and the Role of the State in Thailand. Yale Center for International and Area Studies, Monograph Series 34. New Haven, Conn.: Yale University. Industrial Management Corporation. 1984. A Study of Fiscal Implications of Investment Incentives and Promotion Efficiencv. Bangkok. -. .1985. Industrial Restructuring Study for tlhe NESDB. Bangkok. Ingram, James C. 1971. Economic Clhange in Thiailand: 1850-1970. Stanford, Calif.: Stanford University Press. 244 Thailand's Macroeconomic Miracle: Stable Adjustment aind Sustained Growth International Labour Office, Asian Regional Team for Employment Promotion. 1984. Employment Issues in Thailand's Sixth Plan. Bangkok. International Monetary Fund (IMF). Various years. International Financial Statistics. Washington, D.C. Inthapanya Wutdhithep, and Jumlong Atikul. 1988. "Urban Self-Employment in Bangkok." A Report for the National Economic and Social Development Board of Thailand, New Delhi, India. Jamison, D. T., and L. J. Lau. 1982. Farmer Education and Farm Efficiency. Baltimore, Md.: Johns Hopkins University Press. Jones, L. P. 1975. Public Enterprise and Economic Development: The Korean Case. Seoul: Korean Development Institute. . ed. 1982. Public Enterprise in Less Developed Countries. Cambridge: Cambridge University Press. Juanjai Ajanant, Supote Chunanuntatham, and Sorrayuth Meenaphant. 1986. Trade and Industrialization of Thailand. Bangkok: Social Science Association of Thailand. Kosit Panpiemras, and Somchai Krusuansombat. 1985. "Seasonal Migration and Employment in Thailand." In T. Panayotou, ed., Food Policy Analysis in Thailand. Bangkok: Agricultural Development Council. Kouri, Penti. 1975. "The Hypothesis of Offsetting Capital Flows: A Case Study of Germany." Journal of Monetary Economics 1: 21-39. Kouri, Penti, and Michael Porter. 1974. "International Capital Flows: A Portfolio Equilibrium." Journal of Political Economy 82(May/June): 443-67. Kraiyudht Dhiratayakinant. 1987. "Privatization: An Analysis of the Concept and Its Implementation in Thailand." Bangkok: Thailand Development Research Institute. . 1989. "Privatization of Public Enterprises: The Case of Thailand." A research report commissioned by Asian and Pacific Development Centre, Kua- la Lumpur. . 1993. "Public Enterprises." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Krueger, Anne O., Maurice Schiff, and Alberto Valdes. 1988. "Agricultural Incentives in Developing Countries: Measuring the Effect of Sectoral and Economywide Policies." World Bank Economic Review 2(3): 255-71. Lee, S. Y., and Y. T. Jao. 1982. Financial Structures and Monetary Policies in Southeast Asia. Hong Kong: Macmillan. Leff, Nathaniel, and Kazuo Sato. 1980. "Macroeconornic Adjustment in Develop- ing Countries: Instability, Short-Run Growth and External Dependency." Re- view of Economics and Statistics 62(2): 170-79. Llewellyn, David T. 1980. International Financial Integration. London: Macmillan. Mabry, Bevars D. 1984. "The Development of Labor Institutions in Thailand." Southeast Asia Program, Drake Paper 112. Cornell University, Ithaca, N.Y. Martin, Will, and Peter G. Warr. 1990. "Explaining Agriculture's Declining Share Bibliography 245 of Thai National Income." Chulalongkorn Journal of Economics 2 (August): 178-224. (In Thai.) - . 1994. "The Declining Economic Importance of Agriculture: A Supply Side Analysis for Thailand." Agricultural Economics 11 (November): 219-35. Mathana Phananiramai, and Andrew Mason. 1988, "Enrollment and Educational Cost in Thailand." Paper for the seminar Demographic and Economic Forecast for Thailand, July 2-3, Bangkok. Mathana Phananiramai, and Aphichart Chamrasrithirong. 1984. "Poverty in the Northeast: A Study of Low-income Households in the Northeastern Region of Thailand." A Research Report Submitted to the Council for Asian Manpower Studies, Manila. Medhi Krongkaew. 1993. "Poverty and Income Distribution." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Medhi Krongkaew, and Chintana Chernsiri. 1975. "The Determination of the Poverty Level in Thailand." Thammasat University Journal 5(1): 48-68. (In Thai.) MedhiKrongkaew, and others. 1988. Financing Public Sector Development Ex- penditure in Selected Countries: Thailand. Asian Development Bank, Manila. Medhi Krongkaew, Pranee Tinakorn, and Suphat Suphachalasai. 1991. "Priority Issue and Policy Measure to Alleviate Rural Poverty: The Case of Thailand." Economic Development Resources Center, Asian Development Bank, Manila. Meltzer, Allan H. 1987. "Limits of Short-Run Stabilization Policy." Economic In- quiry 42(January): 1-14. Mishkin, F. S. 1976. "Illiquidity, Consumer Durable Expenditure, and Monetary Policy." American Economic Review 66(4): 642-54. -- .1984. "Are Real Interest Rates Equal Across Countries? An Investigation of International Parity Conditions." Journal of Finance 39: 1345-58. Modigliani, Franco. 1977. "The Monetarist Controversy or, Should We Forsake Stabilization Policies?" American Economic Review 67(2): 1-19. Moerman, Michael. 1968. Agricultural Change and Peasant Choice in a Thai Village. Berkeley, Calif.: University of California Press. Monadjemi, M. 1990. "Testing the Degree of International Capital Mobility." Australian Economic Papers (June) 30-39. Muangchai Tajaroensuk. 1975. "Educational Planning in Thailand: Status and Organization." Bulletin of the UNESCO Regional Office for Education in Asia 16: 128-40. Mundell, Robert A. 1962. "The Appropriate Use of Monetary and Fiscal Policy for Internal and External Stability." IMF Staff Papers 9: 70-79. Muscat, Robert. 1966. Development Strategy in Thailand. New York: Praeger. Myers, C. N. 1986. "Quality of Human Resources in Human Resources Management." Paper prepared for TDRI Year-End Conference, December, Bangkok. Myers, C. N., and Chalongphob Sussangkarn. 1989. "Economic Transformation and Flexibility of the Education System." Paper prepared for Human Resource 246 Thailand's Macroeconomic Miracle: Stable Adjustmetmt and Sustained Growth Problems and Policies, February 24-25, Hua-Hin. Nalinee Homasawin. 1984. "A Regional Analysis of Financial Structure in Thailand: Growth and Distribution." Master's thesis, Thammasat University. Naris Chaiyasoot. 1993. "Commercial Banking." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Naris Chaiyasoot, Varakorn Samakoses, and Cheunruthai Pornpatrakul. 1987. "Structure and Problems of Financial Institutions in the Thai Economy." Paper presented at the symposium The Stability of Financial Institutions, Thammasat University, Bangkok. (In Thai.) Narongchai Akrasanee. 1973. "The Manufacturing Sector in Thailand: A Study of Growth, Import Substitution, and Effective Protection, 1960-1969." Ph.D. dissertation, Johns Hopkins University. . 1977. "The Structure of Effective Protection in Thailand: A Study of In- dustrial and Trade Policies in the Early 1970s." Report prepared for the Minis- try of Finance, the National Economic and Social Development Board of the Government of Thailand, and the International Bank for Reconstruction and Development, Washington, D.C. . 1983. Rural Off-farm Employment in Thailand: Summary and Synthesis of the Rural Off-farm Assessment Project. Bangkok: Industrial Management Consultants. Nimit Nontapanthawat. 1976. "Foreign Borrowing of Commercial Banks in Thailand." In Narongchai Akrasenee and Rangsan Thanapornpun, eds., Rak Muang Thai. Bangkok: Thailand Social Science Association. (In Thai.) Obstfeld, Maurice. 1986. "Capital Mobility in the World Economy: Theory and Measurement." Carnegie-Rochester Conference Series on Public Policy. Journal of Monetary Economics Suppl. Amsterdam: North-Holland. Oey Meesook. 1979. Income, Consumption and Poverty in Thailand, 1962/63 to 1975/76. World Bank Staff Working Paper 364. Washington D.C. Otani, Ichiro, and Siddharth Tiwari. 1981. "Capital Controls and Interest Rate Parity: The Japanese Experience, 1978-9 1" IME Staff Papers 28 (December): 793-816. Pairote Wongwuttiwat. 1975. "The Structure of Differential Incentives in the Manufacturing Sector: A Study of Thailand's Experience during 1945-1974." Master's thesis, Thammasat University. Paisan Chaimonkol. 1976. Public Enterprises. Bangkok: Thai Watana Panich. (In Thai.) Paitoon Wiboonchutikula, Rachain Chintayarangsan, and Nattapong Thongpakde. 1989. "Trade in Manufactured Goods and Mineral Products." The 1989 TDRI Year-End Conference. Thailand Development Research Institute, Bangkok. Pannee Bualek. 1986. An Analysis of Thai Commercial Bank Capitalists: B.E. 2475-2516. Bangkok: Chulalongkorm University. (In Thai.) Pasuk Phongpaichit. 1980. Economic and Social Transformation of Thailand 1957-1976. Bangkok: Chulalongkorn University Social Science Research Institute. Bibliography 247 1982. From Peasant Girls to Bangkok Masseuses. Geneva: International Labour Office. Pasuk Pongpaichit, and Samart Chiasakul. 1993. "Services." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Poonsin Ingavala. 1989. "Privatization in Thailand: Slow Progress Amidst Much Opposition." Asean Economic Bulletin. March. Praipol Koomsup. 1993. "Energy Policy." In Peter G. Warr, ed., The Thai Econommy in Transition. Cambridge: Cambridge University Press. Prateep Sondysuvan, ed. 1975. Finance, Trade and Economic Development in Thailand. Bangkok: Sompong Press. Prayuth Thongtheppairot. 1972. "Return on Investment in Thai Bank." Business in Thailand 3(8):19-20. Reynolds, Clark. 1978. "Why Mexico's Stabilizing Development Was Actually Destabilizing (with Some Implications for the Future)." World Development 6(7/8): 1005-18. Riggs, F. W. 1966. Thailand: The Modernization of a Bureaucratic Policy. Honolulu: East-West Center Press. Rizwanul, Iwan. 1984. "Poverty, Income Distribution and Growth in Rural Thailand." In A. R. Khan and E. Lee, eds., Poverty in Rural Asia. Bangkok: International Labour Office, Asian Regional Team for Employment Promotion. Robinson, David, Yangho Byeon, and Ranjit Teja. 1991. "Thailand: Adjusting to Success, Current Policy Issues." Occasional Paper 85. International Monetary Fund, Washington, D.C. Rozental, A. A. 1970. Finance and Development in Thailand. New York: Praeger. Rungsan Thanapornan. 1985. Tax Buoyancy and Tax Flasticity in Thailand: The State of Knowledge. Bangkok: Thammasat University, Faculty of Economics. (In Thai.) Sataporn Jinachitra. 1987. "Efficiency, Stability, Supervision of the Financial Institutions." In Nopporn Ruengsakul and others, eds., Money and Banking and Implementation of National Economic Policies. Bangkok: Chulalongkorn University Press. (In Thai.) Schlossstein, Steven. 1991. Asia's New Little Dragons: The Dynamic Emergence of Indonesia, Thailand, and Malaysia. Chicago, Ill.: Contemporary Books. Silcock, T. H. 1967. "Promotion of Industry and the Planning Process." In T. H. Silcock, ed., Thailand: Social and Economic Studies in Development. Canberra: Australian National University Press. . ed. 1967. Thailand: Social and Economic Studies in Development. Canberra: Australian National University Press. . 1970. The Economic Development of Thai Agriculture. Canberra: Australian National University Press. Sirilaksana Chutikul-Khoman. 1988. "The Burden of Expenditure on Higher Education in Thailand and the Role of the Private Sector." In Nipon Poapongsakorn and Rangson Thanapornpun, eds., The Economy: On the Path of Peace and Social Justice. Bangkok: Thammasat University Press. (In Thai.) 248 Thailand s Macroeconomic Miracle: Stable Adjustment and Sustained Growth . 1993. "Education Policy." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. Skully, M. T. 1984. Financial Institutions and Markets in Southeast Asia. London: Macmillan. Somchai Jitsuchon. 1987. "Sources and Trend of Income Inequality: Thailand 1975/76 and 1981." Master's thesis, Thammasat University. Sompop Manarungsan. 1989. Economic Development of Thailand, 1850-1950: Response to the Challenge of the World Economy. Institute of Asian Studies Monograph 042. Chulalongkorn University. Somsak Tambunlertchai, 1977. Japanese and American Investment in Thailand's Manufacturing Industries. Tokyo. Institute of Developing Economies. . 1993. "Manufacturing." In Peter G. Warr, ed., The Thai Economy in Transition, Cambridge: Cambridge University Press. Somsak Tambunlertchai, and Chesada Lohawenchit. 1981. "Labour-Intensive and Small-Scale Manufacturing Industries in Thailand." In Ahmed Rashid, ed., The Development of Labour-Intensive Industry in ASEAN Countries. Geneva: International Labour Organisation. Somsak Tambunlertchai, and I. Yamazawa. 1983. Manufactured Export Promo- tion: The Case of Thailand. Institute of Developing Economies Joint Research Program Series 38, Tokyo. Suehiro, Akira. 1985. Capital Accumulation and Industrial Development in Thailand. Bangkok: Chulalongkorn University Social Research Institute. . 1989. Capital Accumulation in Thailand: 1855-1985. Tokyo: Centre for East Asian Cultural Studies. Suganya Hutaserani, and Somchai Jitsuchon. 1988. "Thailand's Income Distribution and Poverty Profile and Their Current Situations." Paper prepared for the 1988 TDRI Year-End Conference, Thailand. Supote Chunanunthatham. 1978. "The Binding of the Exchange Rate Between Thai Baht and U.S. Dollar." In Current Economic Situation and Policy Analy- sis of Thailand, Economics Symposium. Bangkok: Faculty of Economics, Thammasat University. (In Thai.) Swoboda, A. K. 1973. "Monetary Policy under Fixed Exchange Rates: Effectiveness, the Speed of Adjustment, and Proper Use," Economica 40: 136- 45. Tanzi, Vito. 1986. "Fiscal Policy Responses to Exogenous Shocks in Developing Countries." American Economic Review 76(2): 88-91. .1989. "Fiscal Policy, Growth, and the Design of Stabilization Programs." In Mario I. Blejer and Ke-young Chu, eds., Fiscal Policy, Stabilization, and Growth in Developing Countries. Washington, D.C.: International Monetary Fund. Tarisa Derethinan. 1987. "The Development of Thai Financial Institutions." In Nopporn Ruengsakul and others, eds., Money and Banking and Implementation of National Economic Policies. Bangkok: Chulalongkorn University Press, (In Thai.) Bibliographv 249 Terwiel, B. J. 1983. A History of Modern Thailand: 1767-1942. St. Lucia: University of Queensland Press. Thailand, Ministry of Finance. 1990. A History of (;o'ernment Salary in Thai- land: 100th Anniversary of the Department oJ the Compntroller-General. Bangkok: Borpit Printing. Thailand, National Economic and Social Development Board. 1983. "Summary of the Financial Situation and Operation of 60 Public Enterprises during the First Two Years of the Fifth Development Plan." B.E. 2525 and 2526. Bangkok. Processed. Thailand, National Statistical Office. 1988. Report ol'the 1988 Household Socio- Economic Survey: Whole Kingdom. Bangkok. Thompson, Virginia. 1941. Thailand: The News Siamn. New York: Macmillan. Tobin, James. 1978. "Monetary Policy and the Economy: The Transmission Mechanism." Southern Economic Journal 44(3-4): 4211-31. Trairong Suwankiri. 1970. "The Structure of Protection and Import Substitution in Thailand." Master's thesis, University of the Philippines. Trairong Suwankiri, and others. 1975. "The Determination of Minimum Wages in Thailand and Their Economic Impact." Thamninasat Univ ersitN Journal 5(l): 2-47. (In Thai.) Trescott, P. B. 1971. Thailand's Monetary Experience: The Economics of Stabilitv. New York: Praeger. United Nations Development Programme. 1992. Ihioniani Development Report. New York: Oxford University Press. Usher, Dan. 1967. "The Thai Rice Trade." In T. H. Silcock, ed., Thailand: Social and Economic Studies in Developmerit. Canberra: Australian National University Press. Vares Oupatiga. 1986. Roles of Financial Institutions in Thailand in Savings Mobilization and Credit Allocation. Bangkok: Thammasat University, Faculty of Economics. (In Thai.) Virabongsa Ramangkura. 1975. "A Macroeconomic Model for Thailand: A Classical Approach." In Prateep Sondysuvan, ed.. Finance, Trade and Economic Development in Thailand. Bangkok: Sompong Press. Warr, Peter G. 1993. "The Thai Economy." In Peter G. Warr, ed., The Thai Economy in Transition. Cambridge: Cambridge University Press. . ed. 1993. The Thai Economin in Transition. Cambridge: Cambridge University Press. Warr, Peter G., and Bandid Nijathaworn. 1987. "Thailand's Economic Performance: Some Thai Perspectives." Asia;i-Pacific Economic Literature 1:60-74. Reprinted (in Chinese) in Jingiuixe Yicong (Economic translations) 10(1988): 59-66. Williamson, John. 1994. "in Search of a Manual for Technopols." In John Williamson, ed., The Political Economny of Policy Reform. Washington. D.C.: Institute for International Economics. Wilson, Constance M. 1983. Thailanid: A Handbilook of Historical Statistics. 250 Thailand's Macroeconomic Miracle: Stable Adjustment and Sustained Growth Boston: G. K. Hall. Wilson, D. A. 1960. Politics in Thailand. New York: Cornell University Press. World Bank. 1959. A Public Development Program for Thailand. Baltimore, Md.: Johns Hopkins University Press. . 1992. World Development Report 1992. New York: Oxford University Press. . 1993. East Asia's Economic Miracle. New York: Oxford University Press. Wyatt, David K. 1984. Thailand: A Short Historv. New Haven, Conn.: Yale University Press. Index Agricultural: credit policy, 87; employ- Bank of Agriculture and Agricultural ment, 78; exports, 9,48-49, 53; Cooperatives (BAAC), 39 GDP share, 44, 47,49; growth, 9, Bank of England, 9, 10, 232 71,72, 88; interventions, 78-79; la- Bank of Thailand. 9, 10-11, 28, 89, bor force share, 47-49; tax, 34-35 169: assistance to industries, 52; Amnesty bill, 16 bank failures,41-42; bank regula- Anand Panyarachun, 17, 18 tion, 28, 40-41; credit ceiling, Animal Slaughtering and Meat Sale 210-1 1; exchange rate control, Control Act, 79 85; export promotion, 80; finan- Annual Appropriation Act, 28 cial market. 86; foreign ex- Aristocracy: civil service, 21, 68 change, 34, 41, 168; governors, Association of SoutheastAsian Nations 11, 27; independence, 201, 203, (ASEAN): investment in Thai- 233; inflation control, 168, 233; land, 55-56 interest rates, 86, 178, 180, 184, Athit Kaamlang-ek, 212 192-93, 196-97; lending, 165, Australian model, 217 167, 199-203; lending rate, 190- Automatic stabilizers, 144, 231-32 92, 230-31, 233; macroeconomic Ayutthaya, 8 policy decisions, 69; monetary base control, 184; monetary poli- Baht: devaluation, 10-11. 85, 168, 205- cies, 53-54; NDPC, 93; Quarterly 14, 227, 229; exchange rate, 10, Bulletin, 64, 65; sterilization of 19, 44, 85-86, 205-08, 215-16, capital flows, 188 227; 1,000 baht bank note, 12 Bank of Thailand Act, 10 Balance of payments, 55, 71, 72, 73, Bank rate adjustments, 190-92 142; model, 217; shocks, 99-105, Banking regulations. 40-41 217 Banks, commercial: 39-40; control, 28, Balassa, 48 40-41, 86; credit restriction. 196; Bangkok: administration, 69; Chinese, failures, 41; foreign borrowing, 7, 31; founding, 8; industry, 72; 42, 192; foreign exchange, 170; population, 30, 38, 72 government bonds, 11- 12; gov- Bangkok Bank. 40, 210 emient loans, 165; interest Bangkok Mass Transit Authority, 91 rates, 192-93; salary deposits, 12 Bank credits: ratio of GDP, 194 Banks, foreign, 39 Bank deposits: ratio of GDP, 194 Banks, specialized, 39 251 252 Thailand\' MAaci,ve(,olonfi Miracle: Slable Adj;u ;m en I ad Sustained Grawth Bertrand, Trant J., 38 Communists, 13, 14 Bhumipol Adulyadej, King (Rama IX), Comptroller-General Department. 65 7, 14, 18 Constitution: abrogation, 13-14; new, Board of Investment, 13, 79-81 16 Book Club Association, I0 Cook, Sir Edward, 19 Bowring Treaty, 9, 60 Coriuption: Chatichai government, 17; British, 8-9: civil service, 18-19; finan- civil service, 22, 233 cial adviser, 9. 19, 232 Council of Economic Ministers, 69, Buddhism. 7. 31 210 Budget, 68; deficits, 145-54 Crown P'roperty Bureau, 39 Budget Bureau, 28, 69, 92-93, 234 Crude birth rate, 31 Budgetary Law, 145, 161, 239 Crude death rate, 31 Business people: politics, 15 Currency Act, 19, 28-29 Currency Emergency Act, 20 Capital account, 41, 160, 186 Currency reserves, 19-20, 28; ratios, II Capital expenditure, 71, 157 Current account, 186, 196, 199, 201, Capital movement, 178, 186-88; inter- 203, 207, 210 est rate and, 188-89; regulations, Customs Department, 80, 81 170-71 Cash: circulationi, 12 Detense expenditure, 157 Cassava: quotas, 79 Deficit linancing, 163 Chakri dynasty, 8 Democrat party, 14, 15 Chalongphob Sussangkarn. 38 Deemociatic reform: demand, 7 Chamlong Srimuang, 17 Demonstrations: 14, 15, 17-18 Chao Phya River, 8 Development plans, 13, 52, 70-75, 87- Chatichai Choonhavan, 17 89 Checks. 12 DIomestic borrowing, 165 China: growth hoom, 45, 236; invest- DLutch dlisease phenomenon, 220 ment in Thailand, 56; and Thai- land, 14. 27 Eastern Seaboard Development Chinese in Thailand, 7, 31; bank owner- Scheme, 73, 88 ship, 39: education in China, 27 Economic conservatism, 19-29 Chulalongkorn, King (Rama V), 9, 10, Economic growth: effect of shocks, 15, 21. 68 1 12-16; nineteenth century, 9 Civil service: aristocracy, 21, 68; cor- Economic policies, 2, 7-8, 15; World ruption, 22, 233: independence, Bank, 13 18-19,68; loyalty to king, 19,68; Education, 31 salaries, 21, 22-26, 157, 233; Elections, 13-19 trade union ban, 38 Electricity Generating Authority, 84, 134 Colonization: avoidance, 8-9, 231-32 Employment: agriculture, 78; public Commerce Ministry, 80, 84-85 enterprises, 90-91 Commercial Banking Act. 28, 39 Eurodollar interest rate, 174, 180, 192 Commnodity prices: fall, 72, 73, 142; Europe: investment in Thailand, 55-56 rise, 99, 102, 140, 199 Exchange equalization fund, 207 Index 253 Exchange rate policy, 10, 19, 34, 85-86, Foreign exchange: accounts. 41, 168; 168, 192-93, 204, 215-16, 227; reserves, I1, 74, 132, 229, 235 record, 205-09; and relative pric- Foreign investment, 44-45. 55-57, 71, es, 216-26 72. 75, 124, 186-87. 230 Exports, 48-53; agricultural, 9, 48-49, Forward premium exchange rate. 175-77 53; manufactured, 48-51, 53, 74- Free medical care, 14 75; promotion. 52, 79-80; quotas, French, 8-9 79; restrictions, 85; shock, 99; Fund for Rehabilitation and Develop- tax, 52. 75, 78-79, 162 ment of Financial Institutions, External Public Debt Law, 239 41 -42 External shocks, 1-2, 54, 99-105; ef- fects, 106-16; response, 117-35. Germany: capital controls. 175 228-30 Gould, Joseph S., 61 Government: system 68-69 Farmer Aid Fund, 35 Government bonds, 11 - 12,42, 87, 188. Feldstein, Martin, 178 234 Fertilizer project, 88 Government Housing Bank (GHB). 39 Finance companies, 39, 41 Government revenue, 21, 75-77 Finance Ministry: banks, 10, 28, 41; Government Saving Bank (GSB) 39 currency ~ovrnen Savinge Banko 11SB; 39- currency reserve ratio. 11, ex- Governments: civilian. 14-15; military. change rate control. 85; export 13-14. 15-6 quasi civilian, 16- promotion. 80; finance compa- 17 nies, 41; financial institutions, 86; foreign borrowing, 15, 69; Great Depression, 2 1 -22 macroeconomic policy deci- Gross domestic product (GDP): bank sions, 69, monetary stabilization, credits ratio. 194; bank deposits 239; NDPC, 93-95; public enter- ratio, 194: estimates, 60-61, 64- prises, 142, subsidies, 142; tax, 65: growth. 71, 7, 74. 1p24, 181- 75, 81, 170 83; and monetary policy, 1 1s-83; Financial: liberalization, 204; markets, per capita, 45-46; sector shares, 42, 86-87; policy, 19 44, 47, 49; total, 42. 44 Fiscal: adjustments, 138-43; stabilizers, Gross national product (GNP) per capi- 143-65, 230-31; system, 75-77 ta, 1, 42. 44 Fiscal Policy Office: public enterprises Growth boom, 44-45. 236 92-93; taxes 75, 81 Foreign borrowing, 54, 93; budget, 68; Hong Kong: investment in Thailand, external shocks and, 131-32; Fi- 45, 56 nance Ministry, 15, 69; military, Hong Kong and Shanghai Banking 95-96, 127; private, 42; public Corporation, 10 sector, 15-16, 88-89, 93-96, 127, Horioka, Charles, 178 130, 136, 163-65; withholding Household income: percentage share, tax, 42, 170-71 3 1 Foreign debt, 54-55, 74, 89, 123, 127, Hunter, Sir Bernard, I0 235; public enterprises, 93,96, 142 254 Thailaindt' Macroeconomic Miracle: Stable Adjustment and Sustained Growth Imports, 53, 105; prices, 123; restric- Investment Promotion Act, 79 tions, 34,84; tax, 75, 162-63, 166 Investment-savings gap, 186 Income: distribution, 2, 57-60, 72; Irrigation, 88 growth and monetary policy, Islam, 31 199-200; growth stabilization, 203; per capita, 71 Japan: capital controls, 175; investment India: balance of payments, 55 in Thailand, 55-56 Indochina: funds transfer, 102 Japanese occupation, 10, II Indonesia: balance of payments, 55; foreign borrowing, 54; growth Kriangsak Chamanand, 16, 84, 134 boom, 45, 236 Krung Thai Bank, 39, 40 Industrial Finance Corporation of Thai- Kukrit Pramoj, 14-15, 136 land (IFCT), 39, 81 Industrial Promotion Act, 71 Labor Department, 87 Industrialization, 139 Labor force: sectoral shares, 47-48 Industry: interventions, 79-81 Labor market, 37-38; regulation, 87 Industry Ministry, 80 Labor mobility, 38 Infant mortality rate. 31 Labor Relations Act, 38 Inflation, 53-54, 122, 140, 156, 197, Legislated fiscal rules, 234-35 200; control, 168; external Letters of credit (LCs): control, 210- 11 shocks and, 229; and monetary Life expectancy, 30-31 policy stance, 201-03, 233; and Literacy rate, 31 petroleum prices, 142, 194, 201, London interbank offer rate (LIBOR), 204, 209: rate, 2, 53-54, 196, 171 209; World War II, 10,11 - 12,22, 233 Macroeconomic: conservatism, 232- Infrastructure provision, 13, 71, 72, 75, 35; data, 60-66; environment, 34; 77, 87-89, 139 performance, 42-43, 197-98; pol- Insurance companies: government icy decisions, 69 bonds, 11-12 Mahisra Raj Haruetai, Prince, 10 Interest rates: ceilings,41, 170,192-97, Maize: quotas, 79 204,209,210,230-31; domestic, Malaysia: growth boom, 45, 236 171-73; 180, 207-08; Eurodollar, Manufacturing: dispersion, 73; exports, 174, 180, 192; interbank, 180; in- 48-SI, 53, 74-75; GDP share, 44, ternational, 42, 73, 171-74, 180: 47, 49; growth, 72, 75, 230; labor and monetary policy, 181-83; force share, 47-48; private sector, parity, 173-78; and petroleum 13; protection, 35-37; tariffs and price increases, 194; regulation. import restrictions, 34 34, 86, 204; World War 11, 12 Market structure, 31, 34-42 Interior Ministry, 69, 87 Middle class: demand for democratic International Labour Office, 87 reform, 7 International Monetary Fund: loan, 74, Military, 7, 237; coup, 15-16, 17-18, 210 136; foreign borrowing, 95-96, Interventions, 78-87 127; governments, 13-14, 15-16; Index 255 involvement in politics, 18, 136; Nattapong Thongpakde, 37 monetary policy, 229-30; pro- Natural gas, 53, 102, 105, 134 curement, 95 Natural resources: deterioration, 72 Minimum wage, 34, 38, 87, 212 Net claims on financial institutions Monadjemi, M., 178 (NCF). 184-85, 191 Monarchy. 7, 9- 10 Net claims on government, 184-85, Monetary base: control, 184-89; 187. 191 growth, 181-83 Net domestic credit (NDC), 184-85, 191 Monetary policy, 168-69, 229, 233; au- Net export performance ratio (NEPR). tonomy, 169-83; direction chang- 48-52 es. 190-91; and GDP growth. 181- Net foreign assets (NFA), 184-85, 187. 83; and income growth, 199-200; 188 instruments, 190-97; and interest Net other liabilities (NOL), 184-85 rates, 181-83; performance, 197- Nonmetallic manufactured products. 204; World War II, 10 4950) Money: velocity. 183 Nontradables. 217-26, 229 Mongkut, King (Rama IV), 9 Northeast Asia: investmentin Thailand, Mundell-Fleming hypothesis, 169 55 Northeast region: poverty, 59 Narongchai Akrasanee, 36 Nukul Prachuamoh, 210 National accounts, 64-65 National Administrative Reform Offsetting capital flows hypothesis, Council, 15 187-89 National Banking Bureau, 10 Oil fund. 84. 134 National Committee on State Enter- Organisation for Economic Co-opera- prises. 92 tion and Development (OECD), National Debt Policy Committee 144 (NDPC), 93 Organization of Petroleum Exporting National Debt Reform Committee, 89 Countries (OPEC), 44, 72, 73 National Economic and Social Devel- opment Board (NESDB), 13; GDP Paichitr Uathavikul, 212 data, 60-61: macroeconomic pol- Pairote Wongwuttiwat, 36 icy decisions, 69; national ac- Paitoon Wiboonchutikula, 37 counts, 64; national income. 44; Parliament. 16 public enterprises. 92-93 Parliamentary democracy. 14 National Economic Council, 71 Petroleum products: imports, 53, 81. National Economic Development 84, 99, 102, 105, 131-35; price Board (NEDB), 71, 93 increases, 16,44,53.72,73, 102, National Executive Council, 14 105. 133-34, 142. 194, 209,235: National income. 44, 106-12, 229, 235 pricing policy, 81, 84, 105; subsi- National Income Accounts, 64, 65 dies. 134; tax. 84, 134 National Peace Keeping Council, 17 Phibul Songkhram, 13 National Revolutionary Council, 16 Philippines: balance of payments, 55; National Statistical Office (NSO), 57 civil service, 19; economic col- 26 iha11jbiles .& Mar,c F(o fllomic Miracle: Stable Adljustmen,t a,id Sustained (osathl lapse, 73. 235-36; education, 31; Rachain Chintayarangsan, 37 foreign borrowing. 54; foreign Racial conflict, 7 investimient, 445; GDP, 44; house- Rama IV. See MongkuL, King hold incomile 31 : petroleum price Rama V, See Chulalongkorn. King increases. 44 Rama VI, See Vajiravud, King Phote Sarasin. 13 Rama VIl See Prajadhipok, King Political history, 8-19; parties, 14, 18; Rarna IX. See Bhumiiipol Adulyade;, stability, 7, 18-19 King Political Parties Act, 13-14 Real output, 1-2 Politicians: and monetary policy, 233- Recession. 142. 200, 229 35 Rediscounit tacilities, 40 Populationi, 30-31; growth rate. 71 Rice: exports. 44, 53; import ban, 84; Poverty, 57-60; absolute, 2, 59; reduc- price increase, 21: price support tion. 59-60; rural areas. 31. 59, prograrm 14, 35; quotas, 79; re- 7 2 ser\ e scheme, 79: taxation, 35, Pra.iadhipok, King (Rarna VIl), 9, 21 78-79 Prapas Charusathien, 13-14 Roads, 88 Praya Sonaularaj. 10 Rubber: export tax, 34, 79 Prein Tinsulanonda, 16, 17, 84, 95, Rural-urban migration, 38, 72 134,201,210,212,213 Press censorship, 16 Sak(kditit, 21 Price: contr ol, 8 I, 84, 85; indices, 103- Salaries: civil servants, 21, 22-26, 157. 05: movements, 221: stability. 233 20 1-03 Sangad Chaloryu, Admiral, 15-16 Price Setting and Antimonopoly Act, Sanya Dharmasakdi, 14 85 Sarit Thanarat, 13, 90, 234 Private lirms: foreig,n borrowing, 42, Saving Bonds in Emergency, 12 186; trade unions, 38; wages, 38 Savings. 73-74, 123-27, 136. 194, 196; Privatte sector investment, 158-61; pro- Forced, 12 motioin. 13 Sayam Bank. 39 Product markets, 34-37 Securities Exchange of Thailand. 42 Public eniter-prises, 89-93; capital ex- Seni Pramo;, 14, 15, 136 penditure, 141-42; foreign bor- Services: (ti)P share, 44, 47: labor force rowing. 142, 186; foreign debt, share, 47-48 93, 96, 142; trade unions, 38 Siam Commercial Bank, 10, 39, 40 Public expnCIditure, 77, 139-41 Silver: price rise, 21 Public sector, 68-70, 138; data, 65-66; Singapore: investmenit in Thailand, 56 deficits, 154-55; foreign borrow- Slaughterhouses, 79 ing, 15-16, 93-96, 136; invest- Small Industries Finance Office menit, 13, 158-61, 167; role. 67; (sio), 39 savings. 124 Snobi Unalkul, 21 0 Puey Ungphakorn. 27, 93-94, 233 Social Action party, 14, 15 Social indicators, 3 1-33 Quasi money. 196 Socioeconomic surveys, 57, 59 Index 257 Sommai Hoontrakul. 211 Trade: balance. 19-20, 21 1; free, 9; in- Sompop Manarungsan, 45, 46, 60-61 dex, 1: performance, 80-8 1; re- Squire, Lyn. 38 strictions, 84-85; terms, 47-53, State Lottery Bureau, 91 73, 102. 10)6, 206, 235; World State Railway of Thailand, 91 War 11. 10, II Statistical Yearbook of Thailand, 61 Trade unions. 38. 213 Steel project, 88 Trairong Suwankiri. 36 Stock exchange, 42 Transfter pricing. 42 Structural change, 47 Trescoti, Paul B., 61 Student protests. 14. 15 Suchinda Kraprayoon. 17-18 Unemployment, 38 Sugar, 35. 52; import ban, 85 United States: interest rates, 171-74, Sunthorn Kongsompong, 17 177; investment in Thailand. 55- 56: military base closure. 14: Taiwan: investment in Thailand, 44-45, military spending, 71, 72, 95, 56 102, 105 Tambon development funds, 14 United States Agency for International Tariffs, 34, 76-77. 79 Development (USAID), 90 Tax, 75, 162-63; agriculture, 34-35; U.S. dollar: appreciation, 206, 208; and business, 162-63, 166;corporate, baht. 1), 19, 44. 85-86, 205-08, 39; export, 52, 75, 78-79, 162: 215-16, 227 import, 75, 162-63. 166; income, 21, 22, 76. 162-63, 166; indirect, Vajiravud, King (Rama VI), 21 75, 138-40, 166: petroleum prod- Viet Nam: tunds transfer. 102; threat, ucts, 84: reforms, 142; refund IX, 88: U.S. withdrawal, 105 system, 80: revenue, 140; sales, Virabongsa Ramangkura, 21(0 162, 166: time deposits, 196: VAT, 77: withholding, 39, 42. Wages: controls, 34 170-7 1 Welfare programs. 14 Technocrats. 234 Williamson. W. J. F., 21 Textile industry. 49, 52; exports, 53 Wilson. Constance M., 61 Thai Bankers Association, 40 Wiwat, Prince, 11, 20, 27, 233. 237 Thai Farmers Bank, 30, 210 Workers' remittances, 102 Thai Military Bank. 39 World Bank: advisory mission, 13, 28- Thai traditions and institutions, 18 29, 71: loans, 94; macroeconom- Thailand Tobacco Monopoly, 91 ic data, 64 Thammasat University: student pro- World War 11: inflation, 10, 11-12, 22, tests, 14, 15 233: Japanese occupation. 10, Thanin Kraivichien, 15, 136 1 1; monetary policies, 10 Thanom Kittikachorn, 13-14. 15, 136 Total factor productivity, 114 Tourism, 102 Tradables, 217-26, 229 \: D: d \ :: : :654 E 126:54:; lL 0 ffiS i k iA ; 00187 78081 32654 S; 2 : :: S : :$1 . 9