10433 mafts* Economic Development Institute "q*pU of The World Bank Managing Policy Reform in the Real World Asian Experiences Edited by Geoffrey Lamb Rachel Weaving Report No. 10433 EDI SEMINAR SERIES EDI SEMINAR SERIES Managing Policy Reform in the Real World Asian Experiences Edited by Geoffrey Lamb Rachel Weaving The World Bank Washington, D. C. Copyright © 1992 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing February 1992 The Economic Development Institute (EDI) was established by the World Bank in 1955 to train officials concerned with development planning, policymaking, investment analysis, and project implementation in member developing countries. At present the substance of the EDI's work emphasizes macroeconomic and sectoral economic policy analysis. Through a variety of courses, seminars, and workshops, most of which are given overseas in cooperation with local institutions, the EDI seeks to sharpen analytical skills used in policy analysis and to broaden understanding of the experience of individual countries with economic development. In addition to furthering the EDI's pedagogical objectives, Policy Seminars provide forums for policymakers, academics, and Bank staff to exchange views on current development issues, proposals, and practices. Although the EDI's publications are designed to support its training activities, many are of interest to a much broader audience. EDI materials, including any findings, interpretations, and conclusions, are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. Because of the informality of this series and to make the publication available with the least possible delay, the manuscript has not been edited as fully as would be the case with a more formal document, and the World Bank accepts no responsibility for errors. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to 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 photocopy portions for classroom use is not required, though notification of such use having been made will be appreciated. The backlist of publications by the World Bank is shown in the annual Index of Publications, which is available 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, Banque mondiale, 66, avenue d'Iena, 75116 Paris, France. At the time of writing Geoffrey Lamb was chief of Policy Development in the World Bank's Policy and Review Department. Rachel Weaving is information officer, Operations Evaluation, in the World Bank. Library of Congress Cataloging-in-Publication Data Managing policy reform in the real world: Asian experiences / edited by Geoffrey Lamb, Rachel Weaving. p. cm. - (EDI seminar series) Papers presented at the EDI/ KDI Seminar on Economic Policy Reform and Government Processes held in Seoul, Korea, in November 1987. Includes bibliographical references. ISBN 0-8213-1964-7 1. Asia-Economic policy-Case studies-Congresses. 1. Lamb, Geoffrey, 1944- . II. Weaving, Rachel, 1947- . III. Economic Development Institute (Washington, D.C.) IV. Han 'guk Kaebal Yon 'guwon. V. EDI/KDI Seminar on Economic Policy Reform and Government Processes (1987: Seoul, Korea) VI. Series. HC412.M338 1991 338.95-dc20 91-36478 CIP EDI Catalog No. 405/060 FOREWORD This collection of papers explores the process of economic policy reform. The papers were prepared for a policy seminar sponsored by the Economic Development Institute of the World Bank and the Korea Development Institute. Experienced policymakers from several countries, World Bank staff, and academics met to exchange experiences in the management of economic policy, particularly in the context of structural adjustment. The chapters in this volume set forth a number of different national experiences in an effort to identify some common principles about how economic policy reform actually happens. Amnon Golan Director Economic Development Institute iii CONTENTS Acknowledgments vii Preface ix Geoffrey Lamb 1. Economic Policy Change and Government Processes 1 Edgardo Boeninger 2. Sri Lanka: A Decade of Food Policy Reforms, 1977-87 26 M.D.D. Pieris Secretary, Ministry of Food and Cooperatives, Sri Lanka 3. Exchange Rate Reforrn in Pakistan, 1982 44 Javed Hamid and Naved Hamid Professors, Graduate School of Business Administration, Lahore University of Management Sciences 4. Privatization of State Enterprises in Bangladesh, 1975-84 57 Tawfiq E. Chowdhury Joint Secretary, Finance Division, Ministry of Finance, Bangladesh 5. Tax Reform in Indonesia, 1984 71 R. Mansury Director of Tax Regulations, Ismail Tamsir Deputy Director, Directorate of Direct Taxes Directorate General of Taxation, Indonesia 6. Trade Liberalization Policy in the Republic of Korea, 1980-85 86 Ohm Young-Suk Dean Graduate School of Management Information Systems, Hankuk University of Foreign Studies, Republic of Korea 7. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 98 Ammar Siamwalla Program Director, Agriculture and Rural Development, Thailand Development Research Institute Bibliography 119 v ACKNOWLEDGMENTS This volume owes a great deal not only to its authors, but to the participants in the joint EDI-KDI seminar on Economic Policy Reform and Government Processes held in Seoul in November 1987. Their wisdom and experience are reflected, even if incompletely, in these pages. A particular debt of gratitude is owed to Dr. Phaichitr Uathavikul, the moderator of the seminar, whose experience, insights, and occasional sharp discipline did much to focus the debate. A comparable debt is owed to Dr. Whang In Jhoun, then Director of KDI's International Development Exchange Program. Bank staff involved included Suman Bery, Ravi Gulhati, Arturo Israel, and the editors of this volume. vii PREFACE By now we know quite a lot-not enough, but a lot nonetheless-about economic adjustment and policy reform. In the 1980s, developing countries accumulated a great stock of experience, often bitterly won, about what needed to be done, and in roughly what order, for their economies to regain the growth initiative in a new, volatile, and hyper-competitive global economy. But knowing about desirable policy objectives and even sequences of reform is not enough. Reform is above all a matter of politics and government action: unless coalitions can be built and sustained, bureaucracy mobilized, and policies effectively formulated, financed, and implemented, they fail-as we have too often seen. Until fairly recently, the political economy of policy reform in this sense has become something of a gap in the knowledge of the academic community, external financing and advisory institutions, and often policy decision makers themselves. The reformers, after all, are frequently technocrats and former academics thrust into new ministerial or senior official roles, more comfortable discussing first-best solutions than negotiating with fractious political constituencies. And even experienced politicians often mainly know one country, and one way of getting things done. The inspiration of the Economic Development Institute (EDI) and the Korea Development Institute (KDI), the joint sponsors of the seminar on which this volume is based, was to bring a group of policy decision makers, most of whom had occupied senior positions on both the technocratic and political sides of government, together with World Bank staff and academics to explore policy reform as a governmental process. On the Bank side, the seminar was building on a growing body of work on the management of economic policy, particularly in the context of structural adjustment programs.1 In the deliberately varied experiences reflected in the papers in this volume, we were looking for common lessons about how economic policy reform really happens in practice.2 As Edgardo Boeninger's overview chapter vividly demonstrates, there are indeed strong commonalities in the experiences recounted in these pages, notwithstanding the marked differences in policies and in the countries surveyed-from some of Asia's poorest to the richest and most dynamic developing economies of the region. A few reflections may complement Boeninger's discussion. First, changing the course of established policy requires reformers to be eclectic in approach: policy reform is no occupation for the tidy or linear mind. The notion of an orderly policy cycle (analysis, policy formulation, authoritative decision, and implementation) tends to dissolve in practice. This is especially so in the frequent cases where real policy is the outcome of numberless implementation decisions made on the ground, subject to the peculiar rationalities of administrative allocation and of the micropolitics of everyday life. Nevertheless, a clear 1. See, for example, Geoffrey Lamb, Managing Economic Policy Change: Institutional Dimensions, World Bank Discussion Paper No. 14,1987. 2. The authors of the country papers in this volume are shown in the positions they held at the time of the/KDI seminar. Most of them have since moved to other responsibilities; in any case, however, each of them wrote in a personal capacity, although with the knowledge and encouragement of their institutions. ix x Preface ultimate objective and careful preparation can do much to ensure the staying power of reforms- and indeed without that grounding, it is all too easy for reform efforts to cross the line from pragmatic adaptation into pure improvisation. Second, that quality of reasoned pragmatism needs to extend to expectations. Achievements in policy reform mostly turn out to be modest over the medium term. After a while, policies seem to converge to some societal mean: reform can seem a Sisyphean effort, even though with luck and skill the stone will not roll all the way to the bottom of the hill again. Reformers relatively seldom have the opportunity to break the entire mold of economic policy, and when they do-as has happened so dramatically in Eastern Europe and the Soviet Union in recent times-there is a quantum jump in the danger and complexity of the problems they face, as well as in the potential for positive change. Third, the search for reform, whether incremental or more radical, also tends to get more complicated with the advance of pluralism in government and in the society at large. This fact of life-again illustrated in the experiences recounted in this volume-may mean that policy technocrats need to be a little humble, and be more willing to learn some lessons from successful political practitioners. We may need to think less about politics in the way we normally do- as a negative factor, a constraint on "rational" policy choices-and more as a positive way to get things done: to resolve social conflicts, and to build coalitions that have a direct interest in sustaining growth-enhancing reforms. The successful policy reformers of recent years have for the most part used politics brilliantly as an asset, not tolerated it as a constraint. Fourth, in the face of all the political and institutional complexities of implementing really effective economic change, the importance of the reformers themselves-sometimes an individual, sometimes a cohesive group-nevertheless comes through rather forcefully. The political process of policy reform is always complicated and tough-but it is also often malleable, and productive of sudden political opportunities. So there remains, in the last analysis, a critical role for individual clarity of vision, personal initiative, and even sometimes sheer obstinacy. Especially when successful, these qualities on occasion earn a more elevated name: leadership. Successful reform needs that too, perhaps more than most other attributes. Geoffrey Lamb 1 ECONOMIC POLICY CHANGE AND GOVERNMENT PROCESSES Edgardo Boeninger This volume discusses issues in economic policy making and implementation arising from a seminar sponsored by the Economic Development Institute (EDI) of the World Bank and the Korea Development Institute (KDI). The seminar was held in Seoul in November 1987 as part of the Senior Policy Seminar series which EDI started in 1981. Instead of focusing on the substance of policy, the Seoul seminar was concerned with the underlying processes of policy making and policy implementation. It was therefore something of an experiment. In this field, though much has been written by academics, the state of the art is still relatively primitive. In particular, it is difficult to extract from the literature a coherent framework applicable to the task of meeting specific goals in particular settings. Moreover, it is widely felt that the perceptions and experience of practitioners have not been sufficiently explored and systematically related to the analytical literature. By documenting and analyzing practical experience, it was hoped that the case studies and the seminar would contribute to the development of a framework within which-while drawing on theoretical work-experience can be analyzed, understood, and used as a basis for action. Participants prepared case studies for the seminar to illustrate how policy work has been carried out in each of six Asian countries: Bangladesh, Indonesia, Republic of Korea, Pakistan, Sri Lanka, and Thailand. The basic subject of the seminar-how policy change has been and should be managed, was approached with reference to four main themes: * political dimensions of the policy process; * the role of information, professional knowledge, and analysis in policy making; * organization for decision making; and * issues in implementation. In addressing these themes, this paper attempts to relate the findings of the case studies and seminar discussions to existing literature. Defining the Issue of Policy Change What are the processes leading to breaks with the "business as usual" of bureaucratic routine? Sources of Policy Change Major policy reform is often closely linked to a change in government or ideology. This was true in the Bangladesh and Sri Lanka cases examined, and to a lesser extent in the Thailand case. Incoming leadership often reaches power having made reform the basis of its political platform (either in an electoral scenario or as justification for a coup d'etat). But even within an existing governmental power structure, policy change can be expected when certain ideologies or economic policy orientations become more decisively prevalent and gain in technical coherence. 1 2 Edgardo Boeninger A second source of policy change is anticipatory analysis by influential technocrats aware of impending crisis or concerned with long-term issues. (For their analysis to be relevant, technocrats must have influence over policy makers, as will be discussed below.) Politically strong and ideologically committed governments tend to give officials, especially politically sympathetic senior civil servants and advisers, a relatively free hand with the design and elaboration of policy reforms within their overall political programs. Officials in such positions often make full use of such opportunities to influence both the direction of policy and its technical quality. Governmental change may contribute decisively to an increase in the policy role accorded to technocrats. As part of the strong modernizing trend in developing countries, the political programs of incoming governments have been increasingly based on blueprints prepared by teams of economists and experts on specific subjects. The new programs tend to assign a key role to technocrats and to the information and analysis that the latter bring to bear on policy problems. A third situation is that brought about by unforeseen crises requiring significant and speedy changes in standing policies. Though the normal management of policy reform is informed by a long-term view (for example, Thailand may want its economy to become more like Korea's), in a crisis, managers may simply have to act. Obviously, such circumstances place policy makers in a defensive position and reduce their room for manoeuvre. In this realm of crisis management and contingency planning, political survival and the cost of muddling through may be the best one can hope for. This emphasizes the importance of foresight within the policy-making establishment; crises may come up at any time, but where their form is anticipated it is easier to resolve them without undue damage to established long-term objectives. In the experience of participants at the Seoul seminar, policy changes were often undertaken in response to a looming crisis, leaving rather little time for analysis. Where this is true, measures are presumably needed to strengthen the technical capacity to anticipate crisis. If normal economic management is good, however, crises may not develop. For example, if a central bank pursues enlightened policies, exchange rate problems may never arise. Timely action on the exchange rate in the Pakistan case, to delink the rupee from the US dollar, is a good example of such behavior. So is the tax reform undertaken in Indonesia, to diversify sources of tax revenue while the price of oil was still high, successfully avoiding the fiscal crisis that would otherwise have followed the collapse in the price of oil. If a crisis does develop it may produce worthwhile opportunities for substantial reform through shock treatment, an increase in the political influence of technocrats, a new political awareness on the part of the political elite, or changes in leadership. But to improve the quality of the policy process under such circumstances, it is crucial to keep the initiative firmly in the hands of policy makers. By the same token, political leaders may need to be made aware of and persuaded to seize the opportunities for reform that are inherent in a crisis. Most political leaders, unless driven by ideology or forced by circumstances, show an understandable aversion to risk and an equally strong preference for "business as usual" in policy making. The Policy-making Process as the Framework for Policy Change Radical change within the bureaucracy is rare. Most policy reform takes place within the framework of existing practices. What are these practices? What influences them? The case studies show both similarities and differences in the policy processes of countries that differ in their national cultures, bureaucratic traditions, and political systems. Though we are not concerned here with comparing these factors and their influences on the policy-making process across countries, they are factors that cannot be safely ignored. Issues such as the role of, and most advisable administrative locus for, professional expertise; the prevailing style in decision making (for example consensus building, versus direction from the top); or the choice of effective ways to exercise leadership or gain public support for policy, can only be correctly Economic Policy Change and Government Processes 3 addressed if the constraints and behavioral features produced by a country's particular culture and traditions are sufficiently understood. The political system is the single most important influence on a country's policy process. Economic policy reform is both promoted and complicated by the fact that many developing countries are going through periods of profound and difficult political transition. The political context in which governments and official policy makers work has become more plural, whether by moving from military to parliamentary-type rule, as in Pakistan and Bangladesh, and somewhat earlier in Thailand, or by an initially fairly controlled broadening of political debate, as in Indonesia and Korea, or more generally because of a greater degree of political awareness, economic diversification, and rising social demands, as in practically all of the countries featured in the case studies. The trend toward increased social differentiation, political pluralism, and more democratic institutions carries with it a greater need to consider consensus building, public debate, and interest-group participation as key elements in effective policy making. Meanwhile, prevailing cultural traits may make open disagreement about policies less socially acceptable in some countries than in others. (Where open disagreement is not socially acceptable, people use various means of informal and behind-the-scenes consensual decision making, sometimes as part of a bureaucratic tradition involving organized interest groups or influential individuals such as leading businessmen or military commanders.) Political liberalization, including freeing the press as well as allowing an effective role for parliament and political parties, will progressively weaken and put an end to direction from the top, secrecy, and elite hegemony as the prevailing styles of decision making. Where this happens the political dimension of the policy process is enhanced and it must therefore be carefully considered in examining the conditions for successful policy reform. The Need for a Predominant Public Philosophy Effective policy making needs to be carried out in the context of a predominant "public philosophy' (Smith 1983), though this philosophy need not be an ideological or rigid one. In Smith's words, "even hard-headed realists yearn for some measure of coherence. In general, the exercise of leadership, based either on the political will of a governing elite or on the personal charisma of an individual, cannot be expected to have a lasting impact unless sustained by content in the form of a body of ideas giving clear indication of purpose." For policy to be coherent, decision makers in various levels and areas of government need to agree on objectives and on the kind of policy instruments best suited to carry them out. Nationalization in Bangladesh, though partly produced by a series of specific events, was the result of a deliberate political opting for socialism, explicitly written into the country's constitution. Later denationalization reflected a fundamental change, politically reflected in a constitutional amendment that emphasized nationalism and Islam while at the same time redefining the concept of socialism. In Korea, the decision to achieve growth primarily through market forces provided the necessary guiding principles for liberalizing trade. Where ideological dispute, or contradictory preferences as to how goals should be attained, is added to customary bureaucratic rivalry and the need to confront interest groups in the population, weak and wavering policies are the most likely outcome. In Sri Lanka, concern for welfare-oriented policies goes back to the early days of Independence and reflects a set of national values that no social group would dare oppose. Sri Lanka was long considered an example of human-needs oriented development policy, inclusive of large-scale food subsidy programs. When a new growth-oriented government came to power in 1977, redistributive populism was never entirely abandoned and efforts to protect the most vulnerable sectors of the population were therefore kept in place, but the redirection in policy towards employment and away from subsidized food did produce a significant regression in income distribution. Over the ensuing years political pressures mounted and subsidies were eventually forced up to nearly their previous levels. 4 Edgardo Boeninger Rationality and Optimal Solutions, vs. the Need for Pragmatism and Flexibility Because of the complexity of issues, conflicting interests, contradictory judgments, gaps in technical capability, and political opposition, governments usually face uncomfortable trade- offs in policy implementation. Though economic policy analysis ought to be, and usually is, undertaken in the spirit of rationality, seeking optimal solutions to every problem, in implementation the optimal solutions will be unattainable; this is the realm of the second best. To adapt to unforeseen circumstances, one may need to deviate from standing objectives and preferred policies. Ideologues are strongly biased against any change of course and tend to regard any such suggestion with deep suspicion. For the same reason, ideology is hostile to processes of political bargaining and compromise, though these play a key role in pluralistic settings. Within the framework of ideological change that may have given the impetus for reform in the first place, however, policy makers in general need to chart a pragmatic course. Flexibility and pragmatism become inevitable in a participatory environment. The more pluralistic the policy-making process becomes-in the United States, for example, it is exceptionally so-the more muddling through, coalition-building and compromise are probably the best that can be expected. In the presentation or "packaging" of policies, a certain degree of ambiguity has a place, since this makes it easier for different sections of society to interpret a proposed policy differently, with the authority stressing the aspects most advantageous to each. But though ambiguity may be helpful in this context, it should stop short of any significant loss in clarity of the objectives being pursued. Economic reforms are often conceived together as parts of a comprehensive program. However, given administrative constraints and, often, the need to avoid arousing an anti- reform.coalition, their implementation may need to be done piecemeal. In such cases, the leader guiding the process should nonetheless maintain an overall concept of the reform agenda, so as to ensure its coherence over time, even if this agenda may have to be disguised or kept secret. Political and Technical Feasibility Policy must be both politically and technically feasible. Political and technical feasibility, and how these considerations interact in the decision-making process, depends on the nature of the political system and the degree of pluralism in society. Within the framework so defined, it is essential to have mechanisms that provide for joint analysis of both political and technical variables, so that both are understood and effectively considered. In most of the countries featured in the case studies, economists have tended to be market- oriented, and strongly committed to economic growth and to macroeconomic equilibria, especially price stability. Politicians are often more concerned with the distribution of benefits from policies-particularly in an open society in which political elites need to be responsive to their constituencies. Economists point out that policies attempting to achieve greater equity may be self-defeating if implemented at the cost of stagnation. In turn, political leaders and social scientists cite the moral imperative of equity and the need for social peace as a requisite for dynamic growth. Different countries have different institutional mechanisms for informing the legislature and increasing their understanding of economic policy issues. For instance, members of the legislature may have an advisory staff, as they do in the United States. As this may be too expensive a model for many developing countries, an alternative is to give the political elite easy access to civil servants and to systematic explanatory information. This was the model followed in the case of the Indonesian tax reform; to explain the proposed reform to members of parliament, officials from the tax administration spent two weeks with a special task force of parliamentary representatives, going over the new tax laws in detail. Economists within government, for their part, should work in close contact with social scientists and politicians and submit their analysis and proposals to the judgment of the politically sensitive. Ultimately it is not technocrats who make policy. But where individuals Economic Policy Change and Government Processes 5 with technocratic backgrounds and viewpoints play political roles, there is a strong argument for providing them with political insight. The Political Dimension of the Policy Process The importance of the political dimension of the policy process and, especially, of policy change, was rather little emphasized in the case studies but came up repeatedly in the Seoul seminar discussions as the single most crucial issue requiring further analysis and better understanding. Technocrats participating in the policy process naturally have only limited influence on the course of political events, and in many cases it would be improper or counterproductive for them to attempt to exert such influence. Nevertheless, officials obviously have to be keenly aware of the political context-with respect to both constraints and opportunities-within which they work, and within which policy changes take place. The political dimension becomes a more complex and more distinct component of the policy process as countries move toward social differentiation and political pluralism. While this is clearly a very large and complex subject, two points may be noted. First, authoritarian conditions make it easier to push through difficult reforms. With greater pluralism, policy change will almost inevitably tend to become more negotiated, compromising, and based on more explicit recognition of social compensation and equity, however equity may ultimately be politically defined. Second, civil servants and policy advisers who formerly were only responsible to their minister, or head of government or head of state, may now face a much more complex set of accountabilities to a wider political leadership, to parliament, and-especially via press commentary-to the public at large. (And, while this has an immediate impact on the design and implementation of economic policies, it also has broader implications for the management and ethos of the public service in general.) Governance, Problem Solving, and Conflict Resolution For the present discussion a useful distinction may be made between politics, and therefore the exercise of power, in governance; politics as a dimension of problem solving; and politics as a process inherent in conflict resolution. PROBLEM-SOLVING. The main concern of the discussion in the Seoul seminar was with the second of these aspects: how to make problem-solving more effective through policy reform. However, the politics of problem-solving cannot be fully understood unless explicitly related to the first and third aspects: the exercise of power in governance, and the manner in which conflicts are resolved. Optimal policy, whereby problems are solved by policies based on objective fact and scientific analysis, has been the long-standing dream of many economists. But though technocrats often claim their proposals are politically neutral, this is only rarely the case. It certainly does not apply to the broad field of economic and social policy, where they are faced with trade-offs in the distribution of benefits and costs, and where their choices are continually subject to intense ideological debate. GOVERNANCE. Problem-solving takes place within a framework determined by the predominant public philosophy, which was alluded to previously. This framework is the result of choices with regard to fundamental options and orientations, and to the desired goals as well as the admissible policy instruments for meeting them. Governance according to this framework reflects the predominant ideology, and some broad vision of a country's future and of the path the development process is expected to or should be induced to follow. Governance can be expected to contain strong partisan elements reflecting the value- preferences, interests, and biases of those holding power. Policy will conform to such guiding principles; attempts to present proposed policies as nonpolitical are only attempts to disguise 6 Edgardo Boeninger the underlying preferences of policy makers, perhaps as part of a deliberate strategy for policy approval and implementation. CoNFLIcr RESOLUTION. Policies affect different socioeconomic groups in different ways, with some groups gaining and others losing. Policy making is thus inevitably linked to conflict resolution. Two kinds of conflict stand out. First, contests for hegemony, leadership status, or areas of dominant influence are inherent in intrabureaucratic relations. They reflect different interests, ambitions, and rivalries between individuals, groups, and institutions, and can be settled only if a certain distribution of power is defined and imposed on or accepted by the parties involved. The Thailand case illustrates the unstable power-sharing arrangements between the Ministries of Commerce, Agriculture, and Finance, and their relation to recurrent shifts in rice policy. The second type of conflict is disputes over political power at the national level, between social groups and political parties and institutions such as the armed forces. Obviously the outcome of these disputes affects policy. The Thailand case writer contends that the "overthrow of the military in 1973 had left Thai politics with no dominant focus of power. The contending groups engaged in a furious race for rural political support. As the conventional political parties could exert very little direct power, they exerted pressure on the Farmers' Aid Fund, heavily influencing its use since 1975." Organizing for Policy Making Developing an appropriate organization for policy making is a basic challenge for contemporary government. Attempts at radical change, when lacking sufficient support, may end in reversal or even in a government's collapse. If the state lacks power, policies may fail through stalemate and indecision. In a remarkably insightful article, Gawthrop (1983) presents alternative approaches to organizing for policy making: CONTROL: In an authoritarian environment, technocrats and leaders of the elite directly controlling the state (the military, the party) control all aspects of policy making and therefore are the country's only de facto politicians. In such a scenario, decisions are made at the top, relying mostly on control over the behavior of individuals so as to achieve stated objectives of the leadership. ANALYSIS: Organizing for analysis is an approach committed to rational decision making on the basis of a comprehensive information source for all necessary data. In fact, decision making from the top, relying mostly on control, has information and analysis fed directly into the upper echelons of the hierarchy, to facilitate rational decision making. The combination of rationality, hierarchy, and authority provides the ideal world for planners, relying on clarity of hierarchical design, internal procedural orderliness, and on discipline, loyalty, and obedience. B ARGAINING AND RESPONSIVENESS: In highly complex differentiated and pluralist societies, however, in which political and social pressures and open conflict are part of daily life, bargaining and responsiveness become necessary ingredients of the policy-making process. This means that policy decisions reflect agreements between groups, none of which enjoys direct unilateral control over any other. While hierarchy and rational analysis, as well as responsiveness to the demands of specific social groups, may call for radical change, bargaining identifies with incrementalism and with checks and balances, reflecting constraints on the exercise of power. Policy geared to responsiveness requires information on social expectations and demands, obtained through the use of surveys, opinion polls, interaction with leaders of social organizations, and so forth, and it must be effectively delivered. Economic Policy Change and Government Processes 7 In the real world the tension between purposive top-down action and necessary political and social responsiveness has to be accommodated. Seminar participants agreed that though the agents for change are mainly found among the political leadership and the top players of the bureaucracy, the present trend in most of the countries covered is to shift away from authoritarian scenarios in which those in power always get their way. Social acceptability is increasingly becoming a factor that cannot be safely ignored. Successive changes in rice policy in Thailand can most easily be understood as an exercise in conflict solving (between consumers and producers, and between the Ministry of Agriculture and the Ministry of Commerce). Trade liberalization in Korea is almost a textbook case of governmental guidance in pursuit of clearly defined long-term goals. This is also true of privatization in Bangladesh and tax reform in Indonesia. The 1982 devaluation in Pakistan is best identified as crisis management, facilitated by anticipation. Equity Considerations As societies develop and become more diversified and better educated, their demands tend to concentrate more and more on the basic issues of who gets what-that is to say, on the distribution of opportunities and benefits, of income and wealth and the chances for social mobility. Thus equity considerations come to the forefront of political conflict. Economic policy makers have so far found it extremely difficult to deal with the issue of distributional conflict and, often, even to insert the pertinent variables into policy formulation and implementation. In the first place, as was seen above, economics has more to say about efficiency than about equity in allocating resources. For this reason the distributive implications of policy are, perhaps unwittingly, often ignored. Yet the perception of having been unfairly treated or discriminated against breeds opposition to policy. This perception may decisively affect the outcome of the policy process if society becomes sufficiently sensitive to equity considerations. Even in strictly authoritarian settings, unorganized sections of the population can cause policies to fail by simply withholding their cooperation. Hence support-seeking to avoid noncompliance is always an indispensable part of policy. In many instances, an extended policy process has ultimately failed for lack of enough resources. Under heavy pressure as to the distribution of benefits, governments often choose one of two defensive courses of action-neither of which can really be expected to succeed either in solving problems or in keeping political protest within manageable bounds. One such course of action, apparently followed by the Thai authorities in the case of the Farmers' Aid Fund, reduces policy to merely satisfying key political actors, thus usually giving up on problem solving and often letting loose further pressures to benefit from the bounties of patronage. The other common way of facing equity issues consists in establishing a formal right of access to certain benefits for all eligible individuals-frequently through legal provision or some other kind of public commitment-while letting the resources allocated for such purposes severely constrain actual access to the benefits. In such cases, apparent policy may create unrealistic expectations, reinforcing existing pressures and escalating immediate demands, thereby intensifying perceptions of unfairness and government inefficiency. Winning and Keeping Political Support When political support is crucial to the success of a proposed policy, a deliberate choice should be made at the policy formulation stage as to whether the best way to confront political problems is to anticipate opposition, by designing or "packaging" the policy largely to emphasize its responsiveness to expected demands, or whether, alternatively, it is to try to build a coalition of parties or interest groups strong enough to gain approval for the policy and acceptance of its implementation. "Packaging", or the means by which policies are presented to the political leadership, communicated to the public, and combined with other measures, is always a critical factor in 8 Edgardo Boeninger policy success. First, except where policy is initiated by politicians, policy reformers have to "sell" their policies within government itself-in the first instance to their political masters, but often also to parliamentarians, party officials, and so on. This can require great skill, and the ability to sense the political agenda of the particular audience being addressed. Second, a reform-minded government can move to accustom people quickly to a policy change. In Pakistan and elsewhere exchange rate reform is inevitably a major political issue, but deftly handled it can quickly become widely accepted by the public at large and even in most directly affected business communities. The public relations and public education aspects of policy packaging are discussed in more detail later in this paper. Coalition building requires bargaining and compromise, and a willingness to adopt measures to compensate certain groups in society for losses they are likely to bear as a result of the intended policy change. A policy change prepared in a pluralistic setting should have margins of flexibility to allow for adjustments in the course of bargaining, or when support for the new policy begins to erode. The Thailand case writer notes that "Since 1978 the system of government determining the direction of policy has been a de facto coalition among the military, the elected politicians and the monarchy." The fact that "the relative strength of the partners rose and fell somewhat according to short-term events", when compounded with intrabureaucratic rivalries, goes a long way to explaining the variations in what the author describes as the yearly political rice cycle. The method of price support, the fixing of the support price, the amount of funding, and the identification of the institution responsible for implementation were decided "almost at harvest time of the wet season crop." Pluralism, however, seems to have established a need to "pay at least lip service to the idea of supporting rice prices for the farmers." The Thai case also describes another bargaining and support- winning method, when stating that political parties used the Farmers' Aid Fund to "set up a patronage system that allowed them and their supporters to gain financial resources to pay for their political activities." Timing It is perhaps obvious to say that good timing is essential for successful policy change, but in fact the case discussions throw interesting light on how complex political timing can be. First, policy reform is much more likely to succeed if implemented at a time of change in the predominant public philosophy, or political and ideological climate in which debate over policy choices takes place. In several of the cases considered, a process of change in the predominant public philosophy took place gradually over time and paved the way for reform, although actual policy change was usually prompted by a more immediate change in circumstances-whether a crisis situation, governmental change, or pressure from international agencies. In Thailand, for instance, the first proposal to eliminate the rice premium came from the National Economic Development Board as far back as 1955. Tax reform was first suggested in Indonesia in the early 1960s, while in Korea "during 1978-79 there were several reports regarding the need for structural adjustments of the economy, among which the reports of KDI, the Bank of Korea, and the National Council on Economic and Science were the major ones." Second, changes in public philosophy are most often accompanied by a change in government or regime. It is desirable to seize the opportunity of governmental or political change to initiate a policy reform. In other words, reformers have more room for manoeuvre if they act while they can still choose among several options for reform, rather than have events thrust a solution upon them. As stated by the Bangladesh case writer, "change should be completed as quickly as possible if it is a 'one shot' type of policy reform." In Bangladesh, as the case writer notes, "a violent change of government in 1975... led to a significant shift in political and economic strategy, towards the promotion of a mixed economy with emphasis on the private sector. The privatization of industry was initiated and pushed forward by the military leadership, which had to distance itself from the preceding Government in order to justify its presence." Economic Policy Change and Government Processes 9 Participants at the seminar noted how general and urgent was the conviction in Bangladesh, at the time, that nationalized industry had to be dismantled. In view of this, decision makers favored speedy implementation of a less-than-perfect reform package as against the alternative of prolonged analysis and consensus-building. "...A good deal of time pressure was put by senior government officials on the technocrats responsible for designing and implementing the reform, amid awareness that if the momentum of a reform was lost the reform might have had to be aborted." As the Sri Lanka case writer puts it, "A change of government, particularly when there is also a change of political parties, leads to a climate, at least for some time, where the public is more receptive to change." He hastens to add, "Since the honeymoon period is relatively short, it is of importance to make any fundamental changes of policy early and decisively." The concept of the honeymoon period is valid mainly in a democratic political system. If people accept democratic rules, the next election, several years away, gives them the chance to pass judgment on the regime's effectiveness. In such a setting the government, sustained by the legitimacy gained at the polls, is usually given a chance to go ahead with its declared policies. In authoritarian regimes, governments hold greater power and are not so easily challenged. However, their legitimacy rests solely on performance. Popular discontent may produce a social explosion at any time, so that even in such an environment a certain degree of support is essential for stability. Policy reform in such political systems is best introduced during periods of economic boom or other favorable circumstances. Information and Analysis As the Korean case stresses, for the policy process to be effective, policy makers need quick access to relevant information. Deficiencies in information often severely impede both policy making and operations in the public sector. Governments face difficult trade-offs between ignorance and cost, between comprehensiveness and timeliness, and between the analyst's and the political decision maker's perspectives. Information Gaps Decision making on policy matters is often hindered in various ways: • problems are poorly defined. When this is so, no clear signals are available as to the precise content and scope of the information needed to design appropriate policy; * too little information is available about the background of the problem; * too little information is available about alternative courses of action, and their consequences; * too little information is available about the values and interests of those involved in the policy decision; and * the time, skills, and resources for reaching a decision are very limited. (Forrester, 1984). Discussion at the seminar emphasized some recurring information gaps. First was a lack of meaningful fiscal data-making it very difficult to trace adequately the anticipated and possible effects of policy change on government revenues and expenditures. Another gap cited was in information about the domestic social structure and relationships-for example the distribution of wealth and income-which is often unreliable, tardily collected, and patchy. Of course policy makers often ignore such considerations deliberately, but without reasonable data it is all the more likely that policy will be based on ignorance, ideology, or prejudice. Worse, policy-induced changes in the pattern of income distribution then often appear as unexpected or unforeseen results, thereby contributing to the failure or unplanned reversal of policy initiatives. 10 Edgardo Boeninger Participants in the seminar discussion, however, argued that these disabilities, like other policy constraints, can be managed and ameliorated in some political and bureaucratic circumstances. Improving Information Systems In many developing countries, even basic data on population, demographic trends, and the economy may not be easily available or may be of uncertain quality. Wherever this is so, the establishment of a statistical system that can routinely provide basic data, perhaps rudimentary in its early stages, but capable of gradual expansion and improvement, is clearly urgent. For policy purposes information systems must provide systematic, relevant, timely, and reliable data to analysts and decision makers. To make this possible, such a system will need to be firmly set within an administrative routine that explicitly assigns responsibilities to departments and individuals for collecting, processing, and transmitting the information needed. Reliability depends on the quality of primary data sources and on the methods used for processing; relevance and timeliness can only be judged with respect to expected uses, and to political, economic, and social requirements, which change over time. As well as being responsive to such changes in priorities, information systems need to able to respond to special requests and unusual contingencies. The quality of routinely-provided information and statistical analysis within government tends to reflect conditions, expertise, and morale not only within the statistics department but in the civil service as a whole. An enclave of excellence in policy analysis and decision making can be critically hampered by both poor information and poor implementation, if the general conditions in government agencies are poor. This is one of the many reasons why economic and administrative reform may need to be conceived together. Policy analysis requires qualitative as well as quantitative information. Though numerical data must constitute the backbone of an information system, basic elements of political and social reality are often inadequately reflected in numbers. Policy analysts and decision makers will often require qualitative information on, say, social structure and relationships; the level, diversity, and strength of social organization; and the strength of local or interest group leadership (Reischauer, 1984). They have to use their hunches, their political judgment, and their common sense to combine different sources and kinds of information into policy analysis, against the background of past policy history. The Sri Lanka case study, for example, shows that political considerations were an important feature in the design of food subsidy schemes followed in that country since its Independence. All this does not necessarily argue for sophisticated or laborious information processing. The timeliness of information is often at least as important as its completeness. It would be instructive in this context to draw up a sort of information map for key economic issues, outlining, for a wide range of subjects, the data considered essential to policy formulation and indicating the areas where economic decision makers have typically tended to run into uncharted territory. In most of the countries that were featured in the seminar, governments have given a great deal of attention to developing better information systems by manning bureaus of statistics with professional staff, as well as through periodic census-taking and more frequent use of sample surveys. Foreign assistance has been widely available for this purpose. However, in several countries, statistical services appear to have been deteriorating because lagging salaries have led to loss of competent technical staff. Central banks, whose autonomy allows them to pay more competitive salaries, generally stand out as islands of competence. The Role of Analysis Seminar participants almost all agreed that improvements in an information system come about only in response to sustained demand for better information by analysts and decision makers. Analysis is a crucial ingredient of policy making. It serves various purposes: Economic Policy Change and Government Processes 11 anticipating outcomes, giving decision makers a medium-term perspective, supplying criteria for conflict solving and helping to strengthen the exercise of leadership and authority by adding technical legitimacy to political decisions. The anticipatory function of analysis involves throwing light on the probable macroeconomic effects of policy options, attempting an estimate of their costs and benefits, and alerting policy makers to eventual side effects on other sectors or economic variables. It also involves indicating the political and social consequences of proposed policies, notably by identifying winners and losers. Sound anticipatory analysis was a crucial reason for the effectiveness of the policy-making process in the Korea case, and it contributed decisively to the approval and implementation of tax reform in Indonesia. In the Bangladesh case, good design seems to have been sacrificed for speed of implementation. At the time, there appeared to be an important tradeoff: information might improve policy content but would take time to gather and analyze, while the state of industry, and the fact that the government had just come to power, made privatization appear very urgent. The experience, however, shows the importance of information; the reform could have been better executed had it been better conceived. In particular, the terms for transferring ownership should, with hindsight, have been better designed. When the new owners of the privatized industries defaulted on the payments they owed government for their purchases, the government had no recourse. In the Thailand case, an earlier political awareness of the inherent contradictions in the rice policy-whereby a price support program that was intended to stimulate production was financed by an export tax-might have prompted some actors to search for other ways of achieving the political compromise that was needed. By giving the participants in the decision-making process, as well as the people to be affected by the policy, a common background of processed information, analysis can contribute decisively to a rational debate and to consensus building, and thus help to prevent confrontation and resolve conflicts. In Korea, continuing analytical work has no doubt played such a role. At the same time it has provided the authorities with a guide to probable developments in the medium term, thus enabling them to sustain a coherent policy, while making adjustments as required by circumstances. Both the seminar discussions and the case studies revealed a strongly felt consensus with respect to the importance of systematic analysis, as well as a shared judgment as to the need for further progress in enhancing the quality, relevance, and effective role of analysis in decision making. Only in some countries (notably Korea, among those featured in the seminar) does analysis consistently play a central role in the policy process. The short-term obligations and pressures predominant in the minds of government authorities, as well as difficulties with respect to staffing, salaries, and institutional locus of the analytical function, have hindered progress in this area. As societies and political systems become increasingly pluralistic, and in an environment also noted for spectacular progress in available technical knowledge, decision makers can no longer dispense with sound analysis. Whoever can wield superior knowledge with respect to an issue enjoys a decisive advantage over contenders in the policy debate. Analytical Relevance and the Planning Function Recommendations for policy reform must be based on an assessment of past performance, using an analysis of the costs, shortcomings, or outright failure of existing policies to justify change. Recommendations couched in such analysis can be effective even in circumstances where change is not urgent. In Indonesia, the critique of the existing tax system prepared by local and foreign analysts pointed to its complexity, poor performance in raising revenues, narrow tax base, ineffectual provisions for incentives, and poor collection record, and ultimately led the government to move ahead with a substantial tax reform even though the country faced no crisis at the time. 12 Edgardo Boeninger Such analysis stems from a continuous process of evaluation, designed to monitor the economic and political impact of policies and to facilitate the adjustment of governmental action to new circumstances, as well as to unexpected or undesirable effects. The question of the role of anticipatory analysis leads directly to that of the role of planning. Once greeted as the universal prescription for optimal "scientific" policy making, planning has increasingly been discredited by the failure of national plans in most countries, including those formerly centrally planned economies now undertaking sweeping reforms. One should not conclude, however, that planning has no place in the decision-making process. The systematic presentation of likely trends and developments, both nationally and in relation to the world economy, allows the private sector as well as state agencies and public enterprises to project the future on the basis of shared information. Comparing options and examining opportunities improves the chances of coherence in public and private sector decision making, thereby helping to improve relations and establishing closer links between these sectors. Finally, to achieve unified action by a diverse and growing number of government departments and decentralized institutions requires a common frame of reference provided by some kind of planning exercise. Seminar participants stressed that good economic policy typically rested on firmly detailed plans made well in advance of the time they were needed. The Pakistani devaluation and, as stated earlier, the Indonesian tax reforms, were good policy decisions based on substantial advance planning. Seminar discussions also emphasized the need for analysis to be placed within a coherent framework. Though ideological commitment can lead to rigidity and insufficient regard to reality, analysts and policy makers are more often plagued by the reverse: contradictory signals as to the broad convictions and orientations of government. The Need for Objectivity and Balance There is no such thing as objective, purely technical, policy analysis. When analysts assess performance or argue in favor of specific policy recommendations they do so from particular ideological or theoretical perspectives. Awareness of these restrictions is indispensable for proper assessment of the advice provided by analysts. However, greater objectivity still remains a necessary and attainable goal. Prejudice, wishful thinking, misrepresentation of reality, erroneous inference, and outright lack of respect for indispensable scientific rigor are limitations that must be overcome if analysis is to be a relevant tool for policy making. Furthermore, objectivity is essential to uncover mere rhetoric, often attractively packaged, which can be a deadly threat to effective or simply sensible policy making. To be relevant and useful, analysis should satisfy three criteria. First, it should balance short-term and long-term considerations. Planners emphasize the latter dimension while budget analysts and economists working on monetary matters tend to focus mainly on the immediate. Policy makers should request prior consultation between analysts representing these two perspectives or be provided with both perspectives as inputs for their decisions. Second, policy analysis needs to combine independence with loyalty-an issue we shall return below to when discussing organization for decision making. Third, it should give attention to political and social issues. Lack of attention or sensitivity to issues such as the distributional impact of policy recommendations can make politicians resistant to analysts' advice (Verdier, 1984; Leman and Nelson, 1981). Attitudes inculcated during their professional training, and a predilection for market solutions, may also contribute to the neglect, by economists, of critical institutional and managerial aspects of policy decision making, and especially to their customary disregard of implementation-related issues. Think-tanks for policy analysis and advice, whether established at the presidential, ministerial, or agency level, are usually staffed entirely with economists. Private consultants advising on policy matters, too, are most often economists. Experience suggests that such teams should also normally include political scientists and management experts. Economic Policy Change and Government Processes 13 Guidelines for Policy Advisers There is a widespread feeling that policy economists should act as protagonists for efficient (usually implying market-based) solutions to policy problems (Schultz 1968, Arrow 1974). Opponents of this viewpoint argue that it is precisely because many economists have become ideological combatants that their analysis has limited practical relevance. Some useful guidelines for policy advisers have been given in Leman and Nelson (1981), which points out that "the economist often seems better at asking the right questions than at finding the answers", and that "economic theory is concerned with the results of rational choice but has little to say about the processes through which rational decisions can be reached." The following "commandments" are proposed: * be economical in the use of economics; * discount for political demand; * dare to be "quick and dirty"; * think like a manager; * analyze equity as well as efficiency; * know your market (whom are you trying to influence?); * pay your organizational dues (have your minister's or agency's interests at heart); e profit from action-forcing events; a do not oversell economic analysis; and i learn policy economics by practicing it. The art of successful policy advice clearly lies in the accurate assessment of the "bounded rationality" of the actual decision makers; it also involves packaging recommendations so as to meet the decision makers' level of technical expertise and their political or other concerns. Some distinguished commentators have argued strongly that since "scientific" policy prescription is largely bogus, it is important to "blend technical expertise with ordinary knowledge within the reach of most informed citizens, and with the political agenda of the interests likely to be affected" (see for example Lindblom, 1974). There is thus a role in policy making for intuitions, hunches, group loyalties, and, above all, sensitivity born from experience. These are "checkpoints" for theory-based analysis in the same way as objectivity is essential to uncover the weaknesses in political and rhetorical approaches that appear to sweep away practical problems and difficulties. The essential point is that analysis must lead to advice that is both technically and politically feasible. Economic policy advice must respond to the needs of politicians. First, demands by their constituencies mean that politicians try to appear to be taking action to solve problems. Second, it is essential to help politicians satisfy efficiently the income redistribution demands they alone are empowered to address. Therefore, when a politician proposes an action ill-advised on efficiency grounds, good economic and political staff should do more than merely raise objections; they should also suggest better ways of achieving the politician's ultimate ends. For example, when changing income distribution is the desired goal, good economic advice would search for ways of redistributing the desired amount of output to the recipient group with the minimum possible loss to everyone else. A case of bad policy advice mentioned in the discussion at the Seoul seminar was to end price controls on basic foodstuffs without proposing alternative ways of maintaining the same real income of low-income people. Emphasis was also placed on the assertion that policy analysis and recommendations must be clear, concise in the statement of findings and implications, and readily disseminated, so that politicians can grasp its essence quickly and recount it in turn to their constituents. Moreover, it was argued, advisers should lobby policy makers. Beyond simply stating the case for or against a proposal, economists need to make themselves available to explain the issues and to respond to the concerns of politicians. Suggested methods ranged from the 14 Edgardo Boeninger technical advice to politicians cited in Indonesia to widespread intragovernmental consultations in Korea. If analysis is consistently to play a significant role in decision making, its status must be enhanced, so as to ensure high quality staff and effective influence on policy makers and other relevant social and political actors. One prerequisite for status is sustained demand for analysis coming from top policy makers. If the latter are not interested, objective discussion of issues and options will not take root as an indispensable management practice. Organizing Analysis: Loyalty, Independence, and the Role of Think-tanks How is policy analysis best organized? The case studies illustrate a wide range of arrangements and the literature proffers well-argued but somewhat contradictory propositions. It is asserted, for example, that the locus of analysis can either enhance or kill the chances for policy reform, and that analysts whose work is marginal to the interests of the ministries most centrally concerned with the policy will tend to be ignored (Wildavsky, 1979). But it is also convincingly argued that there is a need for independent or autonomous policy thinking to challenge the normal presumptions of the interested agencies. Many commentators are in favor of task forces or commissions with a mixture of outsiders to ensure a broader view of policy problems. At the same time, there is widespread concern that temporary bodies of this kind will be ignored by the bureaucracy, and that such bodies will tend to reflect the interests of well-organized groups, but ignore those of others, such as small business, rural people, or the poor. Combining the prevailing perceptions of seminar participants and theoretical contributions, the following criteria and guidelines reflect a fairly wide consensus. • Some kind of independent policy voice is valuable for a government, especially in countries which lack well-informed policy debate within the legislature or in the media. Such an independent capacity can also help make the government and the bureaucracy more accountable for policy decisions. e Policy makers must perceive of think-tanks and advisers as loyal supporters of existing authorities. Though by the very nature of their functions analysts should be free from administrative routines and bureaucratic restrictions, think-tanks must have direct access to decision makers and share their concerns if advice is to enjoy the credibility required for relevance. * Analysis and policy advice will be more effective if think-tanks and key advisers are closely linked to the dominant policy-making institution or individual. Whether a think-tank is best located in an independent institution closely linked to government and with a strong policy orientation, or whether it should be within government itself but located somewhat independently from the normal bureaucratic machine, for example in the office of the head of government, depends on country circumstances. An advantage claimed for the latter arrangement (e.g., by Plowden, reflecting on British experience) is that policy analysis is then likely to deal better with the institutional and procedural realities of implementation, rather than concentrating solely on the ex ante analysis of policy choice. * Although not strictly part of the policy-making process, critical analysis has a important role to play. Most often located in universities, and rightly so, voices of dissenting analysts are fed into the process as inputs for consideration by "loyal" think-tanks. * Ad hoc task forces may be the most effective way to respond to special needs. Such temporary bodies, however, should be directly related to and report to the dominant decision-making body, so as to keep their recommendations in line with existing long-term objectives and predominant public philosophy. * The largely short-term concerns of central banks and budget bureaus should be adequately balanced against the longer-term perspective of planning commissions or development Economic Policy Change and Government Processes 15 corporations. Face to face discussions attended by key policy makers or reliance on think- tanks located at the highest level of government are arrangements that may be set up for this purpose. * The best policy options arise in intellectually competitive environments. As their numbers increase, economists and other analysts should be dispersed more widely, to encourage competition among different sources of policy advice. As the analysis function develops further, all key ministries and institutions will most likely rely on their own think- tanks. Korea has used several of these approaches to organizing for analysis. Emphasizing the key role of the presidency, the case writer indicates that "a major change in government economic policy toward a market-oriented economy in the early 1980s was strongly advocated by the chief economic secretary to the President, who rose to be a major figure in the economic policy process." At the same time, "Most ministries increase their bureaucratic power by having their own think-tanks, preparing discussion papers that often reflect significant differences in approach but are relevant inputs of the decision making process... Decisions with respect to trade liberalization were made after intense intragovernmental debate, in which each of the contenders relied on a think-tank advocating their own preferred policy options." The influence of analysis and economic thinking in Korea is further enhanced by the fact that "in comparison with other countries, more academicians have been appointed to major posts in economic ministries." Other countries present a range of experience; Thailand has recently established an independent government think-tank with considerable prestige, whereas in Pakistan, Sri Lanka, and Bangladesh economic institutes with academic links have only had a limited influence on civil services, which have a rather inward looking tradition of self-sufficient expertise. Finally, it is worth asking how governments can most effectively use outside expertise from the national and international consultancy industries, and from international institutions, including the World Bank. Are such agencies merely a useful buttress for policy proposals and bureaucratic interests already staked out within the domestic arena, or can they be managed by national decision makers as a source of innovation and expertise (Grindle et al., 1985)? In some cases, outside expertise is mainly used to add the weight of international prestige to domestically favored policy options. External support for privatizing state enterprise in Bangladesh appears to have played such a role. On the other hand, in Indonesia foreign assistance for the study of tax reform was indispensable. In still other situations, the presence of international consultants may simply be an expression of pressure for policy reform by international agencies or foreign states. Organization for Decision Making On policy making as a governmental process, two points should be made at the outset. First, it is clear from the literature, as well as from the case studies and the discussions, that the synoptic, rationalist, view of policy making is rather unrealistic, and that the neat distinctions embodied in that view-means versus ends, political decision versus bureaucratic execution, and a neat sequence of information-decision-implementation-evaluation-do not correspond to how policies are actually made and implemented. At every stage of the policy making process there is a tension between authority and consultation, that is to say between policy making as a highly controlled top-down process and as a much messier matter of political arbitration and compromise. Policy making, particularly on economic issues, is thus inevitably plagued by ambiguity, indecision, and contradictions over time. Second, economic policy decisions are of many types; they vary in complexity, and decision- making styles are heavily influenced by the traditions, political context, and governmental system in which they are situated. 16 Edgardo Boeninger As a result of these features it is difficult to generalize about how to organize for decision making. Where economic thinking is homogeneous, and political and social goals are widely shared and clearly spelled out, consultation relatively easily leads to broad agreements on goals and preferred policy instruments. Authority, in turn, will meet with little opposition in such a case. Even then, however, one should not conclude that the bureaucracy will have a free hand and will shape a strictly hierarchical style of policy formulation. Reality is more complex, as shown in several of the case studies. The pressures of society express themselves through subtle informal channels, personal relationships, or at key points in the administration. Cultural Influences on Decision Making In particular circumstances the actual mix of consultation and authority, control and compromise, governmental direction or responsiveness, is heavily influenced by tradition, bureaucratic culture, the nature of the political regime, and the degree of social cohesion or divisiveness. The Korean case writer, for example, points out that "the predominance of government bureaucrats in the formulation of major policies is ... attributable to the historical and cultural background of Korean Confucianism and its authoritarian political system. [Korean] national culture is also distinguished by face-to-face personal relationships. This fact has allowed lower-level technocrats a good deal of participation and influence within the framework of a hierarchical society. ... Key institutions of the Korean Administration have developed along the lines of the Japanese model of management and public service." In Thailand, the case writer points out, the prevalent policy style is grounded in Thai culture and tradition, favoring bargaining and compromise and allowing for a large degree of bureaucratic freedom. He contends that "written mostly by bureaucrats who do not like to be tied down to a concrete plan, Thai laws are generally vague about such matters [as the aims of the Farmers' Aid Fund]." The nature of the political regime, of course, heavily influences the decision making process. According to the Korea case writer, "given the authoritarian nature of the government, the role of the legislature was kept to a minimum. This was considered necessary to achieve a speedy decision-making process and flexible implementation of economic policies." And in the Bangladesh case, "Martial law obviated the need for detailed interministerial or political debate." In some cases, the approval of the legislature is a mere formality because power is heavily concentrated in the executive. Where this is so, the influence of political parties and social organizations tends to be limited; policy making then becomes an essentially intrabureaucratic process, though a good deal of consultation may take place within the "walls" of the state apparatus, not barring lobbying by interest groups through a network of social acquaintances and personal friendships. Organizing for Complexity First, as mentioned above, organizing for decision making means organizing to establish a policy framework, solve conflicts, attain goals, and manage crises. There is no such thing as separate or parallel organizational solutions to accommodate these diverse purposes. Whatever form of organization is chosen must cope with the fact that all of them will, in varying degrees, feature in any specific policy-making process. One way of handling this complexity, adopted by many governments, is to have a whole range of more or less routine policy formulation and decision making bodies adapted to making policy in particular areas, such as the exchange rate, credit and monetary policy, fiscal matters, and government-wide policies toward key productive sectors. As both the comparative literature and some of the cases indicate, however, this arrangement can have several disadvantages. First, the regular committees or coordinating bodies will often tend to become preoccupied with bargaining and decision making over Economic Policy Change and Government Processes 17 immediate issues in their policy areas-for example next month's discount rate, or the sales tax level for the next fiscal year-and give inadequate weight to longer-term policy issues. Second, such standing specialized bodies can become routinized and parochial in their interests, concerned with settling departmental conflicts and not disturbing the bureaucratic balance too greatly. Third, there is the danger that economic policy in general will then lack clear overall direction and internal coherence among its different components. Some writers (e.g., Lindblom, 1965) hold that "muddling through," coalition building, and compromise are the best that can be expected. But such views tend to be most strongly argued by American commentators, who are clearly faced with an exceptionally pluralistic policy- making process. This brings us to the reactive, incrementalist second-best end of the policy- making spectrum referred to earlier. Rice policy in Thailand, the later stages of the food subsidy policy reform in Sri Lanka, and trade liberalization in Korea before 1979 seem to be close to this kind of policy scenario. The alternative view, that economic policy making can be made considerably more disciplined and effective, is worth exploring further. Up to perhaps a decade ago, conventional wisdom had it that disciplined and effective policy making could only be achieved through economic planning, in which the planning commission or ministry provided a long-term macroeconomic perspective, integrated the different components of economic policy, and was somewhat removed from parochial and bureaucratic interests. Nowadays, however, there is a widespread perception that the need for coherence, consideration of long-term issues, and guidance of policy will not necessarily be met by planning agencies. In some of the cases reviewed for the seminar, decisiveness about policy went hand in hand with a predominantly market-oriented economic ideology. Long-term considerations and explicitly defined purpose are clearly visible in the more recent Korean trade liberalization efforts, in the privatization of state enterprises in Bangladesh, in the design of tax reform in Indonesia, and the delinking of the exchange rate in Pakistan. In none of these cases did institutionalized economic planning play a significant role. Indeed, the Korea case writer indicates that "as soon as the five year plans started, few people, including the Economic Planning Board (EPB) bureaucrats who had prepared them, remembered the major contents of the plans." However, EPB did provide general direction for government economic policies, a function very much in line with the role ascribed to planning in this chapter. There may be a range of explanations for the decisiveness of policy in these instances: the political preeminence of a particular policy agency (the EPB in Korea); political trust reposed in a few individuals (Pakistan); good preparation before taking office (Sri Lanka); and anticipatory planning (Indonesia). In general, how can the longer-term perspective and integrating function that planning is supposed to perform be introduced into government's policy making machinery, if the institutional solution of planning agencies is seen to be ineffectual in most circumstances? Two possible approaches are as follows: * ensuring a balance of influence on decisions between ministries and agencies concerned with long-range objectives and those concerned with short-term issues; * emphasizing "policy packages," while acknowledging the fact that to face certain crucial issues, some key policies must be given priority and that in almost any package, some specific policy will stand out as its most significant component. Far removed from the rather bureaucratic vision of conventional planning, seminar participants stressed the effectiveness of "policy enclaves" created at certain points of the bureaucracy. These are floating or informal groups that take responsibility for policy analysis, and for pushing through the conclusions. Individual and Institutional Leadership LIMITATIONS OF THE CABINET. One way to improve policy performance using institutional or organizational mechanisms may be to strengthen the policy decision-making functions of the 18 Edgardo Boeninger cabinet. Here, though, the literature is almost universally skeptical. In many developing countries, commentators have pointed to the largely ritual or symbolic function of cabinet meetings, their preoccupation with very immediate political issues, the limited ability of cabinet ministers, on the whole, and the short shrift given to policy discussions. In the case of industrial countries, a great deal has been written in recent years about the decline of cabinets relative to the role of the head of government or head of state-the so- called imperial presidency in the United States, the resilience of the French president's office even in the face of a hostile prime minister and cabinet, and the flow of power away from the British cabinet to the prime minister (Hall, 1983). At the same time, cabinets are often seen as merely playing out at the political level the bureaucratic or ministerial rivalries which have not been resolved by lower-level officials. Ministers in fact tend to become spokesmen for the interests of subordinate agencies as well as for their own ministries' clientele or for interest groups related to the ministries' area of concern. Ministers face pressures from managers and directors of agencies or state enterprises demanding large budgets, increased status, and additional functions. Experience shows that institutions consistently attempt to redefine and expand their duties, a process that often becomes an end in itself, thus giving a peculiar meaning to the concept of institutional loyalty to which top executives appear to conform while pursuing their own bureaucratic or political careers. It is no wonder that cabinet meetings tend to become either a series of bilateral exchanges between the head of state and individual ministers or a continuing effort by the former to mediate in interministerial conflict. Only rarely is effective policy coordination practiced in such a body. The Thailand case study highlights these limitations when reporting on the roles played by the Ministries of Finance, Commerce, and Agriculture as to rice policy. In this process of competition, bargaining, and compromise, each ministry at various times controlled specific policy areas and resources. At a later stage a Farmers' Aid Committee dominated by bureaucrats was established, apparently as a result of a bureaucratic compromise. The Cabinet as such did not play a coordinating role. In the Pakistan case, "power was centralized with the President who took all the important decisions. The President had complete confidence in his Finance Minister and treated him like his Prime Minister. Any decision having the support of both did not require an elaborate process of consensus building among other ministers or need to be approved by the Cabinet." Likewise, the Korean case writer indicates that "one of the most important reasons for effective implementation of policies is the dominant role of the Economic Planning Board," adding that "under a system in which all ministries have equal power, decision making as well as implementation of policies can face difficulties when these ministries have conflicts of interest on specific issues." In Korea the Economic Ministers' Roundtable held informal meetings at which major policy issues were freely discussed. The Economic Deputy Ministers' meeting also played a similar role. In this way, policy proposals can be informally approved, after discussion involving even lower-level technocrats, before being officially formulated by the empowered ministry. DOMINANT DECISION MAKERS. Given the limitations of both planning and the cabinet system as mechanisms for decision making, the Korea and Pakistan cases throw crucial light on the leading role of a dominant minister and/or of a head of government in effective decision making. Not only should the national leadership be strongly committed to the economic policy changes under consideration, though that is very important; its close association with a broad economic program clearly seems to confer additional legitimacy on particular policy decisions. The association of leader with program (as with Thatcherism, Reaganomics, and perestroika under Gorbachev) helps to create a policy agenda and a "policy space" within which officials can work (Nelson, 1985, Grindle et al., 1985). It may also serve as a warning (although this is less clear in many instances) that the "business as usual" of vested interests and bureaucratic parochialism will not be permitted to continue. If the role of the head of government is politically and bureaucratically vital to successful economic policy change, he or Economic Policy Change and Government Processes 19 she needs to establish a particularly close relationship with the senior reform minister. For broad economic policy issues this role is normally performed by the minister of finance, who usually enjoys the advantage of holding the purse strings. In Korea, the protracted struggle between protectionist and pro-liberalization technocrats and ministers was resolved in favor of trade liberalization because of the President's strong and continuing support for liberalization. And "EPB's role became stronger when the Budget Bureau was transferred from the Ministry of Finance to EPB in 1961, and the Minister of EPB concurrently holds the position of Deputy Prime Minister." No rule, of course, applies in all cases. First, a head of government, no matter how staunchly he or she supports certain policies, may not be able to ensure their success or continuity in the face of political opposition or administrative shortcomings. Second, as in Pakistan the decisive factor in the decision-making process may be the dominant personality of an individual who was neither minister nor head of government. Whichever model a country follows for decision making at the top level of government, smooth relations and effective coordination between agencies, ministries, and presidential staff are needed. The French "elite corps" and the British civil service, both of them models followed in parts of the developing world, achieve this in different ways. In the French system, close personal ties, shared values, personal relations that often go back to their student days, and a broad agreement on fundamentals have encouraged the development of a class of policy cadres. These public servants are loyal to the individual minister they work for, and are mostly free from the vested-interest syndrome often developed by officials who remain with a specific ministry for most of their careers. The British system has relied on the administrative capability and continuing loyalty of the civil service to whatever government or minister may hold office, although greater mobility at higher levels is developing. The permanent secretary of each ministry plays a key role, balancing political and departmental priorities. Improving Coordination Drawing on the seminar discussions as well as on an earlier paper by this author, various devices can be recommended to improve organization for policy-making purposes: * Establishing ad hoc committees that can take in a wide range of views and make proposals directly to the head of government or dominant minister, though to prevent such committees from becoming excessively politicized, they should not be placed in the center of public attention. * Creating task forces to deal with specific problems that exceed the research or operational capabilities of a given agency. These task forces might include foreign experts or be established by contract with private consultants. Such a procedure may often be appropriate, for instance, in designing export promotion policies, tax reform, or the reduction of a fiscal deficit. When politically sensitive issues are at stake, such a task force would have to be chaired by the dominant minister for economic affairs or report directly to him/her. * Appoint technical advisers and/or political secretaries to the head of government or dominant minister. Such figures played useful roles in the Korea and Thailand cases. The risk here, however, is of developing a dual circuit, because such individuals are not accountable to the bureaucracy. * Concentrating duties and responsibilities in specific areas within a single, largely autonomous, agency, so that formal mechanisms for interagency coordination would only be required for overall policy and for decisions involving the most complex issues. Authority would be delegated to a small number of key institutions, such as, say, the central bank, bureau of the budget, and the ministry of agriculture. 20 Edgardo Boeninger * Confining membership in whatever committees or councils are established to a stable group of individuals associated with certain institutions and posts. Each such committee should be assisted by a technical secretariat. Whenever possible, this function should be entrusted to the research department or economic analysis division most directly linked to the matters directly dealt with by the committee. • Improving the interaction among political leaders, senior civil servants, and interest groups in various informal ways. * Using foreign experts alongside the bureaucracy. The team that designed the Indonesian tax reform, for example, consisted of local legal experts and tax specialists and experts from the Harvard Institute for International Development. In such an arrangement, it is essential for local counterparts to participate on a equal footing; policy change will only be accepted by the bureaucracy if it is seen as coming from within its own ranks. For this reason, private consultants or foreign experts should never be left to work by themselves nor impose their output on the administration. When this happens, resistance to change multiplies, often leading to policy breakdown. Both the case studies and the literature indicate that even the best of arrangements for organizing decision making will be under constant pressure and risk of failure. The "windows of opportunity" for technocrats and for rational and/or consensual policy making tend to close rather quickly. Other political values and interests assert themselves, and the constituency behind any policy erodes over time. Moreover, the experience of implementation, to which we turn in what follows, is often one of the recurrence of old policy problems in a new guise (as in Thailand or Sri Lanka), and of reform or policy change as an iterative and continuous process. To sum up, the effective organization of policy decision making would seem to depend less on complex institutional arrangements than on quick and sound analytical capability; spirited leadership of the policy reform team; a degree of flexibility and even of opportunism; good access to the top; and flexibility in organizational arrangements to adapt to changing circumstances. Policy Implementation Problems in implementation are at least as important a cause of policy paralysis or reversal as a sudden economic or financial crisis or a change in government. In Korea and Pakistan early moves toward trade liberalization gave way to the reintroduction of import controls for a substantial period before more liberal policies were reasserted. In Sri Lanka and Thailand the policy reforms-reduction of food subsidies and establishment of a Farmers' Aid Fund-were never fully in control and have finally been abandoned or lost much of their significance. Getting things done is in the first place a managerial problem, involving the enactment of laws or executive orders to adapt the legal framework where necessary, defining objectives, goals, and procedures, assigning responsibilities and specific tasks to various segments of the bureaucracy, and allocating scarce resources. But successful implementation also requires strong and lasting political support, even in the most hierarchical of environments. In short, implementation of policy is at once an administrative and political process, which itself continuously raises policy issues afresh (Grindle, 1985). The Bangladesh case study, for example, indicates that the basic decision to nationalize banks, jute mills, and textile mills, as well as to divest abandoned industrial units, was taken at the highest level of government. Several cabinet committees and task forces were entrusted with implementing these decisions. Owners had to be identified, assets and liabilities computed, terms negotiated, shares floated, and agreements signed. Had serious conflicts arisen with regard to any of these matters, policy implementation would have been delayed and might have broken down altogether. Planning for implementation therefore involves designing a political strategy as well as assessing-and sometimes planning to create-the bureaucratic capability needed to prevent or Economic Policy Change and Government Processes 21 overcome eventual administrative bottlenecks. The relative intensity or relevance of these factors may vary, but neither of them should ever be ignored. An important distinction can be drawn between policy instruments which demand a great deal of administrative regulation and action, and those which are more indirect, more "stroke of the pen," or which deploy market-based rather than administrative methods. With the latter type, management considerations and administrative capabilities do not play a role in implementation-instead, the basic requirement is acceptance by key interest groups such as the business community, and the avoidance of undesirable effects, say inflation, that might produce social unrest. For policy reforms requiring detailed administrative arrangements and a whole series of interventions, the formulation and implementation phases in practice tend to merge. It becomes critically important for policy planning to take account of the implementation requirements of policies that are management-intensive, both because bureaucratic resources are limited and because of the added scope for repeated and sometimes corrupt political interventions in the implementation process. At the margin, there may often be a choice, for the same policy objective, between these different types of instrument (Lamb, 1987). Especially for countries with severe personnel and administrative weaknesses, simplification and economizing on scarce resources is essential, and choice of the "right" policy instrument consequently more significant. Implementing Management-intensive Policies Rather than two separate stages or processes, policy design and implementation of management-intensive policies are really parts of a continuous process in which decisions have to be taken throughout. Seminar participants agreed that looking at policy making in this way helps to emphasize how easy it may be for opponents to block or distort a reform as originally conceived. It also reveals the complexities of the managerial side of policy making. Planning for the implementation of a policy that is management-intensive must focus on such issues as bureaucratic attitudes and motivations; the technical competence of staff, and opportunities for training; public sector salaries and incentives; institutional strength; the simplicity or complexity of necessary administrative action; benefits and costs of decentralization; modernization of bureaucratic procedures; eventual loopholes in legislation or in specific activities favoring nonbeneficiaries; enforcement; compliance; control; audit; sanctions; and evaluation of progress. ROLEOFCIVILSERVICE. No other variable plays a more decisive role in determnining the quality of public sector management or, more broadly, the operational capacity of the state in economic and social affairs, than the standards of its civil service. First, the educational background and professional level of personnel in top managerial positions, or doing analytical work, is particularly important, though so too are the standards maintained in other functions performed by civil servants. Second, bureaucratic loyalty and willingness to cooperate must be stressed. As mentioned earlier, the latter depends quite heavily on the extent to which the civil service perceives change as coming from within its own ranks. In Indonesia, the tax reform laws were enacted long before administrative capacity was sufficiently developed to carry out the reform. As well as personnel bottlenecks, the computerized tax information system, including the identification of taxpayers by number, which should have started operating in 1984 as part of the new tax reform, has not yet been fully put into place. Such a delay in effective implementation endangers the tax reform itself by encouraging noncompliance on the part of taxpayers. Where the bureaucracy is relatively weak, an objective assessment of eventual administrative bottlenecks and shortcomings becomes an essential part of policy implementation. Simplification of content and necessary procedures is usually high on the list of priorities if the administrative burden associated with a policy is to be lightened, but training will often be an inescapable part of implementation strategy. When massive training 22 Edgardo Boeninger is needed, as in the Indonesian tax reform, the main difficulty arises from the fact that this takes time, because as the first stage it is necessary to train the trainers. The Sri Lanka case emphasizes the quality and spirit of the civil service as decisive. The 1977 rice ration book revalidation involved the removal from the rationing scheme of nearly eight million people. "At this time the Ministry and its allied Departments possessed a core of trained and experienced administrators who had both the confidence and the capacity to carry out a policy change of this nature. All were highly trained and had wide experience working in numerous fields closely related to the policy reform." The implementation process from the issuing of the forms to the completion of the revalidating of ration books took only one month. Administrative capacity is freed when, as in the Indonesia or Sri Lanka case, procedures move to self-assessment and self-declaration (of tax liability or incomes). Audit then replaces overall case-by-case control; such techniques, however, must have been previously mastered by the bureaucracy, through training of staff and appropriate legal and procedural adaptations. INTRABUREAUCRATICCOMPETITION. Policy reform may entail organizational change. Political authorities are easily tempted to create new institutions, because this often seems less contentious than altering the functions of existing ministries or departments, but experience shows that this should be avoided if at all possible. It normally leads to an increase in the number of public employees, to administrative duplication, and to further difficulties in coordination. In Thailand, where there was intrabureaucratic competition and functional overlapping, the case writer argues that rice policy was hampered by failure to build up one institution with sufficient operating capacity and professional ability to administer the policy; neither the Public Warehouse Organization or the Marketing Organization for Farmers developed sufficiently. The government did seek to make rice policy clearer and more consistent by institutionalizing the bargaining process, and thus created the Farmers' Aid Committee. But the Committee "failed to acquire technical capabilities of its own," and "no clear procedures and no coherent way of deciding on resource allocation was established." This allowed the approval of questionable proposals that would not have passed the test of examination by the Bureau of the Budget. It is inherently difficult to achieve coordination when a large number of committees and advisory bodies are involved in a particular policy. If coordinating committees are to operate permanently, continuity of membership by specific individuals as well as a stable technical secretariat are essential requirements. Issues in Policy Enforcement Effective enforcement of a policy will always require a substantial degree of voluntary compliance. People are more likely to comply if the public generally disapproves of violators and if they believe that sanctions will be actually imposed on them. Generally, enforcement and compliance require the clear identification of target groups, avoidance of loopholes in legislation or procedures, adequate control of operations, and a system of credible penalties. Loopholes should receive attention; they may produce outright evasion-for example, of taxes-or they may provide ways for groups not intended to be beneficiaries to obtain access to a program (as, for example, when the middle class commandeers most of the benefits from subsidized housing programs). The Indonesian case exemplifies the point that enforcement may require a change in social attitudes. In Indonesia the long-standing reliance on oil revenue, and the perception of being oil-rich, had led to slack enforcement of taxes. As the case writer notes, "continuous efforts are needed to instill the conviction that paying taxes is a duty inseparable from the democratic right of every citizen to participate in the development of the nation. ... Despite the great improvements [including an impressive public information campaign] to date, DGT still faces problems, the most difficult of which is noncompliance." Economic Policy Change and Government Processes 23 Political Issues in Implementation RESOURCE ALLOCATION. There is no simple answer to the combination of acute need and extreme scarcity of resources in poor countries. But perhaps the single most decisive influence on the success or failure of policy is the allocation of resources in relation to program requirements. Governments hard-pressed by objective needs and social demands are frequently pushed into launching policies and programs with no clear awareness of financial needs and constraints, either deliberately ignoring them or naively relying on budget reallocation. In the Thailand case, for example, the fact that the government never allocated enough resources to withdraw rice into storage for a long enough time was an essential weakness of the rice price support program. Budgetary allocations and budget execution apparently belong to the realm of management. In essence, however, the fight for a larger portion of limited resources is political. A decisive element in the design of a strategy for policy implementation, which must be addressed early in the policy process, is the need to find a way of effectively dealing with the political dimension of resource constraints. A striking response to resource scarcity is Sri Lanka's treatment of food stamps. When food stamps were introduced in the hope of reducing expenditures on food subsidies, no significant reduction of subsidies was actually achieved. In the next few years, however, without any overt political decision, the cost of the subsidies fell from 6 percent to 2 percent of GNP and from 15 percent to 5 percent of government expenditures; the government simply did not adjust the real value of the foodstamps to keep pace with inflation. DECENTRALIZATION. From a strictly administrative point of view, if a program is to be applied for a protracted period over a large area, decentralization contributes to smooth operation, and it becomes inevitable when such programs reach out to large segments of population. From a political standpoint, however, a word of caution is in order. If policy objectives, action programs, target groups, criteria for allocating resources, and rules for identifying eligible individuals are not clearly spelled out when implementation is decentralized, power flows down to lower-level officials, who become an easy prey for interest groups and powerful landlords. Dealing with the Public: Policy Packaging Policy needs to be disseminated clearly and rapidly as soon as it has been decided on, so that politicians, bureaucrats, and particularly the media can grasp it quickly and recount it in turn to their constituents and to the public at large. Timely reports giving available information; analysis of origins and expected effects; and intrabureaucratic debate may all help to build support for policy reform. To prepare the public, policy changes need to be packaged so that those who are to benefit can see this clearly, and so that those who are to lose can either be compensated or have their objections neutralized. The policy package must persuade, coerce, safeguard, compensate or coopt the various groups to be affected. Trade liberalization in Korea, for example, proceeded according to preannounced schedules of liberalization for individual groups of commodities, so as to give the affected firms time to adjust. In Sri Lanka, the government newly elected in 1977 introduced a drastic scaling down of food subsidies, but as part of a wide-ranging economic package to revive the stagnant economy. The package included compensatory measures for people no longer eligible for subsidized food, as is noted below. Meanwhile, the government aroused enthusiasm about national development prospects by stepping up progress on major infrastructure projects (including the Mahaweli hydropower and irrigation project, which was accelerated from 30 years to only 7), and promised to build 100,000 houses within 3-4 years. Simultaneously, it embarked on a sustained 24 Edgardo Boeninger campaign to inform people of the incompatibility of funding such development activities while continuing large-scale subsidies. As the case writer notes, "these undertakings were very helpful in influencing public opinion in favor of the Government's policies." By contrast, in 1985 the government's public announcement of a proposed reduction in coverage and closer targeting of the food stamp scheme maximized opposition to the policy change. The revalued food stamps became a coveted prize for noneligible individuals. "An income survey revealed that 2 million people should be removed from the food stamps scheme. However, members of parliament began to face extensive pressure from constituents who stood to lose their food stamps. They protested to the President and built up pressures for maintaining the status quo that could not be defused." Under these circumstances effective implementation of the proposed policy was almost impossible. Wide coverage by the mass media can play a crucial role in winning initial support for a policy change and for the subsequent sustainability of reform. Policies that keep meeting hostile press criticism tend to lose momentum, as the will and faith of politicians and decision makers weakens. Use of the media requires a careful packaging strategy to highlight the positive aspects of reform. Such a strategy will probably include deliberate appeals to support- gaining symbols and emotions as well as publicity designed to forestall negative reactions. (If the media are dominated by the government, however, the credibility of such a public relations program may suffer.) As was noted earlier, the protagonists in the Pakistan case successfully used a deliberate packaging strategy, successfully presenting the delinking of the currency as "an assertion of national independence in economic policy formulation." Though the main concern behind the delinking was to protect the competitiveness of Pakistan's exports, the case writer notes that "To minimize the negative political impact of the exchange rate adjustment, it was presented as an assertion of national independence in economic policy formulation.... The timing of the decision and its projection by the government were very skillfully managed. As a result, it won the support of the business community and met little or no opposition from the public at large." In many countries the packaging of policy changes is developed to a high art, combining broad public relations with highly group-specific campaigns, briefings, cultivation of particular sections of the press, and so forth. The sophistication of such efforts partly depends, of course, on the openness of public and press debate and on journalistic traditions even where the press is well developed. For example, there may be a much better chance of informed and serious journalistic discussion on economic policy issues in Pakistan than in Thailand, largely because of the character of the newspapers themselves. Compensatory Measures Compensatory measures and anticipation of the need for compromise are very important elements in successful implementation. As noted earlier in this chapter, policy as first formulated and presented should be capable of a certain degree of adjustment in response to social demands and interest group pressures. But it should be stressed that bargaining and compromise are counterproductive if even relatively minor adaptations lead to undesired distortions and prevent objectives from being fu.lfilled. The success of policy proposals, leaving no room for negotiation, depends heavily on the government's effective wielding of power and on its political will. Korea has very effectively used compensatory measures as a key policy instrument. As stated in the case study, firms likely to be affected by trade liberalization were classified into two groups: one requiring complementary adjustments to regain competitiveness; the other judged unlikely to be competitive once trade had been liberalized. For the first group, "plans [were] designed to assist these firms during their transitional period, covering scale of production, quality of product and technological change." For firms deemed incapable of recovering, "the Government [designed] plans to limit new entrance into industry, encourage these firms to move into other, more promising, areas of business, and assist in disposing of Economic Policy Change and Government Processes 25 existing plant and designing industry restructuring plans." "For industries deemed to need complementary measures the Government [designed] plans to assist them in the transition period. These plans are concerned, for example, with the appropriate scale of production, quality of product, possible improvements in technology. Subsides and preferential interest rates on loans and subsidies have become much less significant than before but preferential taxes remain a major means of assistance. According to the Tax Reduction and Exemption Law, firms classified as requiring rationalization programs to improve their competitiveness will receive preferential tax treatment." In Pakistan, the exchange rate adjustment process created some uncertainty with regard to payments for imports and exports. In response, the government introduced a number of measures to cover the foreign exchange risk; at the same time, to alleviate pressures on the budget, it deregulated the prices of a number of key commodities. When Sri Lanka first attempted to reduce food subsidies, the government introduced compensatory measures at the same time. To afford relief to groups no longer eligible for subsidized food, the government introduced subsidies for sugar for children under 12 years old, milk food for infants, a fertilizer subsidy of 75 percent (up from 50 percent), to producers of paddy, a 25 percent wage increase, a dole for unemployed persons, and a higher guaranteed price of paddy. Evaluation of Performance The complexity of policy implementation, as well as the inevitable changes in context created by unforeseen circumstances, make it advisable continuously to evaluate progress and feed back the findings to analysts and policy makers. Administrative tradition emphasizes legal and financial control. Managerial modernization shifts attention to selective audit and evaluation of performance. Policy making should move beyond these routine processes into substantive evaluation of success and failure, systematic examination of managerial bottlenecks, and of the political developments related to support for or opposition to policy. As pointed out in the literature, techniques in this area are still far from perfect. But since large amounts of bureaucratic resources are routinely used in control operations, there is a great deal of room for innovation in this area. Policy making should move beyond administrative reporting into political and managerial assessment of performance, providing new insights as well as timely and systematic feedback to policy analysis and decision making. 2 SRI LANKA: A DECADE OF FOOD POLICY REFORMS, 1977-87 M.D.D. Pieris Sri Lanka, one of the poorest developing countries, has successfully followed a development policy based on meeting human needs. Various measures of welfare show the average standard of living to be much better in Sri Lanka than in many countries with much higher per capita income levels (see Table 2.1). The key element in the social welfare policies pursued since Independence (1948) has been food: essential items of food, particularly rice, were not only made available in every part of the country, but for 30 years have been kept within the reach of the poor through pricing and subsidy policies. Table 2.1 1970-75 Literacy Infant Life Average Age 15 mortality expectancy Physical GNP per and over per at age one quality of Country capita (percent) 1,000 (years) life index Bangladesh 92 22 132 53 35 India 133 34 122 60 43 Pakistan 155 16 121 57 38 Sri Lanka 179 81 45 70 82 Indonesia 203 60 137 55 48 Thailand 318 79 89 63 68 Philippines 342 83 74 62 71 Malaysia 692 53 75 67 66 Taiwan, China 847 85 26 70 86 Brazil 912 66 82 65 68 Average 329 57 72 61 60 In the initial stages, the objective of the rice pricing policy was to stimulate production, which actually tripled during the 25 years since Independence. It would have been impossible to maintain such an incentive price structure without sociopolitical repercussions, were it not for the deliberate consumer subsidy policy that was followed concurrently. But all this also had a cost to the economy. In 1977, on the eve of the policy reforms described here, the budgetary outlays on food subsidies amounted to well over 15 percent of government expenditures or 6 percent of GDP. With population growing fast and economic growth rates declining, a choice had clearly to be made between immediate welfare or economic growth. A key to economic development was to raise the levels of savings and investment. A prerequisite for this was to scale down expenditures on general food subsidies. 26 Sri Lanka: A Decade of Food Policy Reforms, 1977-87 27 This chapter analyzes the conditions that led to the need for an earnest reconsideration of Sri Lanka's food policies. It reviews the events that led up to one of the most significant and far-reaching reforms, the rice ration book revalidation, or removing from the rationing scheme, in 1977, of almost 50 percent of the population. It describes the next major change in policy, the introduction in 1979 of the Food Stamps Scheme, and then traces the implications of these changes for the government budget and their effects on consumption and income levels. In the process the chapter attempts to set these reforms in the context of the broader policy package of which they formed a part, and to show how a system of administration designed to maintain and manage stability adapted itself to be an instrument of major changes. Finally, it describes the most recent policy change, the conversion of the Food Stamps Scheme administered by the Ministry of Food to a poor relief scheme administered by the Ministry of Social Services, and the difficulties experienced at that time in trying to confine the benefits of the scheme to the really needy. Historical Background Food Policies since Independence At Independence, Sri Lanka was importing 60 percent of its food. Agricultural development and the administration of welfare policies through an extensive subsidy scheme formed the central policy package adopted at that time. In the early 1950s, government introduced a guaranteed price for rice, having regard to the high cost of imported rice during the Korean war. Thereafter, the price was not adjusted downward, even though international prices fell. For about fifteen years the price paid to Sri Lanka's paddy farmers was 30-50 percent higher than the international price and by 1966 (when it was adjusted downward to some extent), it was around 75 percent higher. The attractive price to farmers undoubtedly contributed to the rapid increase in production that took place. By the late 1960s, Sri Lanka was 70 percent self-sufficient in rice. Imports had been reduced by half, in spite of an accelerated growth of population. The various forms of input subsidies were estimated at around 25 percent of the cost of cultivation. Food distribution under a rationing scheme was introduced as a wartime measure and continued after the war was over. After the Korean war was over, neither international prices nor shortages gave cause to continue with the rationing scheme. But successive governments decided to continue with the scheme as a welfare measure. The subject soon generated a political momentum which later got out of control. The subsidy on rice, an important item of diet, was considered a vehicle to transfer income to the poor while at the same time ensuring food security and a degree of nutrition to these sections of the population. (See Table 2.2.) Another of the initial reasons for continuing the rice ration appears to have been the need to provide an expanding market for the additional rice production that resulted from the government's policies. Looking back now, it might be surmised that if the consumer price had been allowed to move up, the efforts to expand production might have been seriously impaired. In origin the food rationing scheme was designed to meet the shortages of rice in deficit areas. It gradually came to cover the whole country, for all persons over one year of age. The rationing scheme as well as pricing policies on sales and procurements entailed increasing outlays on subsidies. To contain the growing outlays on the rice consumption subsidy, the government adopted a pricing policy of selling other food items such as wheat flour and sugar at slightly higher prices than their imported costs. This cross-subsidization was possible because the government had the monopoly on imports of rice, wheat flour, and sugar. An unsuccessful attempt was made in 1953 to scale down the rice subsidy; this met with enormous opposition, including large-scale riots in August of that year, and led to the resignation of the Prime Minister. 28 M. D. D. Pieris Table 2.2 Ratio of pro- Guaranteed curement to Ration Market minimum Production Procurement production (1,000 price price Year (1,000 long tons) (percent) long tons) (Rs. pound) (Rs. pound) 1950-54 352 18 5 386 0.40 0.34 1955-59 466 192 41 664 0.37 0.38 1960-64 653 359 55 845 0.36 0.38 1966 638 394 62 926 0.35 0.38 1967 767 185 24 517 0.44 0.45 1968 901 210 23 511 0.49 0.45 1969 920 193 21 530 0.49 0.45 1970 1,081 369 34 630 0.47 0.45 1971 934 455 49 867 0.46 0.45 1972 878 355 40 780 0.47 0.45 1973 878 321 37 675 0.53 0.58 1974 1,072 293 27 550 1.46 1.01 1975 772 159 21 606 1.33 1.06 1976 838 180 21 602 1.19 1.06 By 1965, it was estimated that well over 75 percent of the rice consumed in the country went through the food rationing scheme. In November 1966, when there was an acute shortage of rice in the international market and prices were very high, the ration was halved to 2 pounds a week. However, the governmtnt felt compelled to issue this reduced quantity completely free of charge to all persons over one year of age. That the reduced ration had to be issued free was a measure of the political dynamite that the rice ration had become. The policy change of 1966 was not particularly successful. The price of rice on Sri Lanka's open market rose, initially, by over 25 percent and production readily responded. However, rice consumption gradually began to drop, with some compensating increase in the consumption of wheat flour, and the price of rice gradually came down to its level prior to the policy change. It was then seen that to sustain a higher level of domestic rice production, at least in the country's early stages of economic growth, would require both an expanding market and relatively low consumer prices. Meanwhile, the budgetary cost of the subsidy continued to rise. A new government of socialist orientation took power in May 1970. Its first policy measure was to restore the rice ration to 4 pounds per capita per week. Two pounds were given free and the other two pounds were priced at three times the previous prices. At the same time, the price of wheat flour was fixed higher than in the past. Rice consumption rose immediately- well over half of it being channeled through the rationing scheme. Wheat flour consumption declined. The government's budgetary commitment on the subsidy did not increase, mainly because of the low international prices of rice, sugar, and wheat flour that prevailed in the early 1970s. In addition to rice, the government brought sugar under the rationing scheme for the first time in 1971. Factors Leading to Policy Change Important changes took place beginning in 1973. World prices of cereals and sugar had moved up, as had those of petroleum products. These adverse price movements exacted a terrible toll on the balance of payments. The government did not pass on the entire price increases to consumers, but it had to find additional resources for key imports of diesel fuel and kerosene, Sri Lanka: A Decade of Food Policy Reforms, 1977-87 29 and it had to give additional subsidies for fertilizers. It thus introduced several measures to try to control the escalation of food subsidies. For the first time, an income criterion was introduced, whereby income tax payers could no longer benefit from the subsidies; the price of rice supplied on the ration was increased substantially; and the quantity of rice distributed free was cut by half. Faced with an acute shortage of foreign exchange as well as very high prices on food items, the government began to introduce large-scale restrictions in the internal market, for the first time since World War II. The transport of rice without a permit was banned and paddy was almost compulsorily acquired at prices lower than the import cost. Problems were further compounded by successive droughts which affected the local production of rice. The average rate of growth of paddy production fell for the first time below one percent. Sri Lanka's welfare program would have continued were it not for the hostile economic environment. The increased oil prices meant that oil imports preempted about half of export earnings, and the high prices of cereals and sugar rendered imports at the required level extremely difficult. Preference was given to the imports needed to support the food subsidy program, as well as imports of essential fuel and fertilizers, but the supply of exchange for the import of intermediate goods began to be cut. Economic growth in 1972 to 1977 was only half the rate of the previous decade. Unemployment rose to 25 percent of the labor force. By this time, the choice between welfare and growth had become urgent and compelling. The new government, elected in July 1977 with a massive mandate, began to address the question of a balanced policy of economic growth and social welfare. Before turning, in Section 2.3, to the description of the changes that resulted, we outline some of the effects the food subsidies were having on the economy. Impact on Consumption and Incomes One of the reasons for administering a subsidy was that it was an income transfer to the poorer sections and a way of assuring them of minimum calorie needs. Calculations based on the 1969-70 Socioeconomic Survey showed that for the poorest 10 percent of the population the ration raised total calorie consumption by 115 per person/day. (See Table 2.3.) This tendency declined as one moved up to higher income groups. For the population as a whole, this could be interpreted to mean that rationing by and large was a substitute for commercial purchases. It was also estimated that for each calorie that went to increase the consumption of nutritionally deficient groups, another 13 went to groups who were not nutritionally deficient, or to substitute for commercial purchases. That analysis also clearly revealed, however, that the propensity to buy food under the scheme was much greater for the poorest sections. The food consumption of these groups responded much more to the food subsidy scheme than to other forms of income transfer. For the poorest 10 percent of the population, the value of the rice ration amounted to 16 percent of their real income, as Table 2.3 shows. The scheme also had secondary benefits in rural areas, for incomes that were not spent on rice were used for the consumption of other, local, foods. The effects of the food subsidy on income distribution may have been larger than was often estimated. The rationing scheme, coupled with the high guaranteed price paid for rice, contributed to enhanced production, and consequently raised rural incomes, which largely accrued to low-income groups. It has been estimated that about 30 percent of the total increase in employment in Sri Lanka since 1946 has been generated by the rice economy. The important achievement of the food subsidy policy was the improvement in social welfare. The policy resulted in a increase of the share of income accruing to the poorer 40 percent of the population. About 20 percent of the improvement in income distribution during the period 1953-73 is attributable to subsidies on food. It is estimated that in 1973, those who received below Rs. 200 a month in income received as much as 16 percent from subsidies. (See Figure 2.1.) Table 2.3 .------ - - ----- ....-----------.E--. -------- EstImates of the per capita contutiution of the rice ration to consumption by income percentile, 1969 - 70. ------------------------------------------------- Average Marginal Ration Food propensity propensity value as Sugar Calories consumed RaDon Income expen- to spend to spend on Rabon share of and ---- per day Food --- Elasticities calories --- ..........hMarginal consumption------- contri- percentile tncome diture on food food value income flow tax Cereals Total Expenditure Rice Cereals Total Protein Rice Cereals Total bution (Rs. monh) (Rs. month) (Rs. month) (percent) (caloriesIRs.) (caloriesl day 10 33.08 21.13 0.64 0.58 5.29 16.0 0.97 1,096 2,013 0.91 0.44 0.37 0.36 0.41 333 374 662 115 20 35.64 22.57 0.63 0.55 5.29 14.8 0.97 1,125 2,065 0.86 0.39 0.33 0.33 0.38 283 319 581 101 30 38.52 24.09 0.63 0.51 5.29 13.7 0.97 1,153 2,123 0.81 0.34 0.29 0.30 0.35 236 267 506 88 40 42.01 25.80 0.61 0.47 5.25 12.5 1.07 1,181 2,170 0.77 0.29 0.25 0.27 0.32 191 217 430 74 50 46.45 27.79 0.60 0.43 5.21 11.2 1.19 1,209 2,227 0.72 0.24 0.21 0.24 0.29 146 168 356 61 60 51.88 29.99 0.58 0.39 5.21 10.0 1.19 1,234 2,284 0.67 0.19 0.17 0.21 0.26 106 124 288 49 70 59.28 32.66 0.55 0.34 5.21 8.8 1.19 1,259 2,346 0.61 0.14 0.13 0.18 0.23 67 82 220 38 80 69.60 35.87 0.52 0.29 5.08 7.3 1.39 1,208 2,409 0.56 0.08 0.08 0.15 0.19 34 46 159 26 90 88.58 -0.67 0.46 0.22 4.95 5.6 1.47 1,297 2,486 0.49 0.01 0.03 0.11 0.15 3 11 99 16 Average 51.67 28.95 0.56 0.39 5.20 10.1 1.16 1,196 2,236 0.67 0.24 0.21 0.24 0.29 155 179 323 63 Sri Lanka: A Decade of Food Policy Reforms, 1977-87 31 Figure 2.1 Sri Lanka: Lorenz Curve Showing Distribution of Income for Spending Units Percent of total income 100 90 80 :1 70- Line of equal distribution 60 1973 a 0 50 1953 40- 30 -X 20 10 r '1-963 0 0 10 20 30 40 s0 60 70 80 90 100 Percent of total spending units Source: Reports of the Survey of Ceylon's Consumer Finances 1953,1963, and 1973. Fiscal Aspects In its early years, the burden of the food subsidies did not exceed 3 percent of GDP. In the 1940s and 1950s, the imported cost of rice, flour, and sugar was rather low, and the country had adequate means to support such a subsidy policy for a small population. But the demographic, as well as social and economic, changes and the attendant problems that emerged during the 1960s began to exert considerable pressure on limited resources. The cost of diverting resources for consumption began to be noticed by policy makers. As was noted above, for much of the history of the scheme the government could contain the subsidy on rice by earning surpluses on its sales of sugar and wheat flour; largely because of the low prices of imported cereals and sugar, the net cost of the subsidy on rice remained fairly stable during the period 1950 to 1973. Thereafter, with the rise in the world prices of sugar and wheat, the government's profits on sales of these commodities declined considerably. The net subsidy on rice began to rise rapidly (see Table 2.4). As a share of the government's recurrent expenditure, it climbed from 12 percent in 1959/60 to 23 percent in 1977. By 1975, the net subsidy outlay had risen to 4.6 percent of CDP, well over 20 percent of the government's current expenditures, and around 63 percent of its investment expenditures. The cost of the food subsidy was largely met from deficit financing. As the net subsidy began to rise, substantial pressure on budgetary resources began to surface. By raising the incomes of those who had a higher propensity to consume, rather than save, the rice subsidy generated additional pressure on the balance of payments. Consequent to increased consumption, the overall savings rate began to decline very sharply. The costs mentioned above do not reveal the full burden of the food subsidy program. In the early 1960s, when Sri Lanka began to introduce exchange controls, food imports-particularly rice, flour, and sugar-were given automatic allocations of foreign exchange, while allocations for intermediate goods were cut. This tended to preempt the foreign exchange needed for plantation export crops. Moreover, in order to finance increasing outlays on the subsidy, the plantation sector began to be heavily taxed. This adversely affected investments in, and the long-term future of, this vital sector. Table 2.4 Food Subsidy-1960-1979 (Rs. mn) Food subsidy Food subsidy Food subsidy Budget as a percentage as a percentage as a percentage GDP Government deficit of government's of government's of GDP at market current (surplus) Net Food current capital at market Year prices expenditure (+) subsidy expenditures expenditures prices 1959/60 6,710 1,582 -488 187.0 11.8 61.0 2.7 1960/61 6,874 1,578 -500 246.0 15.5 58.7 3.5 1961/62 6,970 1,626 -501 228.6 14.0 47.5 3.3 1962/63 7,381 1,643 -439 238.0 14.4 59.2 3.2 1963/64 7,793 1,891 -519 369.0 19.5 91.7 4.7 1964/65 8,084 1,893 -520 271.1 14.3 57.5 3.3 1965/66 8,337 1,976 -682 273.0 13.8 51.7 3.2 1966/67 9,037 1,895 -723 201.0 10.6 32.2 2.2 1967/68 10,717 2,186 -850 296.3 13.5 41.5 2.7 1968/69 11,695 2,384 -947 328.8 13.7 38.5 2.8 1969/70 12,782 2,659 -1,150 327.2 12.3 40.2 2.5 1970/71 14,050 2,980 -1,327 536.2 17.8 66.2 3.8 1971/72 15,247 4,233 -1,707 654.0 15.4 59.6 4.2 1973 18,404 3,857 -1,425 679.0 17.6 61.1 3.6 1974 23,771 4,506 -1,599 952.1 21.1 74.5 4.0 1975 26,577 5,153 -2,699 1,230.4 23.8 62.7 4.6 1976 30,203 5,554 -3,576 937.6 16.8 33.6 3.1 1977 36,407 6,148 -3,074 1,424.1 23.1 63.7 3.9 1978 42,665 10,408 -7,165 2,131.9 20.4 39.1 5.0 1979 52,387 11,502 -8,791 2,325.9 20.2 29.7 4.4 Sri Lanka: A Decade of Food Policy Reforms, 1977-87 33 Policy Reforms, 1977-87 Policy Changes of 1977 Before its election the government that took power in 1977 had done a great deal of work in formulating policies. Knowledgeable and experienced people in economics, business, administration, and finance had assisted the party leadership in the contemplation of basic policies as well as matters of policy implementation. Part of their work had been to construct a macro-policy framework that included action on food subsidies. This prior thinking on policies and preparation served the government well. Advance preparation was one of the factors that allowed it to move quickly and decisively in implementing the package of economic reforms of which the food policy reform was a part. Another factor was the government's large majority in Parliament-having won 144 of the 168 seats. This majority gave it the impetus and solid strength that was required for effecting fundamental changes. A third important factor was that while in opposition, the new government had built up the trade unions under its own leadership and could therefore count on their loyalty. It had become clear that the country could no longer sustain the burden of providing free rice to its population. With unemployment at 25 percent, the balance of payments in disarray, and the physical as well as economic infrastructure by now seriously run down, the government's first priority was to free the economy of controls, attempt to create jobs, and relieve the dangerous rate of unemployment. Accelerated development was seen as the only way to make a dent in the very serious unemployment situation. Maintaining large subsidies was seen as incompatible with accelerated development.1 While the large mandate of the government was reassuring, the history of the rice ration was replete with many a grim lesson, having cut short the political careers of promising Prime Ministers and Finance Ministers and having even led to violence and bloodshed. What the new government had in mind, which was to drastically reduce the numbers eligible for the ration, had to be carried out with thoroughness, due care, and anticipation. The government wasted no time. By September (2 months after being elected) it had started work on restructuring and reducing the food subsidies, particularly on rice. It fell to the lot of the Ministry of Food and Cooperatives, the Food department, and the Cooperatives Department to plan out the detailed steps of implementation. At this time the Ministry and its allied departments possessed a core of trained and experienced administrators who had both the confidence and the capacity to carry out a policy change of this nature. All were highly trained and had wide experience working in numerous fields closely related to the policy reform. At Independence, in 1948, the British left in place a sound administrative structure, with a class of professional civil servants at the apex, who were recruited through a rigorous open competitive examination. Although over the succeeding decades various changes had exacted a certain toll on standards of efficiency as well as the prestige of the administration, it was still, by 1977, a cadre of knowledgeable and experienced persons possessed of a basic confidence and esprit de corps. The new economic policy was spelt out in the first Budget Speech of the Minister of Finance and Planning. Its basic objectives were to: "1. revive and resuscitate our economy which is at breaking point; 2. restore vitality to our agriculture, our plantations, our industry, our fisheries, and our trade; 1. For a long time, international lending agencies such as the IMF and the World Bank had been advocating reform of the fiscal, monetary, and exchange rate policies, necessary to put Sri Lanka on a path of high and sustainable growth. Successive governments were reluctant to adopt such measures, fearing their political consequences. The time had now come when the perceptions and philosophy of the government coincided with the recommendations of the Fund and the Bank. 34 M. D. D. Pieris 3. increase employment opportunities, particularly for our youth; 4. stimulate domestic savings and investment; 5. reduce the pressure on our balance of payments; and 6. move forward from economic stagnation to economic growth and development." The policy initiatives to revive the stagnant economy included exchange and interest rate reforms, liberalization of the trading and payments system, removal of price controls, and budgetary reforms, particularly in the areas of subsidies and transfers. The Minister also noted that "As my programme will bring about a major restructuring of prices in the economy, some parts of it will have to extend over several years. It will be necessary to safeguard the standard of living of the lower-income groups." The Minister proposed in his budget to rationalize the food subsidy and distribution program. He announced that the generalized subsidy would be scaled down and would be extended only to those families whose monthly income was below Rs.300. About 7 million out of the then population of 14.5 million fell into this "protected" category and were to continue receiving subsidized rations of rice, sugar, and flour. These 7 million would each receive a weekly ration of one pound free and three pounds at the subsidized price of one rupee per pound, at a cost of Rs. 1,250 million. Flour was to be sold throughout the country at a subsidized price of 60 cents per pound, at a cost of Rs. 770 million; children under 12 years from families in the "protected" category were eligible to purchase sugar at one and a half pounds per child per month at 72 cents per pound, for a subsidized cost of Rs. 55 million; a subsidy on infants' milk food was to be continued, at a cost of Rs. 50 million. A national buffer stock was set up with the goal of stabilizing consumer and producer prices (see later in this section). Concurrently, the Minister announced a wage increase of 25 percent, up to a maximum of Rs. 600 a year, to afford relief to all wage earners who were no longer eligible for subsidized food. He also announced a scheme of income support, up to Rs. 600 a year, for unemployed persons. There was then a total package of measures aimed at cushioning the blow to food subsidies. Along with these proposals, the fertilizer subsidy was increased from 50 to 75 percent to producers of paddy, and the guaranteed price of paddy was raised by 25 percent. To rouse enthusiasm about national development prospects the government focused attention on three major projects. First, it would accelerate, from 30 years to 6-7 years, completion of the huge Mahaweli Scheme, which was a trans-basin river diversion scheme with the potential to more-than double the country's hydro power resources, dramatically increase the irrigable land area, and allay unemployment through huge settlement schemes. Second, the government began a housing and urban development project with an initial target of completing over 100,000 houses within 34 years. Third, it set up an economic investment, or free trade, zone, to enhance employment generation and foreign exchange earnings. Simultaneously, the government embarked on a sustained campaign to inform people of the incompatibility of funding such development activities while continuing large-scale food subsidies. These undertakings were very helpful in influencing public opinion in favor of the government's policies. The new food rationing scheme came into effect from January 30, 1978. At the end of 1977 an estimated 13.7 million rice ration books were in circulation. Ration books were revalidated only for persons whose family income was below Rs 300 per month. To effect this, special income declaration forms were issued through the existing cooperative and authorized distributor network, which consisted of more than 10,000 points covering the whole country. The implementation process from the issuing of the forms to the completion of the revalidating of ration books took only one month. At the height of the exercise the Officers of the Ministry of Food and Cooperatives, the Food Department, and selected officers from the Cooperative Department were available in their offices till almost midnight of each day so that any government agent with a problem could contact them and obtain instant advice. The government agents had in turn made suitable arrangements, in their districts, to ensure that the work was carried out expeditiously and that all the deadlines were met. It was largely this close and Sri Lanka: A Decade of Food Policy Reforms, 1977-87 35 intense coordination that enabled this sensitive and momentous exercise to proceed without a hitch. It must be remembered that there was no press censorship or any curtailment of the expression of views including opposition views. The revalidation exercise reduced the number of ration books by about half, to 7.7 million. Thus it was found that the population eligible for food subsidies had been reduced by about half. The package of policy measures taken resulted in a surge of growth. Imports accelerated to meet the demands of a long-starved economy. Consumer goods flooded the market. Foreign investment and intermediate goods came in expeditiously. Infrastructure development took off. A building boom followed, and thousands of new jobs were created in various sectors, including a fast-growing service sector. The economy grew at the rate of 8.2 percent in 1978 and 6.3 percent in 1979. Both the logic of the government's economic policies and the visible benefits stemming from them led to a consideration of terminating food rationing altogether, and replacing it with a food stamps scheme directed toward the genuinely indigent. Abolition of Rationing, 1979 The Food Stamps Scheme was introduced in October 1979. The results of the Socioeconomic Survey of 1969/70 had shown that for families in the poorest 30 percent of the population, subsidized food covered 15-25 percent of total expenditure on food; its withdrawal would have left this group nutritionally vulnerable. The Food Stamp Scheme was thus targeted to this group by limiting its assistance to families with monetary incomes of less than Rs. 3,600 a year. Though in concept it was an income transfer scheme, the subsidies were still related to specific items of food. The new scheme was formulated by a small committee of officials consisting of the Secretary to the Cabinet, the Secretary, Ministry of Food and Cooperatives, and the Additional Secretary, Ministry of Trade and Shipping, with the Secretary, Ministry of Finance and Planning, joining in at a later stage. Since prices were to be freed, in what was to be basically a free market economy, some sort of cushion had to be devised to protect the weaker sections. The Committee had to keep in mind that: T The government wished finally to dispense with rationing and ration books. * The massive development projects that had been launched needed enormous funds. * The government considered rice rationing per se unscientific, and felt that methods should be evolved to offer the indigent a wider choice of subsidized food. To determine who was eligible for food stamps, an incomes survey was undertaken before stamps were issued. It had been estimated that if a realistic declaration of incomes was made, the 7.5 million people covered by the rice rationing scheme could have been reduced by another 3 million. However, the results of the Survey were such that practically all those families who received assistance under the rice rationing scheme were found eligible to receive food stamps. The survey failed for several reasons. One related to the difficulty of accurately assessing rural incomes; the declarants tended to adjust their answers to make sure they remained in the scheme. The second factor was the inadequacy (recognized by respondents) of the legal framework to penalize those making false declarations. The third, and perhaps the most important, was the political pressure exerted by members of parliament on government officials to accept the income declarations without too exacting a scrutiny. In retrospect, these fears of losing votes appear to have been exaggerated; even five years after the government came to power, local elections and local government elections still confirmed substantial support for the government. The recipients of stamps were categorized in three groups: persons over 12 years old, who received Rs.180 worth of stamps a year; children 8-12 years old, who received Rs. 240 worth; 36 M. D. D. Pieris and children from birth to 8 years old, who received Rs. 300 worth. An eligible family typically received between Rs. 1,140 and Rs. 1,380 of stamps a year, depending on the ages of its children. The stamps could be used to buy rice, wheat flour, sugar, bread, milk foods, and specified pulses. When kerosene prices were increased in 1984, each of these households was also given kerosene stamps to the value of Rs.260 a year. Doing away with rationing was to lift an economic and psychological millstone from the country's neck. It gave a stimulus to the farming community which numbered over 6.5 million people. Freed from controls and the maintenance of artificial prices, rice production soared and by the end of 1984 had reached over 100 million bushels a year, up from about 68 million in 1977. In the initial stages of the scheme, an eligible household obtained about 12 percent more in money terms under the scheme than it had received under rice rationing. Households in the poorest 10 percent of the population received about 20 percent more, reflecting their larger family size. Once the Food Stamp Scheme had become operational and stabilized, however, its benefits were gradually allowed to decline. The prices of major food items were adjusted, to help cover costs. The nominal value of food stamps was kept unchanged from the inception of the Scheme up to December 1985, even though the prices of the items covered almost doubled in this period. As a result, the real value of food stamps fell substantially between 1979 and 1985. The reasons for permitting this deliberate erosion of benefits were as follows: * "leakage" of the benefits to unintended persons, estimated to be about half those receiving stamps under the scheme; * inability to release budgetary resources to fund the additional sums needed to maintain the real value of stamps, due to the large claims of the major investment projects being undertaken; * a decrease in unemployment, as a result of economic growth at an average rate of 6 percent a year. It is also noteworthy that accelerated growth in the paddy sector raised incomes for farmers and agricultural workers. These groups also benefited from a rise of 50 percent in the guaranteed price for paddy, which the government adjusted in line with international price changes. By 1984, the world market prices of sugar, wheat, and rice had begun to fall. Beginning in January of that year the government imposed import duties and other consumer taxes on these three commodities, partly to generate revenue but also to protect local producers of rice and sugar-in the firm belief that economic salvation could only be achieved by developing the productive sectors of the economy. Receipts from these taxes amounted to about Rs. 2 billion in 1985. Assuming that Rs. 1,050 million of these receipts were allocated to the Food Stamps Scheme, whose total budgetary allocation was about Rs. 1,500 million, the net transfer channeled through the Scheme by 1985 was Rs. 450 million, or less than 1 percent of the GDP. Pricing and Marketing of Food Items Policy reforms in the area of pricing and marketing after 1978 made for greater reliance on market forces. Government reduced its direct participation in procurement and distribution. The Paddy Marketing Board, a state agency, procured well over 40 percent of the production in 1970- 72, but less than 6 percent of the production after 1979, although rice production had risen by nearly 70 percent in the intervening years. Food Policy Committee A National Food Policy Committee was appointed after the introduction of the policy reforms in 1977. It consisted of the Secretary to the Cabinet (Chairman); Secretary, Finance and Planning; Secretary, Food and Cooperatives; Secretary, Trade and Shipping; and Secretary, Agricultural Development and Research. Senior Heads of Departments, such as the Food Sri Lanka: A Decade of Food Policy Reforms, 1977-87 37 Commissioner, also attended the meetings. The Committee's task was to keep the reforms under review and to effect proper coordination between the various ministries involved as well as to make ongoing recommendations to government on needs for modifications. Among other things the Committee was responsible for defining a pricing policy, with emphasis on paddy and subsidiary food crops such as legumes and maize. It also ensured a better coordination of the activities of the Paddy Marketing Board and the Department of the Food Commissioner, both in regard to maintaining the floor prices offered to farmers and in delivering rice to consumers without interruption. The Committee was expected to study the price movements of key agricultural products and their costs of production and make recommendations for assistance, either by reducing costs through subsidized inputs or at the market level by increasing support prices. It was also expected to recommend measures to protect local producers through changes in tariff levels. It kept under review the need to expand domestic consumption of rice and therefore the relative prices of rice substitutes, particularly wheat flour. It also studied the availability of food grains and made decisions as to import requirements, particularly of rice. The decisions of the Committee were recommendations in nature, that were put before the relevant ministers, who then decided on suitable action. Buffer Stocks Although no longer committed to providing universal food rations, the government did feel obliged to maintain adequate stocks in the country. It established national buffer stocks, under its own control, of rice, wheat flour, and sugar, as well as some subsidiary commodities. The main reasons for doing so were: * to be able at all times to meet the needs of the Food Stamps Scheme (particularly if items needed to be imported, there could be time lags); - to prevent shortages and black marketeering; * to reduce seasonal fluctuations in prices; • to make up for possible under-importation by the private sector; and - to have stocks readily available to meet special contingencies such as floods, droughts, or civil disturbances. Results of Policy Changes Budgetary Savings The changes in policy contributed substantially to budgetary savings and the use of resources for investments. Though initially the food subsidy continued to account for a high proportion of GDP (5 percent in 1978 and 4.4 percent in 1979), during the period 1980-85 this share declined to 2 percent. After the introduction of food stamps, the outlay on the subsidy accounted for 5 - 6 percent of the government's expenditures, compared with over 15 percent earlier. During the period 1980 to 1985, the cumulative savings on account of the policy change were around Rs. 26 billion. Effects on Income Distribution Increases in the prices of food items in Sri Lanka are reflected in the cost of living index and thus, for workers in the organized private sector (and in the public sector since January 1982), they are offset to a large extent by the payment of allowances. Thus the entire organized working sector is cushioned against increases in food prices. The groups affected most adversely 38 M. D. D. Pieris by higher food prices seem to be the urban unemployed, agricultural workers and to some extent plantation workers. There was apparently a pro-rural bias in the social policies that were followed and more so in the implementation process. As might have been expected, the reduction of subsidies hit the urban sector hardest. When rice rationing gave way to food stamps, the rural sector held on to its subsidies without any marked reduction; 58 percent of the rural population continued to receive benefits. This was not the case in the formal urban sector and the plantation sector where incomes were generally fixed, documented by employers, and therefore ascertainable. Participation was reduced from 40 percent to 32 percent in the urban sector and from 20 percent to 12 percent in the plantation sector. That the rural sector held on to a much larger share of the benefits was entirely due to the difficulty of obtaining realistic estimates of rural incomes. Looking at the occupational distribution of households, as revealed in the surveys of 1978/79 and 1981/82, it was clear that about 80 percent of agricultural laborer households were covered by the Food Stamps Scheme. This group accounted for a substantial share of the poorest twenty five percent of the population. Thus the scheme was quite effective in covering those who were nutritionally at risk, but the assistance declined over time as the real value of the stamps declined. The impact of this decline on nutritional levels is discussed in the following section. As regards leakage of benefits, over 55 percent of those producing rice and other crops, small livestock farmers, and fishermen, were covered. Substantial leakage of benefits probably occurred among these groups. Earlier in this paper it was noted that the food policies in force in the 1950s and 1960s helped greatly to improve the distribution of real income. Surveys of household expenditures made by the Central Bank in 1981/82 suggest that since the food policy changes of the late 1970s the income distribution pattern has worsened (see Table 2.5). Table 2.6 shows that the share of income accruing to the bottom 40 percent declined from 15.05 percent in 1973 to 12.12 percent in 1978/79, and slid further, to 11.78 percent, in 1981/82. Meanwhile, the share of income accruing to the top 20 percent increased from 45.89 percent in 1973 to 54.31 percent in 1978/79, and to 56.52 percent in 1981/82. Table 2.5 Estimates of Gini-Ratios Based on Income Surveys for Income Receivers 1953 1963 1973 1978/79 1981/82 For income receivers 0.50 0.49 0.41 0.49 0.52 Table 2.6 Percentage of Income Share of Each 10 Percent of Income Receivers Decile 1953 1963 1973 1978/79 1981/82 Lowest 1.51 1.17 1.80 1.20 1.21 2nd 3.56 2.70 3.17 2.56 2.49 3rd 3.56 3.56 4.38 3.60 3.47 4th 4.37 4.57 5.70 4.76 4.61 5th 5.71 5.55 7.10 5.93 5.57 6th 6.31 6.82 8.75 7.29 6.93 7th 7.94 8.98 10.56 9.12 8.56 8th 10.39 11.46 12.65 11.23 10.64 9th 14.16 16.01 15.91 15.26 14.82 Highest 10th 42.49 39.24 29.98 39.05 41.70 Sri Lanka: A Decade of Food Policy Reforms, 1977-87 39 Despite the worsening in the income distribution, no serious political pressures developed for its correction. In absolute terms the poor had gained. Many found jobs after 1977 and were able to enhance their family incomes. Higher economic growth both hid and cushioned the real income disparities that began to emerge. Effects on Consumption Levels of the Poor At the beginning of the food stamps scheme, an average eligible family received about Rs. 95 per month. The poorest 20 percent of families, who tended to have more children than the average, received Rs. 115 a month. These income figures were based on the calorie allowances recommended for Sri Lanka's conditions. The recommended allowance was 2,200 units per capita per day; the critical level was 30 percent below this level, that is, 1,540 calories. Table 2.7 shows that the mean per capita calorie intake declined between 1978/79 and 1981/82 for the urban and plantation sector, though it rose in the rural sector. The data for income groups in Table 2.8 show that the bottom ten percent were consuming less than the critical level of 1,540 calories a day in 1978/79. By 1981/82, this was true even for a part of the next income decile-those in urban areas, who seem to have suffered most as a result of the escalation of food prices. Overall, for the bottom 20 percent of the population, there was a marked deterioration in calorie intake of 10-15 percent. This happened despite the significant growth of GDP, at around 6 percent a year. Table 2.7 Mean Calorie Consumption (per capita per day) 1978/79 survey 1981/82 survey All island 2,283 2,271 Urban sector 2,240 2,229 Rural sector 2,230 2,246 Estate sector 2,763 2,639 Table 2.8 Per Capita Dietary Calorie Consumption Per capita expenditure per decile (commencing ---------------1978/79 survey --------- --------------- 1981/82 survey -------- from lowest All All decile) island Urban Rural Estate island Urban Rural Estate 1 1,335 1,288 1,346 1,324 1,181 1,137 1,186 1,214 2 1,663 1,620 1,663 1,821 1,558 1,351 1,586 1,607 3 1,848 1,718 1,855 2,027 1,794 1,589 1,813 1,924 4 1,994 1,824 1,999 2,222 2,008 1,784 2,031 2,122 5 2,157 1,917 2,155 2,490 2,168 1,927 2,184 2,371 6 2,377 2,079 2,385 2,716 2,373 2,088 2,392 2,687 7 2,528 2,260 2,505 3,032 2,553 2,216 3,581 3,024 8 2,738 2,495 2,757 3,160 2,838 2,484 2,869 3,344 9 3,034 2,674 3,071 3,884 3,120 2,705 3,203 3,783 10 3,296 3,181 3,361 3,845 3,216 2,882 3,475 3,549 40 M. D. D. Pieris The agricultural sector grew at nearly 4.4 percent a year from 1976-82; in the case of paddy, growth was around 8 percent. Within this sector agricultural laborers, on plantations as well as farms, experienced more deterioration in calorie intake than farmers. Table 2.9 compares some key prices. The only conclusion that can be drawn is that by 1981/82, as the prices of key commodities rose, and the real value of food stamps declined, changes in incomes and the substitution of less costly sources of calories were inadequate to maintain the calorie intake of 1978/79. The position had worsened by 1985. Table 2.9 Some Prices of Key Commodities, Selected Years (Rs. per kilogram) December December December Commodity 1979 1982 1985 Rice -food stamp 3.56 6.16 7.22 Rice - open market 3.77 6.51 10.05 Flour 2.99 5.95 7.90 Bread 2.76 5.50 6.89 Milk powder 18.70 35.88 52.20 Sugar 6.60 11.90 14.00 Malnutrition among Children In 1977 the new government revamped the specialized organization called the Food and Nutrition (Policy & Planning) Division of the Ministry of Plan Implementation and placed it directly under the President. This division assists the government to develop measures to combat malnutrition, particularly among children. Successive Sri Lankan governments have tried to combat this very widespread problem. This is evident from the allocation of substantial sums for food subsidies, health services, and free education. Surveys taken in 1975/76 and 1980/82 showed a considerable increase in the proportion of children affected by chronic undernutrition in the age group of 6 to 24 months in many districts of the country. The reports pertaining to an increase in malnutrition were studied not only in the ministries concerned, such as the Ministry of Plan Implementation, but also at forums such as the Development Secretaries' Committee, which is a weekly meeting of Secretaries of the various development and service ministries. The President also apprised the Cabinet about the situation from time to time. A series of meetings and discussions, including seminars, led to the formulation and publication of an Agriculture, Food and Nutrition Strategy. This was based on the domestic production of food items, limited to nutrition and consumption needs. The strategy calls for income distributional aspects to be given adequate weight in the evaluation of social returns for development investments, and for the maintenance of an efficient food distribution system, buttressed with nutrition education. It calls for constant monitoring and review of the prices of key foods, not only to promote price stability but also to promote the production and consumption of the cheaper food items consumed by low-income groups. The strategy calls for a coordinated effort in the areas of sanitation, health, education, and food production and supplies. Though the Food Stamp Scheme is one of the most important measures of assistance for the poorer sections of the population, it may not adequately cover children in the age group 1-6 years. To assist this group, the government has adopted feeding programs. Under the Thriposha Programme, protein-fortified biscuits are issued free to malnourished preschool children and pregnant and lactating mothers in low income groups; about 650,000 persons Sri Lanka: A Decade of Food Policy Reforms, 1977-87 41 benefit. The school biscuit feeding program benefits about 1.2 million children between 6 and 9 years of age, while the Kola Kenda Programme provides cooked food to poor children between 5 and 9 years, thus far reaching about 900 project schools. This program has evoked a high level of community participation. Further, for 240,000 persons who, though receiving food stamps, have been identified as extremely poor, the Department of Social Services administers a monthly "dole" style allowance, being guided by such factors as physical incapacity to work, number of young dependents, etc. The scheme is administered through the District Administration headed by the government Agents who are the official heads of administrative districts. The government Agent processes the applications for assistance and sanctions the payment of the monthly allowances. He is assisted by a number of Assistant Government Agents within his district, who in turn rely on the counsel of social service advisory committees consisting of village elders, rural development societies, and the chairman of village/town, or urban councils. A head of an absolutely poor family of, say, 5 persons, together with the food stamps, will receive a total of around Rs. 355 a month, and can thus meet up to 75 percent of the family's calorie needs. 1986 Review of Food Stamp Scheme Between 1977 and the introduction of the Food Stamp Scheme in 1979, unemployment had fallen dramatically (from 24 percent in 1977 to 12 percent in 1985) and incomes had shown a substantial increase. Various income and expenditure surveys undertaken by the Central Bank and by the Department of Census and Statistics showed that at most, only the poorest 30 percent of the population really needed state assistance. There were other inequities. For example, no new applicants had been admitted to the scheme since 1984, when the government had closed the rolls in order to contain its budgetary expenditures. A Committee was appointed to recommend how the Scheme could be reformulated so as to be precisely targeted toward the really vulnerable. Its members were the Secretaries of the Ministries of Finance and Planning, Food and Cooperatives, Social Services, Health, and Home Affairs; the Food Commissioner, Director of Fiscal Policy - General Treasury, and the Director of Social Services also participated. Among the Committee's recommendations were the following: * To increase the value of food stamps, in view of the fact that the prices of food items had more than doubled since 1979. * To eliminate ineligible persons through a new income survey. If such a survey could eliminate about 3 million persons who did not deserve assistance, it could be possible to double the value of the food stamps. To assess incomes realistically, the Committee recommended that the income declarations be obtained at public sittings of officials at the village level, using the participation of the community to ascertain the economic status of applicants. This was the method that had been successfully used earlier to identify landless families. * To enact a new law to regulate the administration of the scheme (so as to permit sanctions to be imposed on those making false income declarations); * To raise the income cut-off point for eligibility for stamps from Rs. 300 to Rs. 700 a month. The Committee welcomed the decision of the Minister of Finance and Planning to transfer the administration of the scheme from the Department of the Food Commissioner to the Department of Social Services. It was felt that this would discourage families who could manage on their own from applying for assistance, because of the social stigma involved in receiving assistance from the Department of Social Services. The government accepted the Committee's recommendations. A new law, initially drafted by the Committee, was enacted in Parliament (the "Poor Relief Act," later renamed the "Public Assistance Act"). This act provided for the administration of poor relief and the Director of Social Services was appointed as Commissioner for Poor Relief. The Government Agent of each 42 M. D. D. Pieris District was ex officio appointed Deputy Commissioner for Poor Relief. In terms of this law, the Commissioner may require from a person who seeks assistance a declaration of his income in such a form and manner as he may prescribe. The law also empowered the Commissioner to discontinue assistance if an applicant was found to have fraudulently obtained assistance or if his status had changed. It further empowered him to impose a penalty by way of a forfeiture equivalent to the value of benefits enjoyed by such as person and to prosecute such a person before the Courts. The income survey was taken on a single day in May 1986. All Government Agents of the Districts, the Assistant Government Agents, and all officers of all government Departments at the District level were drawn into the exercise. It was also decided to publicize the proposals through the media, so as to explain the scheme to the people and elicit their cooperation. Due emphasis was given to the need to direct assistance to the absolute poor in a meaningful way, and at the same time to explain why assistance has to be denied to those with incomes adequate to meet their basic needs, given the need to devote more of the country's scarce resources to development expenditures. The Government Agents and District Officers were put through several training seminars so that they understood the scheme quite well. The survey revealed that 2 million people then receiving food stamps should be removed from the scheme. It was felt that when their income declarations were processed further at public sittings, the number eligible for food stamps could be further scaled down. However, Members of Parliament began to face extensive pressure from those who were to lose their food stamps. At a Government Parliamentary Group Meeting held in July 1986, Members of Parliament protested to the President and requested a review of the scheme. A Committee of them was appointed under the chairmanship of a senior Member to review the technical aspects of the scheme, particularly as regards the obtaining of income declarations, and to make suitable recommendations. After lengthy deliberations, the Committee concluded there was nothing wrong with the reforms and procedures proposed. However, the pressures that had built up could not be defused. The Cabinet of Ministers therefore decided that all members of a family declaring an income of Rs. 300 a month be given stamps without restriction. This decision made an additional 1.3 million persons eligible to receive the stamps. The Scheme as it now stands supports about 7.2 million people, which is around 46 percent of the population, and is virtually at the same level as before the review took place. The wheel has turned full circle. Universal Conclusions This review of experience confirms several general conclusions. These are that: * It is exceedingly difficult to sustain for long an extensive welfare program on a narrow resource base. * It is easier to get into a position of subsidization than to make subsequent adjustments or to remove them completely. * Subsidies are predisposed to proliferate and to grow-particularly in parliamentary democracies where electoral pressures are heavy on every member of parliament and where members have to be accessible to their voters on a regular and continuous basis. * Systems of parliamentary democracy based on universal adult franchise and periodic elections lead to the raising of popular expectations by parties vying for power. There always seems to be a significant gap between the beliefs held and promises made at election time and what a country's resources can actually bear. * A change of governments, particularly where there is also a change of political parties or groups, leads to a climate where, at least for some time, the public is more receptive to policy change. Sri Lanka: A Decade of Food Policy Reforms, 1977-87 43 • Since, even in the case of a recently elected popular government, the honeymoon period is relatively short, it is important to make any fundamental changes of policy early and decisively. This process is greatly assisted if the party coming to power has already mapped out a strategy whilst in opposition. * As to long-entrenched subsidies, particularly on basic foods, even a change of government and the award of a massive mandate may not be adequate to accomplish change. Governments may therefore wish to introduce a total package to the people such as the promise and hope of significant economic development leading to the creation of employment, and the ready availability of goods, as well as various forms of relief such as increased salaries, minimal forms of assistance for the indigent, etc. * When social programs have to be curtailed in a stagnant economy, the impact felt will be much greater than in a rapidly growing one, where income-generating opportunities become available. * In an exercise involving the curtailment or abolition of long-enjoyed subsidies, which have over the years become an economic and social prop, pressures and hostility from an organized working class and powerful trade unions can be a serious impediment. Thus if a party or a government has over the years built up a base within this sector, it becomes easier to carry out the reform and to neutralize opposition. * It seems always better and more prudent to make relatively small changes in pricing, exchange rates, and other matters from time to time than to postpone them in the hope that conditions will get better. Neglecting to make these small changes tends to aggravate economic and political difficulties, leading to the need to make drastic changes later. - In opening an economy, particularly after decades of protection, planning, and administered prices, the release of pent-up demand leads to a period of accelerated activity resulting in expanded employment, greater availabilities, and immediate prosperity relative to the past. At the same time, however, economic disparities tend to widen when prices are released to find their own levels, and when the protection afforded by subsidies is significantly reduced. In such a context time will be required for overall development to seep down, particularly to the poorest 30 percent of the population. In the meantime policies would need to be devised to protect the health and nutrition of the poor. In a context of substantial reductions of general subsidies it may be necessary to identify and target certain limited groups for various intervention programs aimed at providing some increase in their intake of calories and in raising standards of nutrition. Indexation of monetary allowances in relation to the rises in the cost of living may not be feasible for relatively poor developing countries. 3 EXCHANGE RATE REFORM IN PAKISTAN, 1982 Javed Hamid Naved Hamid This is a case study of the exchange rate reform in Pakistan in 1982. Its focus is on the decision-making process involved in delinking the Pakistan rupee from the US dollar and the management of the exchange rate in the period following this decision. The success of the policy change is assessed in the context of Pakistan's economic management at that time. Historical Background Balance of Payments Crisis, 1979 The government that came to power in 1977 inherited a number of serious economic problems. Private industry was demoralized and public enterprise had deteriorated, in part through losses of personnel to the private sector and overseas. Heavy commitments for public sector projects in progress, combined with a need to reduce budgetary deficits and reliance on borrowing from domestic banks, limited the government's ability to reorient public expenditures toward new higher priority investments. The availability of foreign exchange was restricted by poor export performance, declining terms of trade, and falling net aid inflows. The deficit in the current account of the balance of payments had reached US$1.1 billion, or 6.2 percent of GDP, and reserves had fallen to the equivalent of about 6 weeks of imports. In 1979 the second oil shock further aggravated Pakistan's balance of payments difficulties. The monthly oil import bill increased from US $35 million in 1978-79 to US $73 million in 1979- 80. By September 1979, foreign exchange reserves had dropped to less than three weeks of imports. In December 1980, Pakistan borrowed 1,268 million SDRs for three years from the Extended Fund Facility of the IMF, and in January 1981, it obtained an 18-month debt rescheduling agreement which reduced the debt-servicing burden by US$400 million for this period. Some financial assistance was also obtained from the oil-rich middle eastern countries. Pressures for Devaluation Since 1973 the official exchange rate between the Pakistani rupee and the US dollar, to which it was linked, had been maintained unchanged. When the dollar began to rise against SDR currencies in mid-1980, the Pakistani rupee began to appreciate against the currencies of European and other trading-partner countries. As Figure 3.1 shows, between July 1980 and September 1981, the index of the nominal relative exchange rate had increased by about 14 percent and there was a real appreciation of about 22 percent. At the beginning of 1981, A.G.N. Kazi, then Governor of the State Bank of Pakistan, wrote to the then Finance Minister, Mr. Ghulam Ishaq Khan, that with the appreciation of the US dollar, the exchange rate of the rupee must be adjusted if Pakistan's exports were not to suffer. 44 Exchange Rate Reform in Pakistan, 1982 45 Figure 3.1 Movements in the Trade-Weighted Real Exchange Rate and Its Components, 1974-83 January 1973 = 100) 160 150 Relative real exchange rate index1 140- 130 _ - - , - - - - , 120 J, Relative consumer price index2 110 _ -_ 100 Relative exchange rate index3 90 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 Sources: International Financial Statistics, Direction of Trade,and IMF staff estimates, as of end August 1983. 1. Equals (relative consumer price index x relative exchange rate index) + 100; values in excess of 100 are indicative of a relative real appreciation of the Pakistan rupee since January 1973. 2. Pakistan consumer price index relative to trade-weighted consumer price indices of 14 major trading partners. 3. Trade-weighted exchange rates of 14 major trading partners relative to exchange rate of the Pakistan rupee; a rise in the index indicates a relative appreciation of the rupee. Since 1977-78 Pakistan's exports had been increasing rapidly, growing by more than 30 percent a year in 1978-80. In 1980, exports grew by 25 percent, but there were signs of slowing down in the second half of the year. Mr. Kazi was convinced that this slowing of export growth was due to the appreciation of the Pakistan rupee along with the US dollar, but he did not yet have the necessary supporting evidence. At this time, Pakistan was emerging from a period of political isolation from the world community and had recently started receiving support from international financial institutions and bilateral sources. There was thus pressure to adopt economic adjustment measures considered important by the international community. It was these pressures and his sensitivity to the balance of payments problems experienced during 1979 that made Mr. Kazi recommend an adjustment in the exchange rate even before the effects on exports of the recent appreciation of the Pakistan rupee were clearly visible. Popular sentiment was vigorously opposed to devaluation. When Dr. S. Karl Schiller, engaged by the Government of Pakistan to study the Pakistan economy, recommended a devaluation of 20 percent against the dollar (the extent of the rupee's appreciation in real terms since January 1973), this occasioned something of a public outcry. The influential Dawn newspaper, for example, published an editorial entitled "Highly Ill-Advised" on February 21, 1981, arguing that devaluation would give rise to inflation without increasing Pakistan's exports and therefore "we should forget about the devaluation of the rupee." In response, Finance Minister Khan, in an interview on February 23, 1981, stated that "no devaluation of the Pakistani rupee is being considered. All speculation in this regard is totally baseless." This reflected the actual thinking of the Finance Ministry at that time. The Finance Minister well understood the economic and financial problems facing Pakistan, but felt that the US dollar was unlikely to sustain its rise, and was unwilling to make a decision that might have adverse political consequences unless this were absolutely necessary. One of the understandings in Pakistan's EFF agreement with the International Monetary Fund, signed in November 1980, was that to promote exports the government would, by January 1981, broaden its list of export products eligible for compensatory rebates, but the agreement had stipulated nothing regarding exchange rate adjustment. However, when an IMF mission visited Islamabad in May 1981 for Article IV consultations, it indicated to the Government of Pakistan that an exchange rate adjustment was required. 46 Javed Hamid and Naved Hamid Concerned by the continued rise of the US dollar, Kazi analyzed monthly export data. By September 1981 he had enough evidence to show that Pakistan's exports were suffering (see Table 3.1 and Figure 3.2). By October 1981, exports were below those for the same month two years earlier. As a lower-risk alternative to devaluation, he suggested delinking the rupee from the dollar, and setting the exchange rate using a managed float against a basket of currencies. Table 3.1 Pakistan: Exports by Economic Categories 1979-85 (US$ millions) 1978179 1979180 1980181 1981/82 1982/83 1983/84 1984/85 1985/86 Primary commodities 552.9 993.8 1,295.4 863.6 813.1 800.6 722.7 1,130.5 Semimanufactured 352.4 355.5 335.4 332.4 363.6 383.8 439.6 520.6 Manufactured 804.3 1,015.4 1,327.9 1,293.7 1,535.3 1,586.4 1,340.3 1,620.1 Total 1,709.6 2,364.7 2,958.7 2,489.7 2,712.0 2,770.8 2,502.6 3,271.2 Source: Pakistan Economic Survey, 1985-86 (Supplement Edition), Finance Division, Government of Pakistan. Figure 3.2 Monthly Exports for 1979-82 (Thousand U.S. dollars) 340 310 1981 280 /1980 \ , 220 7 130 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Exchange Rate Reform in Pakistan, 1982 47 Decision-Making Process Options Once the Finance Minister was convinced that action was needed to maintain export growth, he had three possible options: EXPORT SUBSIDIES. Pakistan had used export rebates in the past to promote exports of manufactured goods. Many commodities were already eligible for export rebates in 1979-80 and, as noted above, in 1981 the list was broadened to cover all major manufactured exports. These rebates were in the 7.5 to 12.5 percent range. To have raised them would have imposed an unbearable burden on the government budget and at the same time increased the administrative problems of dealing with malpractices. DEVALUATION was politically undesirable. Also, given the instability of major currencies and uncertainty about the future of the US dollar, it was extremely difficult to decide on how far to devalue. Further devaluations might always be needed if the link with the dollar were maintained and the dollar continued to appreciate. Because repeated devaluations would be seen as failures of government economic policies, this alternative was rejected fairly early in the decision-making process. DELINKING. This option offered the government flexibility in managing the exchange rate. The value of the rupee could be adjusted slowly, avoiding the traumatic political, psychological, and economic consequences of a sharp devaluation. The problem with this option was the choice of an alternative reference for fixing the exchange rate. One possibility was to link the rupee with the weakest convertible currency, to ensure the competitiveness of Pakistan's exports. However, it was impossible to predict which currency (if any) would remain the weakest over an extended period. The other possibility was to link the rupee with a basket of the currencies of its major trading partners-i.e. to adopt a managed float. Discussions on the exchange rate between the Central Bank Governor and the Finance Minister lasted almost a year. During this period the question was not formally discussed with anyone except the Finance Secretary. Messrs. Kazi and Khan felt the matter had to be treated with great secrecy to avoid danger of speculation, capital flight, and a build-up of public opinion against an exchange-rate adjustment. (The time taken to reach the decision was not long given the complexity of the issue. By comparison, in 1972, when Mr. Kazi had also been involved, it had taken three years to convince the government to devalue the rupee, which was much more overvalued then than in 1981.) Normally a decision of this magnitude would require extensive analysis of its implications for different sectors of the economy, and consultations with concerned ministries. This was not done. The need for secrecy was an important reason, but it was also because of the government's limited capacity for complex technical analysis. Efforts were not made, for example, to quantify the impact of the adjustment on the budget or develop policy options to deal with such issues as subsidies, regulated prices, increased rupee costs of public and private investment, debt servicing and uncertainty in cost of foreign trade transactions. Kazi felt that the exchange rate adjustment would be relatively small and gradual, and would allow the government to adjust other policies as and when needed. Once the Finance Minister was convinced of the need for action on the exchange rate, he could bring a proposal up for the President's approval. The decision could then go to the Cabinet for ratification. In 1981 Pakistan had no parliament. Power was centralized in the President, who took all important decisions. The President had complete confidence in his Finance Minister, treating him like his Prime Minister. The Cabinet was mostly comprised of technocrats appointed by the President; any decision supported by the President and Finance Minister did not require an elaborate process of consensus building among other ministers. Had the Finance Minister been of lesser influence, a decision of such importance would have required 48 javed Hamid and Naved Hamid more discussion among the important ministers and other senior officials dealing with economic matters. Also, the President would have required more general support for the decision rather than the recommendation of the Finance Minister alone. After a lengthy discussion, the Cabinet decided on January 7, 1982 that the rupee should be delinked from the dollar with effect from the following day. As the Government memorandum prepared for the consortium meeting in Paris in May 1982 put it, "Since the Pakistan rupee was linked to the US dollar, exporters found it difficult to remain competitive, especially in the case of products destined for those countries the currencies of which had depreciated relative to the dollar. ... This step ... will permit flexibility in exchange rate policy which is necessary in the present unstable order of currency values. It will also enable the government to gradually withdraw some export subsidies and concessions that have had to be given in the past to compensate exporters for an unfavorable exchange rate. Since the concessions tended to be specific rather than general in nature, the incentive system had got somewhat distorted. As such, the delinking provides a useful opportunity to introduce a more rational system.. .of incentives." To minimize the negative political impact of the exchange rate adjustment, it was presented as an assertion of national independence in economic policy formulation. For example, in his first interview with the press about the decision, the Finance Minister was reported as saying that "with the diversification of trade, the Government thought it fit to delink the rupee from the dollar and provide Pak rupee an active exchange rate ... fortunes of rupee were buffeted by fortunes or misfortunes of dollar and we had an exchange rate which depended on the US economy. ... We thought time had come that we are no longer able to be governed by the fortunes or misfortunes of dollar." The timing of the decision and its projection by the government were very skillfully managed. As a result, it won the support of the business community and met little or no opposition from the public at large. This was reflected in a number of editorials and other commentaries in the press. Implementation Methods The day-to-day setting of the exchange rate was left to Mr. Kazi, who used three indicators: • a basket consisting of the currencies of Pakistan's 14 most important trading partner countries; • another basket consisting of the currencies of 32 major countries to which Pakistan was exporting; and * exchange rates of the rupee relative to export-competing countries, notably India, Singapore, and Hong Kong. His major concern was the value of the rupee with respect to the trade-weighted basket of currencies of 14 countries. Based on daily figures on the exchange rates for these currencies against the dollar, he estimated the reference rate of the Pakistan rupee against the US dollar. The exchange rate of the rupee against other currencies was then automatically determined on the basis of the market rates of these currencies against the dollar. Changes in Pakistan's trade pattern were reflected in the weights assigned to individual currencies in the basket. Also, since the concern was to manage the real effective trade-weighted value of the Pakistan rupee, projections were made regarding the expected inflation rates in these countries. Finally, to discourage speculation by exporters and people making remittances, the value of the rupee was sometimes adjusted upwards even though this was not warranted by movements in the exchange rates of trading partner currencies. Consequently, there was considerable discretion in the daily fixation of the exchange rate. The basket of currencies with which the rupee was linked was never made public. This enabled Mr. Kazi to exercise his judgement in the day to day fixing, thus allowing him to adjust more smoothly than if the basket were publicly known and the link had been a rigid one. Exchange Rate Reform in Pakistan, 1982 49 Because he did not want to risk anyone profiting from privileged access to information, Mr. Kazi used his staff only for tracking the value of the rupee against different currencies and providing him analyses so he could, where necessary, make adjustments in the longer-term management of the exchange rate. Exchange Rate Policy The initial objective of the delinking was to bring the real effective trade-weighted value of the Pakistan rupee back to where it was in early 1981, before its "ride up" with the US dollar. This was done gradually over a period of about 6 months. By the second quarter of 1982 the real trade-weighted exchange rate was approximately at the same level as in the second quarter of 1980. (See Tables 3.2, 3.3.) Table 3.2 Pakistan: Movements in the Trade-Weighted Real Exchange Rate and Its Components, 1973-83 (January 1973 = 100) Relative Relative Relative consumer exchange real exchange Year/quarter price indexa rate indexb rate indexc 1973 Fourth quarter 119.8 105.8 126.7 1974 First quarter 114.0 104.2 118.8 Second quarter 115.5 105.0 121.3 Third quarter 122.1 106.8 130.4 Fourth quarter 124.3 105.0 130.6 1975 First quarter 125.1 103.0 128.9 Second quarter 127.5 104.7 133.5 Third quarter 128.0 109.4 140.0 Fourth quarter 127.0 109.3 138.9 1976 First quarter 123.1 110.6 136.2 Second quarter 121.6 111.8 136.0 Third quarter 123.4 111.4 137.5 Fourth quarter 126.4 111.1 139.8 1977 First quarter 126.6 107.5 138.6 Second quarter 124.7 108.7 135.6 Third quarter 127.4 109.0 138.8 Fourth quarter 127.0 109.3 136.2 1978 First quarter 125.8 105.5 132.7 Second quarter 125.2 104.4 130.7 Third quarter 128.9 101.1 130.4 Fourth quarter 127.2 101.4 129.1 1979 First quarter 125.4 103.2 129.4 Second quarter 124.0 103.4 128.3 Third quarter 130.8 102.4 133.9 Fourth quarter 125.9 103.2 130.0 Continued ... 50 javed Hamid and Naved Hamid Table 3.2 Continued Relative Relative Relative consumer exchange real exchange Year/quarter price indexa rate indexb rate indexc 1980 First quarter 123.9 107.2 132.8 Second quarter 123.8 101.4 125.5 Third quarter 129.4 101.8 131.8 Fourth quarter 129.4 103.3 133.7 1981 First quarter 127.2 107.0 136.0 Second quarter 126.7 113.8 144.3 Third quarter 132.3 116.0 153.4 Fourth quarter 130.8 113.4 149.5 1982 First quarter 129.3 102.9 133.0 Second quarter 127.9 98.5 126.1 Third quarter 128.9 99.0 127.5 Fourth quarter 128.9 93.4 120.3 1983 First quarter 128.4 94.9 121.9 Second quarter 128.9 95.3 122.9 a. Pakistan consumer price index relative to trade-weighted consumer price indices of 14 major trading partners. b. Exchange rate of the Pakistan rupee relative to trade-weighted exchange rates of 14 major trading partners; a rise in the index indicates a relative appreciation of the rupee. c. Equals (relative consumer price index x relative exchange rate index) x 100; values in excess of 100 are indicative of a relative real appreciation of the Pakistan rupee since January 1973. Source: IMF, International Financial Statistics and Direction of Trade, and staff estimates. Thereafter, in consultation with the IMF, it was decided to maintain the real effective rate at a level about 17 percent below that at the time of the delinking. This level of depreciation was more or less achieved by the end of the first quarter of 1983. The policy in 1983 was to keep the real exchange rate steady against the basket of currencies of 14 important trading partners. During this period the real effective trade-weighted exchange rate fluctuated somewhat because forecasting relative price movements, even in the short term, was difficult. The rupee tended to move up against European currencies, because the dollar was moving sharply up and 50 percent of Pakistan's trade was with the US and with Middle Eastern countries whose currencies were linked to the dollar. By the end of the fourth quarter of 1984, the rupee in terms of its real trade-weighted exchange rate had appreciated by about 6 percent over its level in the first quarter of 1983. When the dollar started falling in 1985, the exchange rate management policy was changed. Mr. Kazi, in consultation with the Finance Minister, decided to maintain parity with the dollar to allow the rupee to slide down against the basket of currencies. As a result, the real effective trade-weighted value of the Pakistan rupee declined to more than compensate for its appreciation since the first quarter of 1983. Exchange Rate Reform in Pakistan, 1982 51 Table 3.3 Pakistan: Movements in the Real Effective Exchange Rate for the Rupee and Its Components, 1981-85 (December 1981 = 100) -----------Trade-weighted basketa --------- -------Export competitors' basket b Relative Relative Nominal Relative real Nominal Relative real Yearlquarter exchange consumer exchange exchange consumer exchange end of period rate indexC price index rate indexc rate indexC price index rate indexC 1981 First quarter 94.0 97.3 91.4 90.4 98.1 88.5 Second quarter 99.2 96.6 95.6 97.6 97.5 95.2 Third quarter 103.5 101.3 104.8 103.8 101.7 105.5 Fourth quarter 100.6 100.9 101.6 100.3 100.9 101.3 1982 First quarter 95.6 98.6 94.3 96.4 98.8 95.3 Second quarter 89.1 97.4 86.7 90.4 97.5 88.1 Third quarter 88.1 98.4 86.8 90.6 98.1 88.8 Fourth quarter 85.9 98.1 84.2 88.7 97.8 86.7 1983 First quarter 82.8 97.8 80.9 85.5 97.7 83.4 Second quarter 83.4 98.9 82.3 87.2 98.6 85.7 Third quarter 82.9 101.8 84.2 88.1 101.0 88.6 Fourth quarter 83.1 103.7 85.8 89.0 102.8 90.9 1984 First quarter 82.4 101.2 83.1 88.6 100.6 88.5 Second quarter 81.3 101.3 82.0 87.4 100.6 87.2 Third quarter 82.9 105.7 87.3 90.5 104.6 93.8 Fourth quarter 81.5 105.5 85.5 89.6 104.1 92.2 1985 First quarter 80.2 106.2 84.6 89.2 104.6 92.2 Second quarter 76.4 105.9 80.2 83.8 104.1 86.0 Third quarter 73.2 108.4 78.8 78.7 106.6 82.7 Fourth quarter 69.7 108.1 74.8 72.7 106.1 76.0 a. Index based on 14 major trade partner weights (excluding petroleum). b. Index based on major export competitor weights for manufactured exports. c. A rise indicates an appreciation. Source: IMF, International Financial Statistics and Direction of Trade, and staff estimates. The 1985 decision to allow the rupee to decline with the dollar attracted little attention, since the Pakistani public only takes serious notice when the rupee moves against the dollar. For the budget, the implications of the change were considered minimal because the foreign exchange allocations for all government departments were made in US dollars and the budget remains unaffected as long as the rupee-dollar rate is unchanged. The Wider Policy Context As discussed earlier, the delinking decision was originally seen in a fairly narrow perspective, being largely designed to ensure export competitiveness. Given Pakistan's 52 Javed Hamid and Naved Hamid dependence on imports (imports account for 20 percent of GDP), however, adjustments in the exchange rate affect every aspect of the country's economy. While the government had not planned for the adjustments that became necessary as a result of the delinking of the rupee, it responded to the pressures as they arose. Successful Demand Management Between Jan 8, 1982 and 1985-86 the rupee depreciated by 42.77 percent against the dollar. Annual inflation during this period was only about 7 percent on average, compared with about 10 percent in the preceding five years, and since 1983-84, the rate of price increase has declined considerably. This was made possible, first, by a gradual reduction in the growth rate of monetary expansion (from an annual average rate of 21 percent between 1976-77 and 1980-81 to 15 percent during the five years ending June 1986), but also by several other developments: real GDP continued to grow rapidly (during the five years ending 1985-86 its real growth stood at 6.8 percent a year, compared with 6 percent a year during the preceding five years); international prices of a number of import items declined; and a continued trade deficit of about US$3 billion during the past five years enhanced availabilities in the economy. (See Table 3.4.) Table 3A Selected Monetary and Price Indicators, 1976-86 ------------------ Annual rate of percentage change ------------------ Consumer GDP price Real Years deflator index M2 GNP 1976-77 10.69 11.77 24.3 4.19 1977-78 9.03 7.79 23.0 10.72 1978-79 5.43 6.63 23.5 6.09 1979-80 10.63 10.71 17.6 7.12 1980-81 10.59 12.36 13.2 5.49 1981-82 9.35 10.00 11.4 6.72 1982-83 5.70 4.48 25.3 8.14 1983-84 9.44 8.56 11.8 3.89 1984-85 4.75 7.46 12.6 7.79 1985-86 4.87 5.3a 14.7a 6.90 a. Annualized growth rate, July-March 1985-86. Source: Pakistan Economic Survey, 1985-86, Finance Division, Government of Pakistan, p. 38. Monetary demand was contained despite growing pressures on the budget. The exchange rate depreciation, through its effect on public sector imports and subsidies on commodities with regulated prices, contributed substantially to an increase in the fiscal deficit-from 5.3 percent of GDP in 1981-82 to an average of 7.1 percent of GDP in 1982-83 to 1985-86 (see Table 3.5). At the time of the delinking decision, Mr. Kazi had calculated that the net rise in government expenditures would be Rsl,000 million for a depreciation of one rupee against the dollar. (This included the additional rupee cost of servicing the foreign debt.) Current government expenditure, the largest component of the fiscal deficit, increased from 14.4 percent of GDP in 1981-82 to 18.5 percent of GDP in 1985-86. Table 3.5 Pakistan: Fiscal Situation (Rs. millions) -------- Expenditures ---------- ------- Revenue ----- --------------------- Financing (net) --------------------- Overall External Domestic Bankin Years Current Development Total Tax Totala deficit (net) (non-bank) system Total 1977-78 25,545 15,353 40,898 21,585 27,005 13,893 6,129 2,817 4,947 13,893 1978-79 30,500 18,494 48,994 25,093 31,676 17,318 6,711 2,108 8,502 17,321 1979-80 35,547 19,082 54,629 32,507 39,966 14,663 6,951 1,407 6,305 14,663 1980-81 40,318 23,321 63,639 38,846 49,021 14,618 6,977 5,286 2,355 14,618 1981-82 46,370 24,643 71,013 43,010 53,846 17,167 5,345 6,313 5,516 17,174 1982-83 59,686 27,435 87,121 49,029 61,467 25,654 5,162 14,368 6,124 25,654 1983-84 73,411 26,591 100,002 53,646 74,855 25,147 5,001 12,280 7,866 25,147 1984-85 85,270 31,549 116,819 55,963 80,042 36,777 5,169 12,873 18,735 36,777 1985-86 97,767 35,127 132,894 61,169 93,447 39,447 9,008 25,682 4,757 39,447 ur As percentage of GDP 1977-78 14.5 8.7 23.2 12.2 15.3 7.9 3.5 1.6 2.8 7.9 1978-79 15.6 9.5 25.1 12.9 16.2 8.9 3.4 1.1 4.4 8.9 1979-80 15.2 8.1 23.3 13.9 17.0 6.3 3.0 0.6 2.7 6.3 1980-81 14.5 8.4 22.9 14.0 17.6 5.3 2.5 1.9 0.8 5.3 1981-82 14.4 7.7 22.1 13.4 16.7 5.3 1.7 2.0 1.7 5.3 1982-83 16.5 7.6 24.1 13.5 17.0 7.1 1.4 4.0 1.7 7.1 1983-84 17.6 6.4 23.9 12.8 17.9 6.0 1.2 2.9 1.9 6.0 1984-85 18.1 6.7 24.9 11.9 17.0 7.8 1.1 2.7 4.0 7.8 1985-86 18.5 6.7 25.2 11.6 17.7 7.5 1.7 4.9 0.9 7.5 a. Includes tax and non tax revenues, as well as surplus of autonomous bodies. b. Is equivalent to budgetary support in domestic financing. Sources: 1. Pakistan Economic Survey, 1985-86, Finance Division, Government of Pakistan. 2. Pakistan Economic Survey, 1985-86 (Supplement Edition), Finance Division, Government of Pakistan. 54 Javed Hamid and Naved Hamid The government managed to lessen the inflationary impact of this increase in the fiscal deficit by financing most of it through nonbank borrowing. For this purpose it introduced financial instruments targeted at both individual small savers and holders of large amounts of "black" money. To prevent government borrowing from crowding out the private sector, monetary policy was essentially conducted through an elaborate system of credit plans. These allowed the achievement of price stability without squeezing credit to key sectors and affecting the growth momentum of the economy. Prices and Subsidies While the regulated prices of imported products such as POL, edible oil, chemical fertilizers, and wheat provided a buffer between the exchange rate depreciation and prices faced by domestic consumers (and thus helped to keep inflation low after the delinking), they also generated pressures on the government budget. Though initially the government managed to finance its additional expenditures through its innovative savings schemes and bond issues, the budgetary pressures rose as the rupee continued to depreciate. This forced a review of the pricing of key subsidized commodities. So far, the prices of sugar, edible oils, nitrogenous fertilizers, and wheat have been deregulated. Such a move would not have been possible without an exchange rate close to its equilibrium level. Trade and Payments When in 1983-84 the dollar continued to appreciate against all other major currencies, the government did not adjust the exchange rate fully, wishing to limit the effects on the domestic economy, but it took steps to maintain export competitiveness by increasing the coverage and rates of export rebates. In 1985, after gaining experience of successfully managing the exchange rate and with the opening provided by the slide in the dollar, the exchange rate began to be used as a policy instrument to allow the elimination of export rebates by making exports more competitive in price. When the dollar weakened in 1985, the government again did not adjust the exchange rate fully but used the opportunity to reduce and finally eliminate the export rebates. Because of budgetary pressures, however, the delinking of the rupee and its subsequent depreciation was not accompanied by a reduction in import duties. In fact, surcharges were levied during the period. This resulted in increased protection for domestic industry and raised the costs of raw materials and intermediate goods, increasing the import-substitution bias in the industrial incentives system. The prime objective of the delinking decision had been to maintain export growth by preventing erosion of international competitiveness. But despite efforts at export promotion and the competitiveness of the exchange rate, total exports in 1985-86 were no higher in dollar terms than in 1980-81. A determinant of the relative profitability of exporting and import substitution is the ratio of the effective exchange rates for imports and exports. This ratio was 1.41 in favor of exports in 1981; by 1984-85 it had risen to 1.51. (World Bank Pakistan Economic Memorandum, January 1987). Pakistan's increasing bias toward import substitution is an important reason for the failure of the exchange rate policy to promote export growth. One of the financial instruments used to mobilize budgetary resources was the Foreign Exchange Bearer Bonds. These bonds carry an attractive rate of interest and though denominated in rupees they can be redeemed in foreign exchange at the prevailing exchange rate on presentation at any Pakistani bank in the world. They are freely tradeable on the Pakistani stock exchange and have resulted in making the rupee almost fully "convertible" at a premium of less than 10 percent. With an exchange rate close to equilibrium and convertibility ensured through these bonds, policy makers have had an opportunity to see that capital flight is not a major problem and are Exchange Rate Reform in Pakistan, 1982 55 considering simplifying and reducing rules and regulations governing foreign trade and foreign exchange transactions. Assessment When the Pakistan rupee was delinked, concerns were voiced about the technical sophistication required for a managed float, the chronic weakness and inconvertibility of the rupee, and the extreme instability of international currency markets. By five years later, it was generally accepted that the exchange rate had been extremely well managed and that the decision to delink had been fundamental to the good performance of the economy in general and the balance of payments in particular. Some of the factors responsible for this were internal and some external. Internal Factors The problem of exchange rate appreciation and its potential effect on export competitiveness was noticed at a very early stage by a key economic policy maker (Mr. Kazi). The implementation was entrusted to a single person who enjoyed the confidence of the President and the Finance Minister, but had full authority to take decisions, and managed the exchange rate in a pragmatic way. The overvaluation of the rupee in 1981 was not very large (22 percent) and thus the initial adjustment required was small enough to be accommodated without having to make immediate drastic changes in other economic policies (viz. subsidies, budget) and economic variables (prices). Since the decision to adjust the exchange rate in 1982 was taken at a time when there was no immediate balance of payments crisis, it did not have to be accompanied by emergency deflationary measures. Inflation was kept at a low level of about 6 percent a year in spite of the exchange rate depreciation and the consequent effects on the prices of imports. The economy continued to grow strongly (6 percent a year) and the negative income and distributional effects that normally accompany a devaluation were avoided. The government responded flexibly to the pressures generated by exchange rate adjustment. For example, (a) in response to the uncertainty in payments for foreign trade and borrowings created by the floating of the rupee, the government introduced a number of measures to provide foreign exchange risk cover for these payments; (b) in response to the pressures on the budget, the prices of a number of key commodities (sugar, edible oil, fertilizers, and cement) were deregulated. External Factors Oil prices collapsed in 1985. This reduced the import cost of petroleum products for Pakistan by about US$300 million in 1985-86, and is likely to save US$800 in 1986-87. These savings on POL products have not been passed on to consumers but used to reduce the pressures on the budget (debt repayments, defense and other public sector imports, subsidies) that partly resulted from the adjustments in the exchange rate. Pressure on the budget and the balance of payments was also reduced by a dramatic fall in the international price of edible oils. Without these windfall gains, the fall of the rupee against the basket of currencies (with the fall in the US dollar) of 26 percent in 1985-86 would have inflicted severe domestic hardship (and therefore might not have been carried out to this extent). In 1985 the US dollar began its sharp decline against other major international currencies. Since in Pakistan it is the rupee-dollar rate that is the focus of public attention, the government could allow the rupee to depreciate against the basket of currencies while maintaining a fixed parity against the declining dollar without attracting too much attention. 56 Javed Hamid and Naved Hamid Maintenance of the exchange rate at close to its equilibrium level has had the beneficial effect of revealing the hidden subsidies and distortions in the economy. This has already led the government to take steps to correct some of these distortions (deregulation) and begin to focus attention on others (tariff reform). The realistic exchange rate also gives the government the opportunity to move toward a more liberal import regime without fear of pressure on the balance of payments or threat to efficient domestic industry. 4 PRIVATIZATION OF STATE ENTERPRISES IN BANGLADESH, 1975-84 Tawfiq E. Chowdhury This chapter examines the transfer of state assets to the private sector that has been taking place in Bangladesh since the mid-1970s. Industry, nationalized at Independence, had been performing very poorly. Levels of output, capacity utilization, productivity, and financial returns were so low as to raise serious questions about the strategy of industrial development through state enterprises. A violent change of government in 1975 led to a significant shift in political and economic strategy, towards the promotion of a mixed economy with emphasis on the private sector. The privatization of industry was initiated and pushed forward by the military leadership, which had to distance itself from the preceding government in order to justify its presence. Privatization proceeded on two distinct fronts: first, beginning earlier, but still continuing today, the divestiture of enterprises that had been taken over by the state after being abandoned at Independence; and second, denationalization, which was a one-shot exercise undertaken in 1982. Implementation of the first enhanced the feasibility of the second. The privatization program had a snowballing effect. One reform created the need for another, and as we kept learning from the process, other reforms were initiated to sustain the change. Historical Background Nationalization Bangladesh became independent in December 1971 through nine months of bloody war. The process had taken a heavy toll on life and property, and the immediate imperatives at Independence were to establish a civil administration and to rehabilitate the economy. The political leadership had not really designed a comprehensive strategy to manage the economy after Independence, but it was soon faced with no alternative but to take over the management of industrial units and financial institutions. Forty-seven percent of the country's fixed assets in the industrial sector had been abandoned by their non-local owners, who had left the country, and the state had to act quickly to take care of these assets. In January 1972 it passed the Abandoned Properties Ordinance, vesting their control in the government. This gave the government control of 725 industrial units. The political leadership had to address the ideological issue arising out of the management of such abandoned enterprises. Within about three months it had decided in favor of a much broader nationalization. Together with all abandoned assets with values of Tk. 1.5 million or more, the entire jute, textile, and sugar industries, irrespective of ownership, were nationalized by Presidential Order No. 27 of 26 March, 1972. Jute is the principal export product of Bangladesh, accounting for about half of the country's export earnings and 10 percent of its employment. The textile industry, though it catered to domestic demand, was the largest single sector after jute. The nationalization of industrial enterprises increased the state's ownership from 34 percent to 92 percent of modern industry. The government created 10 sector corporations to control, 57 58 Tawfiq E. Chowdhury supervise, manage and coordinate different groups of industries. Further, all banks were nationalized and reconstituted into six banks by Presidential Order No. 26 of the same date. This expansion of state ownership took place under a political leadership that had traditionally followed a middle-of-the-road policy. It had neither the ideological conviction nor the cadre nor the organizational capacity to oversee the implementation politically. The process was thus void of intrinsic political content and support. The owners whose assets were nationalized resented their loss, even though they did not express themselves through an organized political platform. They informally mobilized support for their interests and also withdrew from active participation in opportunities offered to them, as reflected in the low level of investment in the private sector in the years that followed. In the same year, the Constitution of Bangladesh was promulgated. Although this gave the country a multi-party parliamentary system, the Constitution strengthened the basis for state ownership, since it incorporated socialism as one of its basic principles. The Constitution also enjoined that state-owned assets could only be disposed of with the consent of a two-thirds majority in Parliament. The contradictions between the constitutional injunctions and the country's political capabilities were later to pose serious difficulties for nationalized industry and finance. Poor Performance, Policy Adjustments The outcome of state ownership and management of industrial enterprises was in general unsuccessful. In the period up to 1975, capacity utilization was low, state-owned industries suffered losses in consecutive years, and the index of industrial production did not show much improvement. In 1972-76 industrial output virtually stagnated, although there were increases in the stock of capital and labor; this suggests that factor productivity declined. Many of the state enterprises showed continuing losses. For example, in 1972-73, five of the sector corporations showed losses totaling Tk. 382 million; the five others showed a combined profit of only Tk. 64 million. The net loss in 1973 on account of public sector operations amounted to about 9 percent of the industrial value added in that year. The corporations' own calculations at the time predicted an accumulated loss of Tk. 1,740 million during the First Plan period (1973-78). The nationalization of industries, banks, and also insurance companies had resulted in a loss of investors' confidence as to the security of private investment. Government policy toward the private sector had been ambivalent, and the response to it negligible. The industrial investment policy announced in January 1973 imposed a ceiling of Tk. 2.5 million on fixed investment by private firms. It gave private firms a tax holiday provided that 60 percent of the tax-exempt profits were ploughed back into the firm or invested in government bonds, and it announced a moratorium on further nationalization for 10 years. Any nationalization that took place after the 10-year moratorium would be compensated fairly. To allow the government to retain control over foreign investment in Bangladesh, foreign investment was permitted with government equity participation of 51 percent. Some months later (August 1973) these provisions were revised. The previous ceiling on private investment was raised to Tk. 30 million. Foreign investment was allowed in collaboration with both private entrepreneurs and government. The moratorium on nationalization was extended to 15 years. Unfortunately, this effort to woo the private sector lacked credibility when most of the modern sector (including concerns formerly owned by Bangladeshis) remained under state control. External agencies had voiced criticism of the industrialization strategy and the poor performance of nationalized industry. For example, a World Bank report (of September 15, 1974) recommended that "further steps [to restore confidence in the private sector] would be to denationalize units below a certain size and sell or lease them to individuals or cooperatives, mainly to permit the public sector to operate larger units more efficiently, also to boost confidence further." Privatization of State Enterprises in Bangladesh, 1975-84 59 Setting the Stage for Privatization With the change of government in 1975 came major shifts in political and economic policies. The coups, counter-coups, and political uncertainty that followed for several months may have reflected the political realignments that were taking place in the country. Once firmly in power, the new government's character and policies were reflected in changes brought about in the Constitution and other legal frameworks, and in the executive decisions that supported them. One of the objectives of the new government was to redirect the economy away from the socialist pattern toward a mixed economy with a greater role for the private sector. The reasons for this may be summed up as follows: * There had not been any explicit consensus building on socialism as a fundamental state principle; * There were no ideologically indoctrinated political and managerial cadres to implement socialist objectives; * The movement away from the multi-party parliamentary form of government, which may imply a pluralistic politics, to a one-party presidential form of government, did not develop through gradual consensus building; it was imposed; * The aggregate performance of the state-owned industrial sector had not improved much over time; • The government was burdened with the accumulated losses of the state-owned enterprises; * Because of the overwhelming dominance of the public sector, active private sector participation was ruled out. The new government seemed to have realized that to encourage the participation of the private sector it would have to restore confidence and at the same time provide greater incentives. Although it made no policy declaration on denationalization, it was preparing the ground in three directions: a greater role for the private sector; the privatization of assets that had been taken over by the state after being abandoned by their former owners-we shall term this process divestiture; and the removal of legal difficulties for eventual privatization of assets that had been nationalized irrespective of their ownership (e.g., jute and textile mills and banks)-we shall refer to this as denationalization. Of the two last, divestiture was easier because it was, in essence, an administrative decision, while denationalization needed far more substantial reforms involving amendments to the Constitution. The main features of the new investment policy introduced in December 1975 were: * to raise the ceiling on fixed investment by private firms to Tk. 100 million; * to reduce the number of industries reserved for public investment, so as to allow public and private firms simultaneously; * to pursue nationalization no further; * to permit industrial finance bodies to participate in, or provide equity and loan capital to, private sector industries; * to reactivate the stock exchange, closed since 1972; * to permit people who disclosed unutilized funds to use them for new investments or to purchase abandoned enterprises without being asked any questions; * to divest the state of those enterprises originally taken over as abandoned by their owners. The investment policy declaration of 1975 thus set the stage for privatization through the divestiture of abandoned industrial enterprises. Even before the formal declaration of policy in 1975, 159 small units had been sold through public tender. During the period 1975-82, 326 60 Tawfiq E. Chowdhury abandoned units were sold to the private sector. Most of these units were small, older firms, usually operating at a loss and representing only a small share of national output. Further Policy Changes To help clear the legal obstacles for denationalization, in 1977 the government issued a Martial Law Proclamation Order which later became part of the Constitution. This effected the following changes in the Constitution: The principles of nationalism, socialism, democracy and secularism, together with the principles derived from them as set out in this part, shall constitute the fundamental principles of state policy. was amended to read: The principles of absolute trust and faith in the Almighty Allah, nationalism, democracy, and socialism meaning economic and social justice, together with the principles derived from them as set out in this part, shall constitute the fundamental principles of state policy. This emendation, among others, diluted the concept of socialism as a basis for the conduct of policy and paved the way for a mixed economy. However, Article 47 of the Constitution, then still in force, strictly limited the procedure for denationalization and subsequent transfer of any assets, because it stipulated that any such transfer would need the consent of a two-thirds majority of the Parliament. The Martial Law Proclamation Order of 1977 also amended this Article, to allow denationalization to be undertaken within the framework of ordinary law. To promote and protect foreign private investment, the Foreign Private Investment (Promotion and Protection) Act was passed in 1980. This accorded fair and equitable treatment to foreign private investment, which would enjoy full protection and security in Bangladesh. Foreign private investment would not be expropriated or nationalized except for a public purpose and against expeditious payment of adequate compensation which would be freely transferable. Repatriation of capital returns including retained earnings would be granted, except for deferment in circumstances of exceptional financial and economic difficulty. Meanwhile, in the financial sector, in line with its emphasis on the growth of the private sector in manufacturing, the government constituted a Cabinet Committee headed by the Vice President in 1980 as the highest body to give approval for setting up privately owned banks. The Denationalization Decision Although the government during 1975-82 had laid the groundwork for denationalization, a formal declaration of such intent was still to be made. One may speculate that the government had lost the steam to carry the privatization policy to its logical end. The new military government that came to power in 1982 inherited the problems of the industrial sector exacerbated by two years of economy-wide recession. After increasing by 6 percent a year during 1976-81, industrial value added grew by only 1.1 percent in 1982. Capacity utilization declined in major industries, particularly jute and textiles. Six public sector corporations made an aggregate loss of Tk. 328 million in 1981. Private investment had suffered a decline. New Industrial Policy, 1982 To give continuity to past economic policies and to reassure the private sector, the new government lost no time in announcing the New Industrial Policy (NIP) of June 1982. The NIP laid out a strategy for accelerating industrial investment through trade and industrial policy; export promotion and development; investment incentives and financing; pricing policy; and Privatization of State Enterprises in Bangladesh, 1975-84 61 adjustments in the relative role of the private and public sector in manufacturing and in the modalities for foreign private investment. With respect to privatization, the NIP was generally guided by the reasons and circumstances of the divestiture that was started earlier. It stated: In order to create a favorable investment climate of confidence in the minds of prospective entrepreneurs, it was decided in principle to return the jute and textile mills owned by only Bangladeshis to them on the same basis as before Independence. The modus operandi for return of the units, including verification of the Bangladeshi owners, will need to be worked out. In doing so it [must] be assured that the sale of jute goods in foreign markets [is] not adversely affected. The NIP thus finally made denationalization part of the announced policy of the government. This was the culmination of the process of privatization that started in 1975. The decision gave the needed signal of the policy posture of the new government and helped to restore the confidence of the private sector. Although in principle a major policy shift could have evoked all kinds of socio-political reactions, these were obviated by the military leadership administering the reform firmly and speedily under martial law. If new banks were to be set up in the private sector, and if the shares of Bangladeshis in the jute and textile industry were to be transferred back to them, why, it was asked, should the shares of banks formerly owned by Bangladeshis remain nationalized? Partly in response to the need for policy consistency, a task force was constituted, headed by the Finance Minister in 1982, to give its recommendations on the denationalization of banks. The decision was formally announced by the Finance Minister in his budget speech of June 1983. Figure 4.1 gives a schematic presentation of the policy reform sequence. Three general points emerge: * Although political changes were crucial in shaping the reform, the discrete policy changes during the two periods all formed part of a gradual process of policy reforms in the same direction. • Each phase of reform was begun with an administrative declaration which was followed by adjustment in the legal framework to make the reform legally tenable. * The reforms were undertaken by military governments immediately on coming to power. The Decision-making Process The following general observations can be made about the decision-making process. First, the information accumulating over time, on stagnation in production, losses by the public sector, low capacity utilization, and so forth, clearly called for policy reform. This information was being continuously analyzed by the administrative agencies both in the corporations and the government. Moreover, the decline in industry was visibly reflected in reduced contributions to the exchequer, increased losses, and bigger state subsidies. The government had tried various options to enhance the autonomy and commercial freedom of state corporations and the enterprises they managed, but without much success. The policy response to this accumulated information was influenced by the assumption that the bureaucracy cannot respond speedily and flexibly to changing business conditions, whereas the private sector can. This faith, coupled with the overwhelming evidence of the inefficiency of the private sector, did not call for further analysis of information to evaluate policy options. Second, the need for such a reform was heightened by the shift in economic philosophy of the government. The nationalization decision of 1982 was derived from the New Industrial Policy and at the same time gave specific content to that policy. Third, the earlier efforts to promote the private sector had not elicited much improvement in private industrial investment. There was thus a need for a further policy to boost the private sector's confidence. 62 Tawfiq E. Chowdhury Figure 4.1 Policy Reform Sequence Toward Denationalization Type of decision Policy sequence Type of government Industrial Policy 1975: Policy declaration to divest abandoned units Military Execufive (~~~~small) and subsequentgoemn implementation Martial Law Proclamnations 1977: amendment of constitutional provisions Military Legal relating to socialism and government transfer of state-owned assets I 1 ~~~~~~govemment | New Industrial Policy and Formal declaration to Executive denationalize jute and textiles (1982) and banks = (1983) X Amendments to P.O. 26 and | Legal P.O. 27 in pursuance of the g denationalization policy 1982/1983 Ministry of Finance Ministry of Industries toMdenatistyofnalize bato denationalize jute to denationalize banks adtxie and textiles Legal and ~~~Continuation of priva- Miitr an. useun Executiv tn ization policy - new civliangvrn mdusqent Execueve ~~~~~~~~dir-ection ihagoemn Fourth, the reform proposals were announced as an administrative decision, so as to test their acceptability while keeping the option open to rescind them if necessary. Once they were perceived to be feasible, the legal framework was established to give a clear basis for their implementation. Fifth, denationalization followed divestiture, as two distinct processes. A solution to intervening problems was worked out with the aim of putting the denationalization speedily in place. The trade-off between prolonged analysis and deliberations (which could have led to aborting the reform) and speedy implementation of a less-than-perfect reform package was in the minds of decision-makers. Privatization of State Enterprises in Bangladesh, 1975-84 63 Sixth, although the reform was ultimately of a political nature, the existence of martial law obviated the need for detailed political debate. The national leadership acted on formal and informal recommendations, suggestions, and advice of bureaucratic institutions as well as of individuals working in them. Seventh, the domestic reform was carried out in an international environment favoring such changes. The international agencies had been pointing out the inefficiency of public sector enterprises and their consequent drag on economic growth. Being an aid-dependent country, Bangladesh had to pay serious attention to these observations. Implementation Denationalization of Jute and Textile Mills The government published a formal notice of the terms and conditions for denationalization of jute and textile mills on September 27, 1982. This invited applications to buy shares within a 21-day period. It spelled out the following conditions: 1. Identification of mills formerly owned by Bangladeshis. The mills eligible for such divestiture were as follows: mills with more than half of their shares formerly owned by Bangladeshis; mills with more than half of their shares belonging to Bangladeshis and financial institutions together; mills with not less than 30 percent of their shares belonging to Bangladeshis but whose share holders would like to buy further shares, to bring their ownership up to 51 percent. (The reference date for the above was the date of nationalization.) 2. Terms and conditions of transfer. Once the mills had been identified, using the above criteria, government offered to sell the shares on the following terms and conditions: the price of the shares would be the same as originally paid to the owners in compensation; those owning less than 50 percent would have to buy shares to own up to 51 percent, and these shares would be sold at a revalued price. The payment would be. in cash, or 50 percent in cash and the rest to be paid within 12 months, backed by guarantee. All other shares were also up for sale. The Bangladeshi owner of 52 percent would have the first option to buy, followed by other private parties and financial institutions. All assets and liabilities, including those of banks, financial institutions, etc., would stand transferred to the denationalized companies on the existing terms and conditions. government's investments would be repaid over a period of 12/15 years (with a moratorium of 2 years), the rate of interest remaining the same. The denationalized companies would take over all the employees, none of whom could be dismissed before the expiry of one year from the date of the transfer. If the denationalized company defaulted on its contractual obligations, or did not operate the mills for any reason within its control, the government would have the right to intervene in the affairs of the company. To keep exports of jute stable, a Pricing Committee composed of representatives of Bangladesh Jute Mills Corporation (BJMC), denationalized companies, and the Bangladesh Bank was set up, in the proportion 3:2:1. All sales of jute under barter/tender to foreign governments and local sales to government or corporations would continue to be negotiated by the BJMC and allocated to the concerned mills on the basis of their capacity. Common expenses in connection with such sales and allied functions would be shared proportionately by BJMC and denationalized mills. Details would be worked out by the Pricing Committee. To reinforce these efforts to safeguard exports of jute goods, the government promulgated the Jute Goods Trading Order, 1982. However, the private sector mills considered the Order as undue interference, and it was rescinded two years later. Once the political will was clearly demonstrated the denationalization was carried out speedily by a working group in the Ministry of Industries. (see Figure 4.2). This efficient implementation was possible because: 64 Tawfiq E. Chowdhury Figure 4.2 Denationalization of Jute and Textile Mills-Implementation* Inhouse working group in the M10 Industries with Corporation representatives-1982 Working group recommendations put up to president and chief martial law administrator (CMLA) Negotiation with previous owners Referred again to CMLA Agreement reached with former owners Transfer of shares through detail legal documentation and executive procedure-1983 * This was basically a one-shot exercise. * the Minister of Industries was a former civil servant with long and diverse experience, with a reputation for being a "doer;' * senior officials of the Ministry were professional civil servants trained to implement the decision; * the chiefs of the two corporations (Jute Mills Corporation and Textile Mills Corporation) whose enterprises were to be nationalized were cadre-service officials working on deputation and as such did not have any vested interest. Protests by other interest groups were discouraged by the existence of martial law. It should nonetheless be noted that prolonged negotiations took place between the government and the former mill owners, as to the terms and conditions of transfer, and particularly the liabilities of Privatization of State Enterprises in Bangladesh, 1975-84 65 the mills. The former owners were not disposed to accept the liabilities that the companies accrued during the period of nationalization. Interventions at the level of Minister and finally, the President, who was the Chief Martial Law Administrator, were needed to make them agree to take over all the liabilities on the existing terms and conditions. The owners of the mills kept contending this issue years after the denationalization. The bulk of the denationalization of jute and textile mills was completed within a year. Today, 27 out of 68 textile mills, and 35 out of 68 jute mills, have been transferred back to the private sector (see Tables 4.1 and 4.2). Denationalization of Banks The denationalization of banks was carried out in the shadow of the overall policy decisions with respect to private sector participation and denationalization of jute and textile mills. Of the six nationalized commercial banks only two (Pubali Bank and Uttara Bank) had had registered head offices in Bangladesh at the time of Independence. A Task Force in the Ministry of Finance, headed by the Finance Minister along with representatives of the Central Bank and the Chief Executives of Pubali and Uttara Banks Ltd., oversaw the process of implementation. (See Figure 4.3.) It discussed whether to sell 51 percent of the shares of these banks to their previous owners, with the state retaining 49 percent; to sell Table 4.1 Textile Industry, 1985-86* -------- Production ---------- No. of Yarn Cloth Employment Type mills (million lbs. ) (million yds.) (thousands of people) Nationalized 41 61 38 42 (60) (56) (58) (58) Denationalized 27 47 27 30 (40) (44) (42) (42) Total 68 108 65 72 (100) (100) (100) (100) * Figures in parenthesis are percentages of the total. Table 4.2 Jute Industry, 1985-86* No. of Production Employment Export Type mills ('000 mt) ('000 people) ('000 mt) Nationalized 33 275 161 279 (49) (60) (65) (60) Denationalized 35 183 86 186 (41) (40) (35) (40) Total 68 458 247 465 (100) (100) (100) (100) * Figures in parenthesis are percentages of the total. Note: The denationalized mills are generally of smaller size. 66 Tawfiq E. Chowdhury Figure 4.3 Denationalization of Banks-Implementation* Cabinet committee headed by vice-president set up for allowing banks in the private sector-1980 Task force headed by finance minister to denationalize banks-1982 Recommendation of task force to cabinet committee now headed by chief martial law administrator: modalities for denationalization approved-1983 Formal declaration of denationalization by the finance minister - 1983 Amendment of law and preparation of other legal documents Assets and liabilities transferred to new banking company through a vendor's agreement Shares floated for subscription by earlier owners and public * This was one-shot exercise. Privatization of State Enterprises in Bangladesh, 1975-84 67 Figure 4.4 Divestiture of Abandoned Units--Implementation* Minister for industry approves units to be taken up for divestment Industry profile prepared for public tendering and also the notional reserve price is fixed I11 Invitation of bid through public notice Bids evaluated by interministerial working group and recommendations made, Disinvestment board (interministerial) headed by minister for industries gives final decision * This is an ongoing process. all the shares to the general public; or to sell shares to their former owners and the remainder to the public (the last was the option eventually adopted). The proposals of the Task Force were presented to the President and were by and large approved and carried out. The Cabinet Committee had a lot of confidence in the Task Force, who handled the subject very professionally. Pubali Bank and Uttara Bank were reconstituted as public limited companies and their assets and liabilities were revalued at market prices. Revalued shares of the banks were open for purchase by the public. Former Bangladeshi shareholders were given preference to buy shares proportionate to their previous holdings. The rest of the shares were to be sold to the general public (including former shareholders), with preference given to wage earners. Former shareholders were allowed compensation to the extent of the earlier face value of the shares, which would be adjusted against the payments they made to purchase shares of the reconstituted banks. The banks would have to pay immediately to the government 50 percent of the sale price of shares. The other 50 percent could remain with the two banks as a loan from the government, bearing an annual interest rate of 14 percent, repayable in five years in equal annual installments. No appropriation of profits or declaration of dividends could be made until the instalment had been paid. 68 Tawfiq E. Chowdhury The existing employees of the two banks would be retained and their services would continue to be governed by the regulations already in force, which could not be varied to their disadvantage. Divestiture of Abandoned Enterprises The divestiture of abandoned units was continued under the New Industrial Policy. The broad sequence of implementation that followed the formal declaration of 1975 is shown in Figure 4.4. The process was complicated because of the large number of units, which were managed by various authorities such as the sector corporations, Director General of Industries, and local bodies, depending on their size and location. Moreover, the abandoned units did not all have the same legal status, being either sole proprietary, partnership, or public/private limited companies. This affected the choice of procedures for divestiture. Further, some of the units were partly abandoned, while others were fully abandoned in the context of existing laws of title and ownership. Generally speaking, if these were companies, then shares were sold. In other cases, the assets were sold but the purchaser had to undertake the long-term liabilities of the enterprise and government assumed all the current liabilities in excess of current assets. The procedure for divestiture was, and still is, as follows: * The Minister for Industries initially approves the units to be divested. i An enterprise profile of the unit is prepared for the guidance of potential buyers, and a notional reserve price fixed. i Public tenders are floated and publicized through newspapers and through embassies abroad. * Prospective buyers may visit and inspect the enterprises. * On the appointed date tenders are opened publicly in the presence of buyers; they are then evaluated by the Disinvestment Board. * Claims by former owners or shareholders are examined by a Scrutiny Committee who in turn places its findings before the Working Group. * The Disinvestment Board in its formal meeting gives the final decision which, among others, includes the decision to sell or invite new tenders. 3 The buyer has to make 20 percent of the sale price as a downpayment and the rest in three equal annual installments with a moratorium for two years. In case of default, 7 percent penal interest is charged. In such a situation the government has the option to take over the unit and forfeit the payments made by the buyer. The title is transferred after the payment of the total sale price. Nearly 500 abandoned units have been privatized so far. The process of divestiture is still going on. Unlike the denationalization exercise, which was a one-shot job, the divestiture of abandoned units has been modified over time to incorporate gains from experience. Because of its continuing nature there are permanent institutional layers for implementation. This has slowed down the process. Possibly the diversity of abandoned units and the continuity of the process call for intensive broadly based consultation to ensure, improve, and sustain implementation. Assessing the Reform: Looking Back and Forward What has been the performance of industry since privatization? Two studies were undertaken by the Bangladesh Institute of Development Studies-one relating to the profile and performance of mills that had been privatized and the other a comparative analysis of jute and textile mills after denationalization. Unfortunately, both may be too deficient in data and methodology to lead to any meaningful findings. The first study concluded that "the trends in Privatization of State Enterprises in Bangladesh, 1975-84 69 the improvement in production and sales are mixed whilst the capacity to improve the financial performance of enterprises is if anything more negative than positive..." The paper raised the question of the concentration of wealth through the divestiture process. The other study collected information, although covering only a short period, on production, capital productivity, material productivity, and financial performance of jute and textile mills in the public and private sector. Its findings are inconclusive. The following issues that have emerged from the process of denationalization and divestiture could be used to improve upon such policy in future. • Although the new owners of denationalized jute and textile mills agreed to assume the liabilities to banks and financial institutions, later on they kept contesting these liabilities in court. An independent committee had to be set up to examine some of the associated accounting disputes. Most of the mills have since come to a reasonable settlement with respect to their long-term loan liabilities with banks and financial institutions, though in the process the banks have had to waive some of their accumulated interest on these accounts. * Because of the depreciation of the domestic currency, loan liabilities denominated in foreign exchange had been ballooning, seriously affecting the viability of some enterprises that had borrowed heavily abroad. Government introduced a scheme in 1983 through which some of the past burden arising out of the exchange risk was reduced and future exchange risks were eliminated, in lieu of paying an exchange risk premium. This scheme, the EFAS, may need more fine tuning to enhance its effectiveness. * Many of the new owners of abandoned units defaulted on their installment payments. About 40 percent of such installments have become overdue. The penal interest of about 7 percent has been a disincentive for regular payment since money in the bank can earn a much higher return. There are discussions now on how to revise this condition. • While the terms of divestiture specifically stipulated that if the private owner failed to make the agreed payments, the unit would revert to the state, the denationalization procedure had an unspecific clause allowing for government intervention. In situations of default on payment, this clause has not been legally enforceable because of its vagueness. * The contract laws of Bangladesh are more than 70 years old. Their deficiency, coupled with a slow judicial process, have made them ineffective. Given that the private sector is coming to play a larger role, the relevant laws need to be modernized. The government has already finished drafting a new law. * The substantive law relating to banking business is also over 20 years old and not adequately framed to deal with banks in the private sector. The government is in the process of drafting a new law for regulation of banks. * The overall lack of financial discipline in Bangladesh has encouraged noncompliance with contractual obligations. Compounding this attitude has been the underdeveloped nature of the institutions responsible for enforcing such obligations and, in particular, for collecting the dues owed to the government or its agencies. Experience led to the appointment of a Commission on Money, Banking, and Credit to report on the financial sector. New measures to enforce financial and credit discipline have begun to show results. * With the enhanced role of the private sector, the fiscal, monetary, and institutional structure needed realignment to favor competition and remove market distortions. A study was commissioned with the assistance of Harvard Institute for International Development to recommend reforms in trade and industry. The Industrial Policy of 1986 and Budget of 1987-88 are important milestones in this direction. Recently, the privatization process has taken a new direction. Government has designed and started implementing the partial privatization of fully-owned state enterprises by the open sale of shares to the public, where employees have a reserved quota and would find a place in the boardroom. 70 Tawfiq E. Chowdhury The privatization of state-owned enterprises is but one significant step toward the establishment of a competitive market economy. The efficient functioning of such an economy can be constrained by distortions in markets both at home and abroad. The full test of privatization and the market economy in Bangladesh can only be achieved under reasonable competitive conditions. The country is now on this transition path. With all the inadequacies of a market economy in its formative stage during the period 1978-83, besides large-scale privatization, annual private sector industrial investment increased fivefold, contributing nearly half of domestic industrial investment. This is a good beginning. 5 TAX REFORM IN INDONESIA, 1984 R. Mansury Ismail Tamsir Aware of the weak potential of the tax system in generating revenues and the looming threat of declining oil prices, the Indonesian Government undertook a comprehensive Tax Reform. The reform was enacted in a series of laws in 1984-85, at the end of the Third Five-Year Development Plan, Repelita III. This paper analyzes the reasons for the reform, the process of its conception and its passing into law, and its main features. The last section offers an assessment of the results of the reform thus far. We look first at developments in the five year period leading up to the reform. Historical Background Government Finance In fiscal year 1981/82 the government drew 70 percent of its revenues from taxes on oil and gas, which it relied upon to cover 65 percent of public expenditures. While the proceeds from these taxes consistently rose between 1974/75 and 1981/82, tax revenue from the rest of the economy was consistently decreasing in relation to GDP. Taxes on incomes elsewhere in the economy contributed only 13 percent of revenues; taxes on domestic consumption supplied about 8 percent, and so did taxes on imports and exports. (see Table 5.1). In the period 1979-84 oil and natural gas accounted for nearly 21 percent of Indonesia's output and 76 percent of its export earnings. Signs of the Dutch Disease had been particularly prevalent in 1974-78, marked by accelerated inflation, rising unemployment, primarily in agriculture, which produced traditional export products, and stagnating or declining exports of non-oil and gas products, in the midst of rising foreign exchange earnings from the petroleum sector. Enthralled by the oil bonanza, the government paid relatively little attention to developing other exports or other sources of revenue for its own budget. High government revenues from the corporate income tax on oil had made it possible to launch an ambitious development plan, but left its financing very vulnerable to changes in the oil price (see Table 5.2). Between fiscal 1979/80 and 1983/84, government revenues rose at an annual rate of 21 percent-well above the inflation rate-and they more than doubled during Repelita III. When the second oil boom reached its peak in 1981, government expenditures amounted to nearly 26 percent of GDP. Government revenue from oil had reached 16 percent of GDP, while non-oil revenue, never more than 9 percent of GDP in the last two decades, stood at under 7 percent of GDP, equivalent to less than one fourth of total government expenditures. The remaining portion, of 3 percent, came from foreign aid (see Table 5.3). By 1982, tax revenues from the non-oil sector had fallen to 6.6 percent of GDP-only twice the share of foreign aid in GDP. (See Table 5.3.) This ratio compared very unfavorably with the average figure for low income countries at that time, of 17 percent of GDP. (See Table 5.4 for selected Asian countries.) 71 72 R. Mansutry and Ismail Tamsir Table 5.1 Indonesia Central Government Receipts, 1979/80-1983/84 (Rp. billion) 1979/80 1980/81 1981/82 1982/83 1983/84 Total Percent Taxes on income 5,129.3 8,230.3 10,100.3 10,009.9 11,650.1 45,074.9 80.4 Income tax 148.1 164.2 207.2 288.8 398.8 Corporate tax 297.1 447.6 559.1 674.5 757.4 Corporate tax on oil 4,259.6 7,019.6 8,627.8 8,170.4 9,520.2 37,597.6 67.1 Withholding tax 291.2 433.5 513.0 641.9 628.1 IPEDA/Property Taxa 71.4 87.2 94.5 105.2 132.4 Others 61.8 78.2 98.7 129.1 168.2 Taxes on domestic consumption 537.2 732.9 888.0 1,137.4 1,392.1 4,687.6 8.4 Sales/value added tax 192.2 265.6 310.7 476.6 575.2 Excises 326.4 437.9 544.2 620.1 773.2 Other oil revenuesb - - - - - Miscellaneous levies 18.6 29.4 33.1 40.7 43.7 Taxes on international trade 843.0 948.1 887.9 835.4 916.0 4,430.4 7.9 Import duties 316.7 448.0 536.2 521.9 557.0 Sales tax on imports 137.2 195.1 223.3 231.0 255.0 Export tax 389.1 305.0 128.4 82.5 104.0 Nontax receipt 187.3 315.7 336.4 435.6 519.5 1,794.5 3.2 Domestic revenue 6,696.8 10,277.0 12,212.6 12,418.3 14,432.7 56,037.4 100.0 Development funds 1,381.1 1,493.8 1,709.0 1,940.0 3,882.4 Counterpart fundsc 64.8 64.1 45.1 15.1 14.9 Project aid 1,316.3 1,429.7 1,663.9 1,924.9 3,867.5 Total revenue 8,077.9 11,720.8 13,921.6 14,358.3 18,315.1 a. Since January 1986, IPEDA replaced by Property Tax. b. Oil subsidies shown as Government expenditures from 1977/78. c. Program aid. d. Includes commercial bank and suppliers' credits for development programs/projects. Source: Ministry of Finance. Tax Reform in Indonesia, 1984 73 Table 5.2 Indonesia's Oil Export Prices, 1969-86 (US$/bbl) January 1969 1.67 January 1980 27.50 April 1971 2.21 February 1980 29.50 October 1971 2.60 May 1980 31.50 April 1972 2.96 January 1981 35.00 April 1973 3.73 November 1982 34.53 October 1973 4.75 February 1983 29.53 November 1973 6.00 February 1985 28.53 January 1974 10.80 January 1986 25.13 April 1974 11.70 February 1986 21.00 July 1974 12.60 March 1986 14.45 October 1975 12.80 April 1986 10.66 January 1977 13.55 May 1986 10.38 January 1977 13.90 June 1986 12.11 April 1977 15.65 July 1986 10.25 May 1979 16.15 August 1986 9.83 June 1979 18.25 September 1986 12.20 July 1979 21.12 November 1979 23.50 December 1979 25.50 Source: Bank of Indonesia. Table 5.3 Government Expenditure, Tax Revenues from Oil and Non-oil Sector and Foreign Aid in Percent of GDP-Indonesia 1968-1983 Government ------------ Tax revenues ----------- expenditure Oil and gas in Non-oil and in percent percent of gas in percent Foreign aid in Year of GDP GDP of GDP percent of GDP 1968 8.8 1.2 6.0 1.6 1969 12.3 2.4 6.6 3.3 1970 14.1 3.1 7.6 3.4 1971 14.8 3.8 7.8 2.4 1972 16.1 5.1 7.9 3.1 1973 17.2 5.7 8.7 2.8 1974 18.5 8.9 7.4 2.2 1975 21.6 9.9 7.9 3.8 1976 23.9 10.6 8.2 5.1 1977 22.6 10.2 8.4 4.0 1978 23.3 10.2 8.6 4.5 1979 25.2 13.3 7.6 4.3 1980 26.4 15.8 7.2 3.4 1981 25.8 16.0 6.6 3.2 1982 24.1 13.7 7.1 3.3 1983 23.8a 14.4a 6.8a 266a a. Estimated. Source: Ministry of Finance. 74 R. Mansury and Ismail Tamsir Table 5.4 Tax Efforts in Selected Asian Countries (percentage of GDP) External Average for Total Nonoil Income Consumption trade Country fiscal years taxes taxes taxes taxes taxes Low income India 1982-84 10.9 10.9 25 5.3 3.1 Pakistan 1982-83 12.7 12.7 2.6 5.2 5.0 Lower-middle income Indonesia 1983-86 19.7 6.3 2.7 2.5 0.8 Philippines 1980-83 10.4 10.4 2.5 4.8 2.9 Thailand 1981-84 13.0 13.0 2.9 6.6 3.2 Upper-middle income Malaysia 1982-84 22.2 19.0 7.2 5.2 6.2 Republic of Korea 1982-84 16.6 16.6 4.3 8.5 2.8 Source: World Bank staff estimates. Deficiencies in the Tax System Under the old tax law, then in force, the petroleum sector was the only one able to generate revenues effectively. Oil prices were high, but there was also a special auditing agency and the sector was not eligible for tax incentives that might have reduced its tax burden. Deficiencies in the tax system were well known to Indonesia's policy makers and the subject of constant criticism by international donor organizations. With the start of the first oil boom in 1973, however, efforts made in the early 1970s to overhaul the system came to little-the new government (in power since 1966) had brought inflation under control and domestic revenues had more than doubled in real terms without major changes in the tax system. Thoughts of a comprehensive tax reform, in the air again when oil prices declined in 1975- 78, were similarly shelved with the arrival of the second oil boom, which started in 1978 and reached its peak in 1981-revenues from the oil and gas sector provided the government with enough funds for its development program, notwithstanding the fall in non-oil tax revenues to a low of 6.6 percent of GDP. Unfortunately, the heavy and sudden inflow of funds from the second oil boom could not be absorbed without inflation and disruptions. The fiscal policy options that were put forward in 1981 had been carefully considered. If inflation, which greatly hampered development, was to be prevented, the government could drastically cut its budget or undertake fundamental tax reform. Moreover, there was serious concern in some quarters about the vulnerability to oil prices. Should these decline, the funding of Repelita IV, the next development plan, would have to depend largely on slowly-growing revenues from income and sales tax. Import tariffs could not be raised by much, because Indonesia's industrial sector depended on imported raw materials and capital goods. Export revenues were likely to develop only slowly. The Minister of Finance decided in January 1981 to prepare for a tax reform. In 1981, oil prices reached a high of $35 a barrel. Oil and gas export prices were forecast to continue rising in real terms by 3 percent a year until the end of 1985. Thus at the time the tax reform was proposed there was no fiscal crisis, in the sense of a large deficit which could not be overcome, nor any general economic crisis. It was argued that with oil prices so high, the right time had come for tax reform, for if tax revenues from other sources declined in the initial stages of the reform, the drop would be readily offset by revenues from oil and gas. And, as it turned out, no serious political difficulties resulted from proposing tax reform at a time when the state budget was showing a surplus resulting from high oil prices. Tax Reform in Indonesia, 1984 75 Existing Tax Structures and Administration Most of Indonesia's tax laws dated back to the Dutch Colonial period. The hundreds of ad hoc amendments that had been made to them were too superficial to serve the need for increasing development funds. The old tax legislation was aimed less at generating revenue than at influencing incentives. Examples included incentives for firms to go public, and to use public accountants in preparing their annual reports, incentives for projects in priority areas, and incentives for the development of cooperatives. Studies showed these to be of dubious value in promoting efficiency and economic growth. The system of incentives also created difficult problems for the tax administration. The old system had, moreover, become so complicated and so open to evasion that it could not provide any guarantee that a target level of revenues could be achieved. The vagueness of its provisions was a heavy burden to the tax administration as well as to the taxpayer. Tax collection procedures were so poor that they easily led to embezzlements. Some of the sanctions imposed for failure to pay taxes were so unrealistic that they were seldom applied, while others were so light that they failed to function as deterrents. A simplification of the tax system was imperative not only to guarantee revenues but also to support equity among tax payers and to simplify tax administration. Summary of Factors Leading to Reform • The dangers of heavy dependence on oil began to be widely recognized. Since it was not possible to predict how long the oil boom would last, it was felt necessary to have an alternative source of income for the development program. • Lending countries and institutions put increasing pressure on Indonesia to raise tax revenues from the nonoil sector. It was hoped that this, coupled with the continuing increase in revenues from the oil and gas sector, would help reduce the heavy dependence on foreign aid. • Domestic earnings from oil and gas in 1981 were very high, so that the temporary fall in tax revenues from other sources, likely to accompany the reform process, could be offset from the oil sector. - There was a high degree of political stability, necessary for carrying out a contentious reform without serious political reverberations. * There was increasing evidence that the ad hoc amendments made before 1980 and aimed at increasing tax revenue from the nonoil sector were not achieving the expected results. Initiation of Reform When the reform was conceived it was decided to spend two and a half years on technical studies as a basis for the design of the new system. As in many other developing countries, expatriate experts were asked to assist with the design of the reform. A team from the Harvard Institute of International Development was assisted by experts from Canada, Colombia, the Netherlands, and West Germany. Indonesian law experts and tax specialists, together with expatriate law experts from the Harvard Institute for International Development, were put in charge of preparing a new tax policy and tax law, formulated together. The time used from the start of the technical studies to the final drafting of the Tax Bill was 33 months. The decision-makers adopted the method used in Colombia: proposals were discussed with decision-makers as they were formulated, and the revisions sent back for approval. A Steering Committee was established, whose members were officials of the Department of Finance and officials of the Directorate General of Taxation. Its task was to 76 R. Mansury and Ismail Tamsir control, monitor, and assist the expatriate technical team and to analyze the specific policy options proposed. Colombia's experience had shown that the acceptance or adoption of a tax reform plan depends a great deal on the involvement of local officials and academics in its formulation. In contrast with Colombia, however, it was decided to give little publicity to the reform in its first two years of preparation. This was to prevent protracted debates with the private sector, which might have delayed the progress of the reform. It was also decided that the results of the preparatory work would not be presented as a formal report. The chief aim of the team preparing the reform was structural reform. Aspects of procedure, administration, and implementation were largely considered later. However, the period used to prepare the reform was also used to formulate rules on the education, both formal and informal, of tax administrators. The government decided to create a cadre of well-trained tax administrators to implement the new system in due time. Sizeable funds were allocated for sending young tax administrators abroad for study. Within six months, 50 of them had been sent abroad for degree and non-degree programs, and by 1984 many of them had returned with advanced degrees. Not long after the technical studies were begun, it was decided to set up a computerized tax information system which would start operating in 1984 as part of the new tax regime. The Decision-Making Process Before the tax bill was presented to the Dewan Perwakilan Rakyat (the House of People's Representatives), it was reviewed in the Department of Finance. During this review the senior tax officials, as people with practical experience, contributed important feedback and suggestions. Other reviews were made by the Chamber of Commerce and by prominent tax experts. After this review, the bill was submitted to the DPR with an introductory speech by the Finance Minister. The DPR then formed a work committee, representing its various political groupings, to review the bill, article by article, with representatives of the Department of Finance, Department of Justice, and Cabinet Secretariat. Once the DPR had approved the bill it was forwarded to the President for ratification. The Reformed System Aims The aims of the reform were to: * create a tax system to stimulate the growth of domestic revenues from outside the petroleum sector; * create a more effective tax apparatus by simplifying the tax structure, and raise taxpayer compliance by lowering tax rates and applying strict enforcement; * create a more equitable tax system by broadening the tax base and maintaining progressive tax rates; * provide certainty by formulating clearer definitions or limitations in areas where disputes commonly arose between tax officials and tax payers; * plug the loopholes for tax abuse and tax evasion, for example by not allowing fringe benefits to be tax deductible, subjecting small businesses to tax on a deemed-profit basis, creating provisions for transactions between affiliated companies, adopting a taxpayers' identification system, providing education to tax payers and tax officials, and using a computerized tax information system. None of the old tax laws was retained. Tax Reform in Indonesia, 1984 77 Main Features The reform of the system was accomplished by a series of tax laws enacted between January 1984 and the end of 1985: Law No. 6 of 1984 on General Tax Provisions and Procedures, enacted simultaneously with Law No 7 of that year, on Income Tax; Law No. 8 of 1983, which became effective in April 1985, on Value Added Tax and Luxury Sales Tax; and Law No 12 of 1985 on Land and Building Tax. The income tax law applies both to individuals and business firms. For greater simplicity and equity, income from all sources is consolidated and taxed at the same rate. This applies to both firms and individuals. Income is broadly defined to include "any increase in the economic prosperity received or accrued" by the taxpayer. All income is now, for any given level, taxed at the same rate, thereby eliminating the former practice of taxing different types of income at different rates. Income tax rates have been lowered to a maximum marginal rate of 35 percent applied to both individuals and firms, instead of 50 percent for individuals and 45 percent for firms under the old law. Tax rates are low, but there are relatively few exemptions. The tax incentives valid under the old system have largely been eliminated. Even though tax rates would be lower than before, the simplification was expected to yield higher revenues. The new system is progressive: it effectively excludes the poorest 90 percent of the population from tax obligations. On the other hand, higher taxes are now imposed on big companies and industries with high incomes. The new income tax law introduces the system of self-assessment in determining the tax payer's tax obligation, and emphasizes the withholding of tax at source. Under the former system, official assessment had created several administrative problems. It placed a heavy work load on the tax administration; it necessitated personal contacts between tax official and taxpayer, which usually led to collusion; and it forced the tax payer to depend on the issuance of an assessment to determine his tax liability. Under the new system no assessment will be made by the tax authority, provided that the tax liability as stated in the tax return has been computed in accordance with the tax provisions. The new tax incentives for businesses, foreign and domestic, featured lower rates of tax and higher depreciation allowances. The lowering of the maximum corporate tax rate from 45 percent to 35 percent was expected to save firms as much, in present value terms, as the tax holidays they would have received under the old system. The new law is believed to contain many elements attractive to domestic and foreign investors. The law on Value Added Tax (VAT) took some time to become effective, while the apparatus was prepared to execute it and taxpayers were prepared to comply with it. For VAT to be administered successfully, taxpayers need to keep accurate records of their transactions. Thus in the period from January 1984 through March 1985, the Tax department informed and taught the taxpaying public about the new taxes and the need to keep accurate records. It also developed a system of control to deal with the possible recurrence of tax evasion. VAT is payable by anyone manufacturing taxable goods or acting as a main distributor of imported manufactured goods. The agricultural sector, wholesalers, retailers, and services other than construction are not liable for VAT, but the law gives the Minister of Finance authority to broaden the statutory tax base, for example by extending the VAT to the retail level, without additional legislative approval. The revenue potential of the VAT was greatly increased by applying it to domestic sales of petroleum and tobacco products. VAT is expected to become a major source of revenue. Luxury sales taxes, currently at rates of 10 and 20 percent, are collected simultaneously with VAT, and can if necessary be raised to 35 percent by government regulation. No tax is imposed on exports of luxury goods. New and simpler provisions for a land and building tax became effective in January 1985. Currently set at a low rate, this tax has a good revenue potential provided it is supported by a well-prepared and well-run tax administration. 78 R. Mansury and Ismail Tamsir Implementation Tax Administration In the period of transition from official assessment to self-assessment, the government granted a tax amnesty: all activities relating to tax assessment and tax verification in 1984 were suspended until the tax amnesty period expired on June 30, 1985. A crucial preparatory step in the implementation of the new tax law, the amnesty was intended to encourage taxpayers to report the true state of their financial condition, their assets and their business, while it offered the opportunity for final assessments of all past tax liabilities. The amnesty applied to all individuals and establishments, whether or not they had previously registered as taxpayers. Amnesty was given for taxes that had never been or had only partially been levied or collected. Taxes for which amnesty was requested were subject to a ransom payment: those who had filed their tax returns for FY83 and FY84 paid one percent of the assets used to compute the tax on which amnesty was requested, while those who had not filed their returns paid 10 percent. After the enactment of the new tax law, the government requested help from the Fiscal Affairs Department of the International Monetary Fund to improve the structural organization and technique of the tax administration. Internal reforms in the Directorate General of Taxation were designed to improve: * organizational structure (still in progress); * methods of operation, including the introduction of new regulations for control; * quality of tax officials; * quality of service to the public. These reforms were based on a growing realization that the provision of good services to taxpayers will increase their tax compliance. The task of setting up a taxpayer ID system and a computerized tax information system was accomplished much faster than had been anticipated. Taxpayers' registers were reorganized; computers were installed in district tax offices, regional offices, and head office. Computerization will facilitate accurate record keeping, auditing, and cross checking of withheld amounts vs credits claimed. Continuing efforts are being made to increase the professional competence of the officials of the Directorate, through appropriate training programs, domestically as well as abroad. Efforts are also being made to promote dedication, motivation, and a constructive mental attitude, so as to nurture the attitude of serving society to the best of one's ability. In-depth explanations have been given to all tax officials about the difference between the new tax law and the old, seen from both the philosophical and technical sides, to help them identify with and act according to the spirit of the new law. A new policy was developed for control and for strict discipline of any officials who violate regulations or disciplinary rules. Tax information centers have been set up in district tax offices, with a fixed schedule of information sessions. In this way, tax officials not only act as sources of information but also play a role as civil servants charged with making a success of national development through taxation. Taxpayers The heavy burden of taxation during the colonial period, when taxes were used for the benefit of the rulers and not of the people, left effects that are still felt today. Continuous efforts are needed to instill the conviction that paying taxes is a duty inseparable from the democratic right of every citizen to participate in the development of the nation. Tax Reform in Indonesia, 1984 79 An extensive information campaign has been undertaken with these principles in mind. It includes education of school and university students, by tax officials; establishment of a public relations office of the Directorate General of Taxation; establishment of a Campaign Work Group that conducts seminars and workshops on the workings of the new tax system; opening of taxpayer service units in district tax offices, as noted above; publications explaining the provisions of the new laws, and a magazine called Weekly Tax News, published by the DGT; and guide-books, stickers, and slogans which are put up in places where they can easily be seen by the public. These various forms of media dissemination play an important role not only in promoting understanding of the new tax laws, but also in practical matters (helping taxpayers file income tax returns using the correct forms, for example). Tax information is also carried on the radio and television and in national newspapers. To promote taxpayers' compliance it is imperative that taxes be used properly, to finance the execution of government functions and development programs whose benefits can be enjoyed by the people. If the feeling prevails that tax revenues are not being properly used, the compliance of taxpayers will be hard to maintain. It is hoped that the new system will create a sense of fairness and promote correctness, since under the self-assessment system it is the taxpayer himself who knows best how much he should pay. In implementing the new tax legislation the government is hoping that the openness and honesty of the tax-paying public will continue to increase. It has to keep in mind that the role of tax revenue in national development is still relatively small and so is the share of the population paying taxes. Effects More Taxpayers After the enactment of the new law the Directorate General of Taxation took steps to increase the number of taxpayers. The results are shown below: -------------Income tax---------- ------------------ VAT----------- Period Individual Corporate Individual Corporate December 1984 431.287 n.a. n.a. 18.696 December 1985 536.769 130.016 5.783 30.152 December 1986 639.500 155.718 11.427 57.477 The number of businesses subject to VAT declined under the new provisions. The overall rise in the number of taxpayers cannot be viewed separately from the efforts of the Directorate General of Taxation to familiarize the general public with the new tax laws. The tax amnesty, which expired in June 1985, also greatly contributed to the rise in the number of registered taxpayers. Toward the end of the amnesty period, many citizens registered themselves as taxpayers for the first time. Increase in Tax Revenues Tables 5.5-5.8 compare revenues before and after the enactment of the new tax law. (These tables reflect the tax returns lodged by taxpayers. No audits or investigations had yet been carried out to determine their correctness.) Though the data stop at the end of FY85/86, they show a rapid response to the reforms. As a proportion of domestic revenues, non-oil tax revenues increased from 18 percent in 1982/83 to 27 percent in 1985/86, and as a proportion of GDP, from 80 R. Mansury and Ismail Tamsir Table 5.5 Items Old tax system New tax system Depreciation Straight line method Straight line method for building Declining balance method for non-building Rate 5-20 percent Rate 5-50 percent Carry forward (a) 4 years (a) 5 years of losses (b) Unlimited for losses (b) 8 years for specific suffered within the first types of businesses 6 years of establishment Tax rates General tax rates General tax rates Corporate income tax: Corporate income tax: up to Rp 25M 25% up to Rp 10M 15% over RP 25-Rp 50M 30% over Rp 10-Rp 50M 25% over Rp 50 45% over Rp 50M 35% Individual income tax: 5-50% Tax refund No time limit for finalization Finalization within 12 months after application Tax objection No time limit for finalization Finalization within 12 months after the filing of objection Tax collection Official assessment Self-assessment 3.8 percent in 1982/83 to 5.2 percent in 1985/86, according to Ministry of Finance figures (see Table 5.6). Table 5.8, based on Ministry of Finance and World Bank staff estimates, suggests there was an even greater rise in the non-oil tax revenues as a share of GDP, to 6.8 percent for 1985/86. Estimates for 1986/87 show a further rise in this share, to 8.2 percent. These figures compare with a fall in the oil and gas tax revenue from 14.8 percent of GDP in 1981/82 to 10.8 percent in 1985/86. The VAT system in particular, according to the World Bank, is a most effective revenue system. Even though the new legislation on VAT did not go into effect until April 1986, in the period 1984/85 to 1985/86 revenues from VAT increased by 260 percent, while revenue from income tax rose only 9 percent. Based on World Bank estimates, revenue from VAT increased from 0.9 percent of GDP in fiscal 1981/82 to 2.9 percent in 1986/87. The increase is mainly attributable to the imposition of VAT on petroleum and tobacco products. Tax Reform in Indonesia, 1984 81 Table 5.6 Tax Collection in Relation to Total Domestic Revenues and GDP (Rp. billion, at current market prices) Fiscal Tax collected Total domestic Percent Percent year by DGT revenues (2: 3) GDP (2: 5) (1) (2) (3) (4) (5) (6) 1969/70 59.4 243.7 24.4 2,718.0 2.2 1970/71 73.9 344.6 21.4 3,238.0 2.3 1971/72 98.7 428.0 23.1 3,672.0 2.7 1972/73 134.1 590.6 22.7 4,564.0 2.9 1973/74 209.3 967.7 21.6 6,753.4 3.1 1974/75 334.1 1,753.7 19.1 10,708.0 3.1 1975/76 454.6 2,241.9 20.3 12,642.5 3.6 1976/77 567.1 2,906.0 19.5 15,466.7 3.7 1977/78 711.9 3,535.4 20.1 19,033.0 3.7 1978/79 838.9 4,266.1 19.7 22,746.0 3.7 1979/80 1,041.5 6,696.8 15.6 32,025.4 3.3 1980/81 1,439.5 10,227.0 14.1 45,445.7 3.2 1981/82 1,770.0 12,212.6 14.5 54,027.0 3.3 1982/83 2,248.6 12,418.3 18.1 59,632.6 3.8 1983/84 2,742.2 14,432.7 19.0 73,697.6 3.7 1984/85 3,053.7 15,905.5 19.2 85,914.4 3.6 1985/86 5,015.4 18,674.5 26.9 96,066.4 5.2 Source: Ministry of Finance. Revenue from personal income tax also showed an important increase in 1986/87, rising by 22 percent a year in nominal terms from the level attained in fiscal year 1983/84. This increase came largely from wage earners taxed under the new withholding system. Tax revenue from the self-employed remains low. Revenue from corporate income tax did not increase markedly up to 1985/86, but showed some improvement in fiscal 1986/87. The relatively slow increase in income tax revenues reflects the reduction in income tax rates; the revision of the original income tax base, so that some tax payers were now subject to VAT but no longer to income tax; and a slackening in business and industry outside the petroleum sector. The Way Ahead Revenues from direct and indirect taxes can be expected to play an increasingly important role in the Indonesian economy. In the future, it will be imperative to audit enough taxpayers often enough-no self-assessment system will work without continuous control and supervision. Despite the great improvements to date, DGT still faces problems, the most difficult of which is noncompliance. This is partly reflected in the low filing of tax returns. For example, in January 1987 only 39 percent of the taxpayers registered for VAT filed returns, while the filing ratios for personal and corporate income tax were 85 percent and 55 percent respectively. Another element of noncompliance is the underreporting of tax liabilities for both VAT and income tax. Table 5.7 Collection by Types of Tax in Relation to Total Domestic Revenue (fiscal years) ---------1982-1983-------- -------- 1983-1984 -------- -------- 1984-1985 -------- ------- 1985-1986 ----- Tax Tax Tax Tax Tax Tax Tax Tax collected domestic collected domestic collected domestic collected domestic Type of tax by DGT revenue 2:3 by DGT revenue 5:6 by DGT revenue 8:9 by DGT revenue 1 2 3 4 5 6 7 8 9 10 11 12 Income tax (individual & corporate) 1,706.5 1,932.3 2,121.0 2,313.0 Sales tax/ VAT 476.6 575.2 637.2 2,326.7 Other tax revenue 68.3 64.3 138.4 208.3 Property tax 105.2 132.4 157.1 167.4 Total 2,248.6 12,418.3 18.1 2,742.2 14,432.7 19.0 3,053.7 15,905.4 19.0 5,015.5 18,674.5 Tax Reform in Indonesia, 1984 83 Table 5.8 Trends in Central Government Revenues (percent of GDP) 1974/75 1978/79 1981/82 1983/84 1985/86 1986187a OilILNG taxes 8.6 9.6 14.8 14.1 10.8 6.0 Non-oil taxes 6.3 7.4 5.6 6.0 6.8 8.2 On incomeb 2.2 2.7 2.5 2.6 2.6 2.9 Personal tax (0.4) (0.5) (0.4) (0.5) (0.7) (0.7) Corporate tax (0.8) (0.9) (1.0) (1.0) (1.7) (2.0) Withholding taxC (0.7) (1.0) (0.9) (0.9) Property tax (0.3) (0.3) (0.2) (0.2) (0.2) (0.2) On consumption 2.0 2.5 1.8 2.2 3.4 4.0 Sales tax/VAT (1.4) (1.4) (0.9) (1.1) (2.4) (2.9) Excise tax (0.6) (1.1) (0.9) (1.1) (1.0) (1.1) On external trade 2.0 1.9 1.1 0.9 0.7 1.1 Import duty (1.4) (1.2) (0.9) (0.8) (0.6) (1.0) Export tax (0.6) (0.7) (0.2) (0.1) (0.1) (0.1) Miscellaneous 0.1 0.3 0.2 0.3 0.2 0.2 Non-tax revenues 0.6 0.8 0.6 0.7 1.6 2.1d Total domestic revenues 15.5 17.8 20.9 20.8 19.2 16.3 a. Estimated. b. Unadjusted for restitution of income tax payments from earlier years. c. Withholding tax was abolished as a separate category after the tax reform. Most withholding tax revenues are from the corporate sector. d. Includes domestic oil surplus. Sources: Ministry of Finance and World Bank staff estimates. Government measures to overcome these problems include computerization of tax information, more emphasis on field audits, staff training and increasing the number of tax offices, so as to aid collection. So far not many audits have been undertaken of VAT or income tax, for lack of trained staff. To step up tax audits, about 850 auditors are being trained at present for both VAT and income tax. 84 R. Mansury and Ismail Tamsir Table 5.9 Revenue Impact of Tax Reform Before reform After reform 1974/75 - 1983/84 1984/85 - 1986/87 Percent of GDP" Non-oil taxes 6.0 8.2 On income 2.6 2.9 On consumption 2.2 4.0 On external trade 0.9 1.1 Real growth (percent per annum)b Non-oil taxes 5.9 16.7 On income 8.8 9.0 On consumption 6.5 29.4 On external trade -1.2 10.0 Tax buoyancyC Non-oil taxes 0.9 2.1 On income 1.1 1.4 On consumption 1.0 3.3 On external trade 0.6 1.5 a. End of period. b. Tax revenues deflated by the Consumer Price Index. c. Using non-oil GDP as the tax base. Sources: Ministry of Finance and World Bank staff estimates. Table 5.10 Indonesia: Central Government Budget Realization Summary, 1979/80-1983/84 (Rp. billion) Annual growth 1979/80 1980/81 1981/82 1982/83 1983/84 rate (%) 1. Domestic revenues 6,696.8 10,227.0 12,212.6 12,418.3 14,432.7 21.1 2. Routine expendituresa 4,061.8 5,800.0 6,977.6 6,996.3 8,411.8 20.0 3. Government savings (1-2) 2,635.0 4,427.0 5,235.0 5,422.0 6,020.9 229 4. Development expenditure 4,014.2 5,916.1 6,940.0 7,359.6 9,899.2 25.3 5. Balance (3-4) -1,379.2 -1,489.1 -1,705.0 -1,937.6 -3,878.3 29.5 Finance by: 6. Counterpart fundsb 64.8 64.1 45.1 15.1 14.9 7. Project aid 1,316.3 1,429.7 1,663.9 1,924.9 3,867.5 8. Change in balances -1.9 -4.7 -4.0 -2.4 -4.1 (- increase) a. Includes debt service payments. b. Program aid. Source: Ministry of Finance. Tax Reform in Indonesia, 1984 85 Table 5.11 Indonesia: Central Government Expenditures, 1979/80-1983/84 (Rp. billion) 1979/80 1980/81 1981/82 1982/83 1983/84 Total Percent Personnel expenditures 1,419.9 2,023.3 2,277.7 2,418.1 2,757.0 10,896.0 33.8 Wages and salaries 1,053.9 1,482.9 1,660.4 1,749.0 1,996.0 Rice allowance 179.9 252.0 253.3 289.9 346.1 Food allowance 109.9 193.2 241.0 254.9 261.3 Others 47.1 61.2 80.0 78.6 87.6 External 29.1 34.0 43.0 45.7 66.0 Material expenditures 569.0 670.6 922.4 1,041.2 1,057.1 4,260.3 13.2 Domestic 539.6 637.8 890.5 1,007.4 1,007.0 External 29A 32.8 31.9 33.8 50.1 Subsidies to regions 669.9 976.1 1,209.4 1,315.4 1,546.9 5,717.7 17.7 Irian Jaya 25.0 33.9 42.0 43.0 41.5 Other regions 644.9 942.2 1,167.4 1,272.4 1,505.4 Debt service payments 684.1 784.8 931.0 1,224.5 2,102.7 5,727.1 17.8 Internal 36.5 30.8 16.0 19.8 29.8 External 647.6 754.0 915.0 1,204.7 2,072.9 Other expenditures 718.9 1,345.2 1,637.1 997.1 948.1 3,646.4 17.8 Food subsidy 124.9 281.6 224.0 1.0 - Oil subsidy 534.9 1,022.0 1,316.0 962.0 928.1 Others 59 1a 41.5 97.1b 34.1 20.0 Routine expenditures 4,061.8 5,800.0 6,977.6 6,996.3 8,411.8 32,247.5 100.0 Development expenditures 4,014.2 5,916.1 6,940.0 7,359.6 9,899.2 Total expenditures 8,076.0 11,716.1 13,917.6 14,355.9 18,311.0 a. Includes PERTAMINA subsidy (Rp. 48.1 billion). b. Includes general election (Rp. 81.0 billion). Source: Ministry of Finance. The government has also initiated a number of technical assistance schemes to strengthen tax planning. For example, with the help of German experts, a system is being developed to assist managers in controlling tax payments and assessing tax potential. A training program for the heads of district offices is being organized under the supervision of American IRS experts. To secure future successes, a number of areas should be focused on: • Making the best use of tax information that has been collected. Cooperation between DGT and other agencies, public as well as private, should be enhanced for this purpose. For example, noncompliance can be significantly reduced by cross-checking information on taxpayers. Data from VAT returns may be used to indicate the likely level of corporate income tax due. The cross-checking of data on recipients of government payments against those on registered taxpayers will also help to identify those recipients who ought to register as taxpayers. * Income tax collection can be improved by imposing stiff penalties for nonregistration and nonfilers. * The problem of underreporting can be reduced by effective field audits on a selective basis. To ensure effective audits, a system for selecting tax returns for audit should be developed. * Training programs should not be neglected. With adequate progress in these areas, according to World Bank forecasts, it should be possible to raise nonoil tax revenues from 8 percent of GDP in 1986/87 to 9-10 percent of GDP in the next two to three years. 6 TRADE LIBERALIZATION POLICY IN THE REPUBLIC OF KOREA, 1980-85 Ohm Young-Suk In 1967, when it joined GATT, Korea took a major step toward import liberalization by changing from a positive to a negative import licensing system-in which only import products required to have approval were listed-and made efforts to reduce tariffs. Its subsequent efforts to liberalize trade were negligible until 1978, when the government opened markets for a considerable number of foreign products. After this, a surplus in the current account and changes in the international economic environment reduced the impetus for liberalization somewhat, but in the early 1980s the Korean government has made remarkable progress in trade liberalization. In the mid 1970s, half of the products Korea imported were subject to restrictions, but by 1986 this proportion had fallen to 8.4 percent, and it is expected to fall to 5 percent by 1988. By then, most commodities except agricultural products will have completely open markets. The formulation and implementation of this change in policy provide an idea of the unique decision-making process in Korea. Traditional Korean culture is based on Confucianism, and the authoritarian bureaucratic system has a long history in Korean government. Government bureaucrats dominate decision-making and the private sector and research institutes play minor and supporting roles. This chapter addresses the following questions: * Was Korea's import liberalization program an inevitable response to changing economic conditions, or did it reflect deliberate changes in government economic policy? * What were the controversies regarding liberalization programs and how was the government able to implement liberalization programs in the face of resistance from both inside and outside its ranks? * In implementation, how effectively did different governmental organizations coordinate their work? * How well did the program fit with industrial policy? * How did the relatively rapid liberalization program affect domestic industries, and import-competing industries in particular? Historical Perspective To understand the trade liberalization that took place in the early 1980s, one must analyze the economic conditions and policy responses of the late 1970s. Korea's major export items, such as textile products and footwear, had been losing competitiveness as a result of Korea's relatively high wage levels and of market penetration by other developing countries. In response, the government had changed its policy emphasis from light and labor-intensive products to heavy and technology-intensive products. By the late 1970s, however, 86 Trade Liberalization Policy in the Republic of Korea, 1980-85 87 overinvestment in the heavy and chemical industries was adding to a heavy debt burden and causing growing inflationary pressure. Meanwhile, the slowdown in the world economy worsened the export prospects of heavy industrial products and resulted in overcapacity. These unfavorable results led to a rethinking of economic policy. During 1978-79, the findings of several reports on the need for structural adjustment in the Korean economy (notably those of the Korea Development Institute (KDI), the Bank of Korea, and the National Council on Economy and Science) were presented to the President and summarized by the Economic Planning Board (EPB). In contrast with the prevailing view of the 1970s, that the Korean economy needed continued protection, a small number of economists and government officials now began to indicate the merits of trade liberalization. They argued, first, that liberalization would induce firms to increase productivity, and second, that resources in the less competitive and protected industries would be transferred to more competitive industries. Meanwhile, for the first time since the first five-year economic plan (1962), in 1978 the current account recorded a small surplus of $12.3 million. The surplus arose mainly from the increased flow of foreign exchange from construction businesses in the Middle East and from a sharp increase in exports of goods and services. Policy makers became quite optimistic about the future course of the economy, assuming that the balance of payments outlook would continue to improve. This optimistic outlook facilitated trade liberalization; indeed, the decision taken in 1978 to liberalize imports can be seen more as a reaction to the current account surplus than as a reflection of well-formulated economic policy. The CMES EPB's introduction, in 1979, of the Comprehensive Measures for Economic Stabilization (CMES) was a turning point in the government's economic policy. The measures included price stabilization through conservative monetary and fiscal policy, and reduction and realignment of overinvested heavy industries. To curb excess liquidity, an upper limit was set for the growth rates of monetary aggregates, and extensive reforms of the banking sector were proposed. To reduce excess investment in the heavy and chemical industries, the government temporarily suspended new projects in these industries. Changes were made in the incentive systems for different industries. Trade liberalization measures were an important feature of the CMES. The sudden increased inflows of foreign exchange, mentioned above, had resulted in an oversupply of money, and subsequent inflationary pressure. The need to realign a distorted price structure and reduce inflationary pressure provided favorable conditions for liberalizing trade: import liberalization would reduce the inflationary pressure not only by slowing down the rate of increase in the money supply but also through increased pressure on local firms to cut prices of import-competing products. The CMES were introduced to alleviate structural difficulties arising from the unrealistic and overambitious heavy industrialization programs. It is important to note that the buoyant export sectors and the resultant current account surplus of the period somewhat lessened the urgency of the structural adjustment. The efforts that were made nonetheless allowed the Korean economy to cope with the worldwide recession and the political transition in 1980. The major contents of the CMES were reaffirmed by the new government under President Chun Doo Hwan, which espoused competitive free-market ideas as the basis of economic policy. In retrospect, one may say that a small number of EPB technocrats and economists at KDI were instrumental in persuading the head of the government to follow a market-oriented economic strategy. Thus, the economic policy of the new government put high priority on price stability, elimination of relative price distortions, reduction of economic concentration, and a free trade regime. Considering that high priority was placed on the free operation of the market, trade liberalization in the late 1970s was more a natural outcome of the general economic reforms than of any pressures from trading partners. 88 Ohm Young-Suk Import Liberalization, 1981-82 Unlike that of 1978-79, the import liberalization of 1981-82 was pursued in the face of growing opposition in an unfavorable economic environment. The world economy was in recession and export markets were very sluggish. Slow growth in the domestic market was responsible for a growing mood of protectionism. However, import liberalization formed part of an overall economic readjustment program to which the government was already committed: * The government was deeply involved in reducing controls in the monetary sector by releasing its commercial bank ownerships and freeing interest rates. . It had launched an industry realignment policy to reduce the preferential tax-subsidy incentive systems. * The foreign exchange rate, fixed at 484 won per dollar since 1974, had been devalued to 659 won per dollar in 1980, and was left to fluctuate according to market forces without much government intervention. Exchange rate reform greatly reduced the need to provide arbitrary incentives to export and to restrict imports to protect domestic industries. EPB and KDI publicized the economic readjustment program vigorously and held a series of meetings with bureaucrats of other ministries and economists of other research institutes and universities. EPB's strong position on import liberalization was not well accepted in some of the other ministries. A case in point was MTI, which had initiated most of the old industry- protection laws and had relatively close contacts with business people adversely affected by import liberalization. Despite strong opposition, the EPB technocrats and KDI economists managed, with the support of the President, to carry out the program as envisaged. The import liberalization measures also encountered strong opposition from the business sector, which had not been officially represented in the process of their formulation. Throughout the 1960s and 1970s, private firms had been accustomed to protection, and the bigger they were, the greater the benefits they received from the government. Although some businessmen advocated market-oriented policies in general, they were reluctant to accept lower protection for their own businesses. Even so, it has been rare in Korea for major government economic policy to be scrapped simply because of strong opposition from the private sector. The private sector was informed of the final policy measures and persuaded to cooperate. As export performance continued to improve in 1982-84, the liberalization proceeded on schedule. Import Liberalization, 1985-86 Import liberalization in the mid-1980s presented a very different picture. By then, Korea had become a surplus country against the United States. For several years before 1984, neither Korea nor the US had enjoyed a continued trade surplus, and trade negotiations between the two countries had centered mostly around the import protection of the US and the Generalized System of Preferences. By 1984, Korea's trade surplus against the US amounted to $3.6 billion; it rose to $4.3 billion in 1985, and to $7.4 billion in 1986. As a result, the American trade negotiation team made a strong request to open up Korean markets for US commodities, and went further, pressing for the liberalization of insurance, intellectual rights, and commercial banking. After nine months of trade negotiations, ending in July 1986, the Korean government accepted most of the American requests for trade liberalization. Recently, it seems that further trade liberalization is proceeding because of heavy pressures from the US. Import liberalization in Korea has mainly taken the form of simplifications in the import- licensing system and reductions in tariff rates. Means of protection still in use in Korea as contingency measures include import surveillance, an import diversification scheme, and a flexible tariff system. Import surveillance has been used, though rarely, to discourage imports of products which might disrupt domestic markets for a period of time after the products were permitted to be imported freely. Trade Liberalization Policy in the Republic of Korea, 1980-85 89 Under the import diversification scheme, the government announces a list of commodities which require approval when imported from Japan-in an effort to reduce the chronic trade imbalance against Japan. The flexible tariff system includes emergency tariff, anti-dumping provisions, and countervailing duty. The emergency tariff rate is used when it is necessary to raise tariff rates on the products which have been liberalized, and is regarded as the Korean version of safeguards. By 1988, most industries except agriculture will have completely open markets for foreign products. Foreigners will have easier access to service sectors such as banking, insurance, and recreation. Import liberalization means that hitherto-protected domestic industries become vulnerable to foreign competition. To achieve the intended results of import liberalization a package of industrial reforms had to be implemented at the same time. Hence the Korean trade liberalization measures cannot be properly evaluated except in the context of the policy packages of which they formed a part. Preferential treatment awarded to industries in their infancy during the 1960s and 1970s (e.g. through subsidies, preferential taxes, preferential bank loans, discounted public utility fees, and government subsidized industrial complexes) had continued in effect in the early 1980s, after the industries had become quite competitive. In the belief that a freer competitive environment was essential to the efficient allocation of limited resources and to correct over- investment in certain industries, the government replaced the existing protective industry promotion laws with the Manufacturing Development Law of October 1986. To give the affected firms time to adjust, the government prepares preannounced schedules of import liberalization by commodities. The government classifies the affected industries into two groups: one unable to be competitive under the new policy and the other group requiring complementary adjustments to regain competitiveness: * improving production technology and renovating facilities; or * securing larger value added by saving energy; or * eliminating bottlenecks in production and management. For the industries deemed to need complementary measures the government is designing plans to assist them in the transition period. These plans are concerned, for example, with the appropriate scale of production, quality of product, possible improvements in technology. Subsides and preferential interest rates on loans and subsidies have become much less significant than before but preferential taxes remain a major means of assistance. According to the Tax Reduction and Exemption Law, firms classified as requiring rationalization programs to improve their competitiveness will receive preferential tax treatment. In industries deemed incapable of recovering competitiveness the government is designing plans to limit new entrants, encourage firms to move into more promising businesses, and assist in disposing of existing plants. Government assistance to industry is thus much more limited now than before. Before the implementation of the Manufacturing Development Law, government assistance was specific to certain industries and remained without serious review. Current assistance is highly limited in scope and time period, and its basic philosophy is to let the affected firms take responsibility for adjustment. Overview As stressed above, trade liberalization was undertaken as part of a more general economic policy package designed to promote a more market-oriented economic system, in place of strong government-directed economic policies. Another objective of these economic reforms was to enhance the generality of economic policies. The high priority accorded export and heavy industries had brought about numerous special incentive schemes, and it was recognized that inconsistency among individual policies can ultimately reduce rather than increase economic 90 Ohm Young-Suk efficiency. To promote exports, for instance, the government had provided various incentive systems to firms in the export sector. These included preferential tax packages, lower interest rates, and subsidies. These measures were extensively used in the 1960s and 1970s. The CMES rationalized the incentive systems for different industries and the Manufacturing Development Law of 1986 superseded the several individual laws designed to protect specific industries. (Despite these efforts, particularism continues to be an important element in Korean policy formulation.) Though some of their results were unfavorable initially, the measures were able to correct a distorted price structure, stabilize prices, and improve the productivity of previously protected industries. Two different opinions exist regarding import liberalization in Korea. One holds that the US, so long as it has a trade deficit against Korea, will always ask for more trade liberalization, regardless of the Korean government's past efforts to open the Korean market, and that the liberalization programs undertaken voluntarily since 1978 have left Korea little scope for making further concessions. The other opinion holds that trade liberalization has benefited the economy by increasing the competitiveness of domestic industries and has helped to build a better Korean image for trade partners. Formulation of Economic Policy Sources of Influence What is the nature of the decision-making process for economic policy in Korea? Ever since Park Chung Hee took over the government in 1961, economic development has received the highest priority. While the major economic policies reflected the opinions of various sectors of the economy, the government has played the predominant role in decision-making. Given the authoritarian nature of the government, the role of the legislature was kept to a minimum. This was considered necessary to achieve a speedy decision-making process and flexible implementation of economic policies. The predominance of government bureaucrats in the formulation of major policies may be attributed to the historical and cultural background of Korean Confucianism and its authoritarian political system. Korean national culture is also distinguished by face-to-face personal relationships, and this fact has allowed lower-level technocrats a good deal of participation and influence within the framework of a hierarchical society. Key institutions of the Korean administration have developed along the lines of the Japanese model of management and public service. Usually policy proposals are prepared by the bureaucratic elite and, after being finalized in a ministry, are presented to the Economic Ministers' Meeting for formal discussion and approval. (Often problems that are raised in the Economic Ministers' Meeting are further discussed in the Economic Deputy Ministers' Meeting.) Opinions on major policy issues are more freely exchanged in the Economic Ministers' Roundtable. Economic policy proposals prepared in one ministry are likely to touch upon the major responsibilities and functions of others, and thus require informal approval from all ministries before being officially formulated. The Roundtable has contributed greatly to smooth cooperation in the formulation stage of major economic policies. Many important economic policy ideas are initiated by economic ministers, either in the Economic Ministers' Meeting or in the Economic Ministers' Roundtable. Lower-ranking government technocrats formulate economic policy packages, but the directions of major economic policies are usually set by high-ranking decision-makers of the government. To reduce conflicts of opinion among different ministries, a committee of lower-ranking technocrats has been extensively used in early stages of the decision-making process. Most of the major economic policies have been initiated and implemented by the EPB, which also prepares five-year economic plans. It was influential technocrats at EPB and their Trade Liberalization Policy in the Republic of Korea, 1980-85 91 counterparts at the Blue House, provided the major impetus for import liberalization. EPB's think tank, KDI, provided the necessary background support. Under a system in which all ministries have equal power, decision-making as well as implementation of policies can face difficulties when these ministries' interests conflict. The existence of a dominant ministry, responsible for coordinating the different views of other ministries, facilitates the formulation and implementation of policies. EPB has come to fill this role. Its position was strengthened when the Budget Bureau was transferred from the Ministry of Finance to EPB in 1981, and the minister of EPB concurrently holds the position of Deputy Prime Minister. The economic advisers at the Blue House have played very influential roles, and in fact the chief economic secretary to the President has often been more influential than the economic ministers in formulating major economic policies. The major changes in policy toward a market- oriented economy in the early 1980s were very strongly advocated by the chief economic secretary to the President, Kim Jae Ik, an ex-EPB technocrat who became a figure in the economic policy making process in the Fifth Republic. While the decision-making process in Korea has generally followed a top-down approach, because of the cultural and political background, the participation of lower level-technocrats cannot be ignored. As in the 1960s and 1970s, most able college students still aspire to become government officers. The high quality of government officers has been an important factor in the effectiveness of Korea's policies. The authoritarian nature of the government has permitted the role of the legislature to be kept to a minimum. This allows speedy decision-making and flexible implementation of economic policies. In formulating policies the executive has received cooperation from institutions such as research institutes, business sector organizations, and academics that is outstanding in comparison with other developing countries. Most economic ministries have drawn upon the research institutes in their fields. KDI was the most influential research institute under EPB, and was called upon to provide EPB with theoretical justification and information on which to base policies. Quite often economists at research institutes were appointed as government officers and became very influential decision-makers. Since most ministries have their own think tanks, differences of opinion among research institutes can be as strong as those among ministries. With regard to liberalization programs, EPB and MTI had significantly different approaches and their different views were reflected in the discussion papers prepared by KDI and KIET. KDI during the early 1980s was a strong advocate of trade liberalization and provided theoretical arguments in support of the EPB's position, while a paper issued by KIET, reflecting the position of MTI, took a cautious approach to trade liberalization-arguing that the hasty progress of liberalization might hurt import- competing domestic industries and that most incentive systems created in the past would continue to serve useful purposes. The Korea Rural Economic Institute (KREI) under the Ministry of Agriculture and Fisheries opposed trade liberalization in regard to agricultural products. Academics have had only limited chances to participate in policy formulation and the opinions they present at committees or meetings tend to be more general than specific. But more academics have been appointed to major posts in economic ministries than in most other countries. For instance, the Deputy Prime Minister is the former KDI president, and the chief economic secretary to the President is the former KIET president. The Minister of MTI is a former economics professor at Seoul National University. The major business organizations include the Federation of Korean Industries (FKI), the Korea Traders' Association (KTA), the Cooperative Association of Small and Medium Firms, and many individual associations representing particular industries. Resulting Characteristics Centralization in the formulation of policy is the natural result of strong commitment to economic growth by top decision-makers and the predominance of the executive branch in policy 92 Ohm Young-Suk formulation. The dominance of economic policy making by the executive has meant that policies could be quickly conceived and put into action, if necessary minimizing the influence of interest groups. In many cases, major economic policies were formulated by a small number of selected technocrats in secret. At least until recently, Korean policy makers have been much more concerned with the effectiveness of policies in achieving intended goals than with the acceptability of the policies to the various sectors of the economy. In many cases, government policies effective in accelerating economic growth were not welcomed by the majority of people. As economic activities become more complicated, the important roles of middle and small industries, agriculture, and labor unions will need to be recognized. The government cannot indefinitely ignore the opinions of the majority on the grounds that the growth-first approach, once achieved, will eventually increase the welfare of the entire people. Implementation of Economic Policy As has often been pointed out, the success of Korean economic planning has been due less to formal plans than to their flexible implementation. In the execution of five year economic plans, most emphasis is placed on the effective implementation of major projects, and though much effort goes into their preparation, the contents of plans are apt to be quickly forgotten, even by the EPB bureaucrats who prepared them. Since the economy depends heavily on export performance, changes in the world economic environment such as the oil shock and recessions in major trading partners make frequent revisions a necessary routine. Outcomes at the end of the plan period have always differed from those envisaged-and in most cases have surpassed them by a wide margin. While the actual function of the five-year economic plans has thus been to provide no more than the general direction for government economic policies, the government prepares and follows Annual Economic Management Plans, using these to reflect changes in economic conditions and to list the major projects to be undertaken during the coming year. Project- oriented implementation of policies has been highly emphasized. One of the most important reasons for the effective implementation of economic policies has been the dominant role of EPB, referred to earlier. For example, although the import liberalization program introduced in the recession years of 1981-82 was not popular, it was able to be sustained (though the pace of liberalization slowed somewhat) because it was initiated by EPB and received strong support from the political leadership. The government has sometimes intervened at the implementation stage with restrictions or bans on consumer products but more often with actions directed toward business firms. Many firms were forced to go public and subjected to price-wage controls. Foreign exchange control is an other example of coercive intervention. Some of the penalties for not following government directives involve economic loss such as the cancellation of previous benefits. So-called informal guidance has often played an important role in the implementation of policies. In a society such as Korea's where business depends heavily on government supports, informal guidelines or moral suasion can be as effective as any formal coercive intervention. To get the support and participation of the private sector, several important meetings are held in the President's office: the monthly economic review, the monthly export promotion conference, and the quarterly science and technology promotion meeting. The President chairs these meetings with high-ranking government officers and business leaders, representing such organizations as FKI and KTA and the Chamber of Commerce. In particular, the monthly export promotion conference reviews the export performance in the past month, problems preventing the smooth implementation of export policy, and import liberalization programs. Even though export promotion has been the major topic in the meetings in recent years, much attention has been given to the progress of import liberalization, particularly in view of Korea's unexpectedly large current account surplus with regard to the United States. Trade Liberalization Policy in the Republic of Korea, 1980-85 93 Results of Import Liberalization Programs Despite the partial liberalization of imports in 1967, during the 1970s tariffs in Korea remained relatively high compared with those of industrial countries. The average tariff rate, 39.1 percent in 1968, gradually declined to 29.7 percent in 1977 and to 24.8 percent in 1979, following the tariff reform of 1978, and to 21.9 percent in 1984. Progress in import liberalization can also be measured by the change in the proportion of imported products requiring no import permits. In Korea this import liberalization ratio continued to worsen during the second half of the 1960s and the first half of the 1970s and stood at the very low level of 49.9 percent in 1977. But by 1986, the ratio had improved to 91.6 percent. As Table 6.1 shows, significant import liberalization was achieved in most product groups in the first half of the 1980s. In 1981, several major industries were still considerably Table 6.1 Import Licensing Liberalization Program in Korea, 1981-881 (as of July 1 , each year) Proportion of items subject to automatic import Number of approval, under the regular Trade Notice (%) items2 1981 1982 1983 1984 1985 19863 19873 19883 Primary products, foods and beverages 1,386 68.5 70.6 73.2 75.8 78.2 79.7 80.1 80.5 Chemical products, paper and ceramics 2,182 93.4 94.0 94.4 95.0 95.6 97.7 99.1 99.6 Steel and metal products 802 88.9 89.7 90.9 92.8 95.6 99.4 100.0 100.0 General machinery 1,414 64.2 65.5 68.7 78.0 83.0 89.4 93.3 100.0 Electrical and electronic machinery 495 40.9 46.1 53.6 62.4 73.0 87.0 95.5 100.0 Textile products, including leather garments 1,089 65.4 68.4 80.4 90.3 93.1 95.1 96.9 97.8 Others 547 71.2 75.7 81.2 82.1 82.8 85.7 88.2 88.2 Total 7,915 74.7 76.6 80.4 84.8 87.7 91.6 93.6 95.4 1. Ministry of Trade and Industry, 1986. 2. At the 8-digit level under CCCN as of 1984. 3. The program for 1986-88 is an announced plan. Source: Soogil Yang, Import Liberalization and Industrial Adjustment in Korea, KDI Working Paper 8613, 1986, KDI. protected. The least liberalized industry was electrical and electronic machinery, followed by general machinery, textile products, and primary products, food and beverages. Electrical and electronic machinery made the most progress, with their import liberalization ratio increasing from 40.9 percent in 1981 to 73.0 percent in 1985, and were closely followed by textile products, (including garments), whose import liberalization ratio increased from 65.4 percent in 1981 to 93.1 percent in 1985. By 1985, only electrical and electronic machinery and primary products, foods, and beverages had import liberalization ratios below 80 percent. According to the import liberalization program, the average import liberalization ratio will reach 95.4 percent in 1988; most industries will have almost complete liberalization by that time. One sector that continues to require a certain amount of protection is primary products, food and beverages; its import liberalization ratio will remain at about 80 percent in 1988. 94 Ohm Young-Suk How were imports affected by the liberalization? Though it is difficult to distinguish the effects of the policy changes from other factors, the following analysis compares the actual imports of specific commodities after import liberalization with the import levels projected by extrapolating the trend of the previous three years. Imports of electrical and electronic machinery, shown in Table 6.2, were liberalized in three stages: in July 1982, in July 1983, and in July 1984. After the first liberalization, actual imports for the seven months following were lower than the trend estimates by $9.7 million dollars. Imports were much lower than trend values for almost all the period, and the gap widened to $44 million in the first half of 1985. As Table 6.3 shows, the broad pattern for electrical and electronic machinery prevailed for most of the commodities liberalized in the 1980s. Table 6.2 Impact of Import Liberalization-Electrical and Electronic Industry (US$ '000) 1982 7-12 1983 1-6 1983 7-12 1984 1-6 1984 7-12 1985 1-6 1982. July liberalization Household products -2,783 -401 -2,330 -537 -6,560 -1,767 Industrial products -7,958 -20,173 -7,304 -23,424 -13,419 -42,291 Total -9,741 -20,574 -9,634 -23,961 -19,979 -44,058 1983. July liberalization Household products -723 452 108 617 Industrial products 4,913 5,254 1,624 1,628 Total 4,190 5,706 1,732 2,245 1984. July liberalization Household products -1,716 -9,001 Industrial products 47,469 -28,182 Total 45,753 -37,183 Grand total -9,741 -20,574 -5,444 -18,255 27,506 -78,996 It should be borne in mind that the Korean economy in the first half of the 1980s was somewhat sluggish, and that the rate of growth of imports has been declining over a long period. In addition, the contingency measures such as import surveillance, import diversification, and emergency tariffs were effective to a certain extent in reducing imports after liberalization. Nevertheless, the statistical analysis suggests that many of the affected industries had already reached the stage where they could compete successfully with foreign firms in the domestic market. The lifting of protection forced them to adopt cost-cutting and technology- improving measures, resulting in an improvement in competitiveness. Trade Liberalization Policy in the Republic of Korea, 1980-85 95 Table 6.3 Import Trends After Import Liberalization, by Industry (US$ '000) 1982 7-22 1983 1-6 1983 7-12 1984 1-6 1984 7-12 1985 1-6 Electrical and electronic -9,741 -20,574 -5,444 -18,255 27,506 -78,996 Steel and iron 228 -4,784 2,074 -3,508 -2,089 -43,733 Machinery -686 -1,025 3,661 9,719 -6,583 -35,013 Chemical 375 645 -4,651 556 -8,278 -12,103 Textiles -1,053 -1,211 342 34,211 -6,519 -77,846 Primary products and agricultural products -3,906 -2,880 -13,323 -3,954 -18,799 4,828 Miscellaneous goods 1,913 1,854 8,889 -4,604 -5,432 -13,278 Total -12,871 -27,975 -17,341 14,165 -20,194 -256,141 Liberalization appears to have affected the composition of total imports somewhat. For products of heavy industry, which were those most affected by the liberalization, the share of total imports rose from about 74 percent in 1980 to 77 percent in 1983. The share of products of light industry fell from 26 percent in 1980 to 23 percent in 1983. Heavy industry's share of total exports continued to rise during the period of import liberalization, perpetuating the trend caused by the heavy industrialization policies of the 1970s; the increase was from 44 percent in 1980 to 56 percent in 1983. The increases in heavy industry's shares in both exports and imports reflect growing intra-industry specialization. Korea's heavy industries are characterized by weak vertical integration, with most firms depending on imports for many of their raw materials and parts. Import liberalization made access to foreign inputs much easier. As a result, both exports and imports of heavy industry products increased faster than before. Many of the imported inputs used by heavy industry came from Japan, and because their rise exacerbated Korea's payments imbalance with Japan, the Korean government encouraged firms to obtain them from other sources, particularly domestic ones, where possible. In analyzing the impacts of import liberalization programs, it is useful to ask not only how total imports and shares of affected products have changed, but also how individual firms behaved in the face of increased competition. Although not much research has been carried out at the firm level, one can draw some conclusions from a survey carried out by Dong Kuk University Overseas Development Institute in 1985 (see Table 6.4). The survey covered 207 firms each producing at least one of the 311 products liberalized in 1983 and 1984. It found results of this survey strongly suggest that import liberalization was timely and produced a better environment for improving competitiveness. 96 Ohm Young-Suk Table 6.4 Distribution of Firm Responses to Survey on Impacts of Import Liberalization Unaffected Raised Lowered Total sales 52.6 21.3 26.0 Domestic sales/total sales 82.1 5.3 12.6 Domestic market share 77.4 2.9 19.7 Output prices 76.3 7.2 16.5 Production costs 88.8 6.9 4.3 Net profit/sales 82.8 2.3 14.9 Marketing strategy 22.7 - - Marketing expenses 57.8 39.8 2.4 Technology development 46.8 48.4 4.8 Technology level 50.7 45.5 3.8 Product quality 57.4 39.0 3.6 Competitiveness 48.4 26.3 33.5 Note: Derived from Dongkuk University Overseas Development Institute (1985), various tables. Source: Soogil Yang, Import Liberalization and Industrial Adjustment in Korea, KDI Working Paper 8613, 1986, KDI. Conclusion The experience of import liberalization since 1978 gives useful insights into how major economic policies were formulated and implemented in Korea. Trade policies in Korea must reflect the economy's heavy dependence on external conditions, but they cannot be conceived separately from other, domestic, economic policies. Changes in trade policies have affected domestic policies as much as overall domestic policies have influenced trade policy. For instance, the import liberalization program of 1978 was directly influenced by the buoyant export sector and the resultant current account surplus, and became an important factor in accelerating reform toward a market-oriented policy. By the early 1980s, when external economic conditions were less favorable, the market-oriented approach that characterized the Comprehensive Measures for Economic Stabilization had become firmly established. Thus the import liberalization of those years could be carried out as an integral part of a broader economic reform, even though domestic and external economic conditions were unfavorable. Many foreign analysts have observed that the cooperation among different institutions was a major reason for Korea's effective formulation and implementation of major economic policies. Because of the predominant position of EPB in relation to other ministries and the strong commitment of the political leaderships to the new policy, the opposition from other ministries was not as strong as might have been expected. Different ministries have cooperated well with one another once the final position was decided by the government. Recently, the character of the Korean decision-making process has been changing. The role of the private sector has increased significantly. It has become increasingly clear that speed, flexibility and pragmatism may need to be somewhat sacrificed to accommodate the that most of the affected firms were able to maintain their original market shares, their ratio of domestic to total sales, output prices, production costs, and net profit rate, even after liberalization. A significant proportion of the firms were able to improve technology, product Trade Liberalization Policy in the Republic of Korea, 1980-85 97 quality and competitiveness. Like those of the import analysis, the standard rules of the free market economy. The government experienced some difficulties over trade liberalization in the early 1980s when the economy was sluggish. But presently, having seen the results of the liberalization programs, most Koreans consider that they have to live with open markets if the Korean economy is to prosper. 7 THE GENESIS AND AFTERMATH OF THE FARMERS' AID FUND ACT, THAILAND, 1974 Ammar Siamwalla Thailand shares with other countries in Monsoon Asia a preoccupation with rice. Not only is rice a major staple in consumption; it is also a major export item, accounting now for less than 10 percent of Thailand's total commodity exports, but closer to one fourth as recently as two decades ago. Given its strategic importance, it is not surprising that rice has attracted a great deal of government attention. Government rice policy has had two distinct strands. First, since early this century, the government has invested massively in irrigation facilities for rice. Second, set against this promotion of production have been the government's price policies, at times including heavy taxes on exports. Rice export taxation has been subject to continual tinkering: at times the government increased taxes while simultaneously embarking on price support programs; at times it introduced new devices to tax exports; and at times it reduced the rate of one type of tax while raising the rate of another. This case study examines a piece of legislation with important effects on rice pricing policies: the Farmers' Aid Fund Act of 1974. The Act was a purely administrative measure that directed the use to which the revenue from one of the export taxes was to be put. As such it had no direct impact on tax rates. Rather, it affected the way rice policies were formulated, and the relative bargaining strengths among the various agencies from which a consensus had to be created. In this respect, the Act was typical of many Thai laws, particularly those pertaining to agricultural commodities. Though these laws have very little policy content they assign powers to the various agencies of government and thus affect the distribution of bargaining strength in the "horse-trading" that is inherent in Thai bureaucratic culture. Rice Policies Before 1973 Rice pricing policies in Thailand are most easily applied through instruments at the export level. By controlling export volumes, the government can easily control the domestic price structure. Four specific measures were employed in the period covered by this study: * An export duty which was set in 1952 at 5 percent, reduced in 1985 to 2.5 percent, and eliminated in 1986. Being one of Thailand's general taxation measures, the export duty was handled by the Ministry of Finance. * The export premium, a specific tax levied by the Department of Foreign Trade (DFT), Ministry of Commerce. Until the passage of the Farmers' Aid Fund Act of 1974, the revenues from the premium went to the central budget. * A requirement, known as the rice reserve requirement, for exporters to surrender rice to the government at below-market prices. Managed by DFT and at one time used to procure rice for low-priced distribution to civil servants and Bangkok residents, this measure was later used as a simple export management tool by the Ministry of Commerce. * Quantitative restrictions on exports, also managed by DFT, with the beneficiaries of the quota rent being the exporters. 98 The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 99 DFT used the last three measures with different intensities over the years to suit the circumstances (Siamwalla and Setboonsarng, 1987). Figure 7.1 indicates the tax-rate equivalents of the four measures over the last three decades. The discussion in this chapter concentrates on the export premium. Figure 7.1 Tax Equivalent of Four Government Measures on Rice Exports (Percent of export price) 70 60- 50- Quota rent 40 20 10 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 Sources: Export Duty: Thailand, Ministry of Commerce for tax rates. Royal Thai Gazette for assessed price. Premium and Reserve Requirement: Thailand, Ministry of Commerce, Department of Foreign Trade. Goals of the Premium The goals of the Thai rice export premium evolved from a preoccupation with generating revenue, in the 1950s and early 1960s, to a concern for domestic price stabilization from the mid- 1960s onwards. In the 1980s, export management received increasing attention as a means of increasing market share in the international market. From the late 1940s to the early 1960s, revenue from taxes on rice was crucial to the government's finances. It contributed as much as one-third of government revenue in the late 1940s, falling to less than 10 percent by 1965 as a result of growth in other parts of the economy. The critical importance of the rice premium to the Treasury precluded any major adjustments in the rate of this tax, even for price stabilization purposes, for fear of destabilizing government revenues. In 1955, a series of reforms introduced by officials from the Ministry of Finance gave control over the rate-setting and collection of the premium to the Ministry of Commerce, yet the Ministry of Finance continued to play a substantial role in decisions on these matters. After 1965, however, when the share of the premium in total government revenue had declined, decisions on the rice premium came to be dominated by the Commerce Ministry, which varied the premium in order to stabilize domestic prices. After remaining unchanged in the four years 1963-1966, premium rates thus changed 11 times in 1967, 5 times in 1968, and 4 in 1969. When rice prices began to soar in 1973, so did the premium rates. In the period before the passage of the Farmers' Aid Fund in 1974, the government's authority to levy the export premium had origins, though obscure ones, in law. The Import and Export Act of 1939 gave very broad powers to the Ministry of Commerce to control both domestic 100 Ammar Siamwalla and foreign rice trade. The Ministry used this law, together with a Rice Trade Act of 1946, to require that exports by private firms be licensed. The government could then levy a fee for issuing export licenses. Two consequences followed from this legal framework. First, the final authority for changing the premium rates rested largely with the Commerce Ministry. For the latter, freedom of manoeuvre in the use of the premium to stabilize prices was of considerable value. (It is worth noting that the stability so achieved was only against the upside risk; the law gave no help against downside risk, because it provided only for taxation, not subsidy, of exports.) The second, less well-known, consequence of the legislation is that it enabled the Commerce Ministry to vary the premium rate shipment-by-shipment. It has never been suggested that the Ministry or its officials abused their authority for private gain in levying the premium on private exporters, although it has been suggested that before 1974, some discounts on the premium were given for specific deals involving a cartel of all exporters. Generally the premium on private export trade was fully collected at the announced rate and the proceeds passed on to the Finance Ministry. Government to government (G-to-G) trade was treated differently. The Thai government's rice trader was DFT. In G-to-G trade, the government arranges the foreign part of the deal, while the tasks of procuring the rice domestically and loading it into ships are usually farmed out to private exporters. The risks of price movements between the time contracts were signed and the time rice was procured were usually borne by the Thai government. Administratively, DFT would handle adverse price movements by giving itself a discount on the premium. DFT's ability to set the premium rates and to vary them from shipment to shipment was therefore very useful to it. Interestingly, this combination of the roles of regulator and trader did not give rise to complaints from the private exporters who sometimes competed with DFT in foreign markets as sellers and domestic markets as buyers, and at other times helped DFT to fill its orders. Flexibility in the collection of the premium also had implications for financial administration. Before 1974, all premium revenues from private trade in rice were to be handed over to the Ministry of Finance, as were profits from G-to-G trade. The rules and regulations on the use of profits from G-to-G trade were (and are) somewhat more lax than those on the use of profits from private trade; in budget documents, the profits from G-to-G trade were put in the same category as those from public enterprises. The consequence of this administrative legerdemain was that the Ministry of Commerce had access to funds which it could use to quick effect. Generally speaking, the decline in the importance of rice premium revenue in total government revenue, and the legal framework just described, implied that rice price policies were very much the preserve of the Commerce Ministry. The general perception of the population, as well as that of bureaucrats in that Ministry, was that the Ministry's main mission was to keep the cost of living low. This perception was quite natural, because the tasks assigned to the Ministry included price control, "anti-profiteering," and the calculation of the retail price index. True, it was also assigned the task of looking after Thailand's foreign trade, but it shared this responsibility with other agencies, and treated it as of secondary importance to keeping the cost of living low. Assigning the rice policy-making power to the Commerce Ministry thus imparted a bias toward a low-priced policy. From an economic analyst's point of view, the export premium policies had three possible justifications. These rationales were used to justify Thai government actions in the decades before the enactment of the 1974 law: * The optimal export tax argument. Thailand's position as a major supplier of rice in the world market (Table 7.1) meant that it was not a price-taker in this market. Given its influence on world prices, an export tax could be justified, the optimal rate of such taxation being the reciprocal of the foreign elasticity of demand for Thai rice. Furthermore, since the world rice market was quite fragmented by quality and grade- The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 101 presumably implying different foreign elasticities of demand for different grades and markets-discriminatory tax treatment on a shipment-by-shipment basis could perhaps also be justified. The optimal export tax argument implies an always-positive level of export taxation, and hence always-lower levels of rice prices for Thai farmers. The rate may however vary over time, depending on movements in the foreign elasticity of demand. In years when world rice trade is a buyers' market, the elasticity will be higher than in a sellers' market. In a buyer's market, a lower premium is called for, and when world supplies of rice are short, a higher rate is justified. * The price stabilization argument. The world rice market has been extremely volatile because of its residual nature (Siamwalla and Haykin, 1984). It may be deemed undesirable to allow domestic rice markets to be fully exposed to this volatility. As Figure 7.2 shows, Thailand's stabilization measures chopped off the peaks without adding to the troughs. Thus, government policies were consistently costly to domestic rice producers. The net incentive to produce rice in Thailand has been negative, but this is consistent with the argument that to enable Thailand to reap profits from scarcity in the world market, it has to restrict its rice output. By the same token, Thai consumers gained, but the gain was because the average price was shifted downward and was a counterpart of what producers had lost. The gain from price stabilization per se is unclear. * Another possible economic argument is that the stability of the price of wage goods has made for social stability, and in particular for a lack of upheaval in labor markets over the years. Figure 7.2 Border and Domestic "Paddy" Prices (Thousand baht/ton, deflated by GDP deflator, 1972=100) 4.5 5 / \ Border price 3.5 / \ 3 I \ 2.5 1.5 Domestic price 0.51 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 Notes: One ton of "paddy' is defined as 450 kg of white rice 5%, plus 150 kg of broken rice, Al extra, plus 30 kg of broken rice Cl extra, plus 30 kg of broken rice C3. Sources: Border price: Thailand, Ministry of Finance, Department of Customs. Domestic price: Thailand, Ministry of Commerce, Department of Internal Trade for 1959-1984. Thailand, Ministry of Commerce, Department of Business Economics for 1985-1986. 102 Ammar Siamwalla Table 7.1 Thailand's Market Share and Rank among Rice Exporters Year Market share Rank 1962 20.02 2 1963 19.31 2 1964 24.52 1 1965 22.98 1 1966 19.14 1 1967 19.06 3 1968 14.90 3 1969 13.65 3 1970 12.98 3 1971 18.43 1 1972 24.22 1 1973 10.13 3 1974 13.65 3 1975 12.74 3 1976 22.26 2 1977 27.49 1 1978 16.39 2 1979 22.53 1 1980 21.19 2 1981 23.30 1 1982 30.62 1 1983 31.03 1 1984 33.13 1 1985 36.75 1 Source: U.S. Department of Agriculture. The second argument in favor of the intervention merges into the third-that is, that the rice premium also acted as a macroeconomic stabilization device, by skimming off and sterilizing the windfall gains at times when rice prices rose. Depending on the share of revenues from the tax in total government revenue (small in Thailand after 1965), such a strategy may destabilize government revenue. (See Figure 7.2.) Though precise quantitative estimates were not made, it may be argued that the decisions on premium rates made by the Ministry of Commerce worked well. Thus Wong (1978) has shown that the average premium rate prevailing in the 1960s at least led to an improvement in static welfare relative to free trade, even if the outcome was not optimal. The variations in the premium rates might be regarded as fine-tuning by the Commerce Ministry in response to changes in the elasticities of foreign demand, but no econometrician actually estimated these changes. Equity considerations formed an important backdrop to decisions on the premium rates, but precise estimates were not attempted until a study by Trairatvorukul (1984). The major redistributive impact of the premium is likely to have been between rural and urban households-the differential in their incomes in Thailand being particularly large (Meesook, 1979). Within the rural sector, the better-off rice farmers would have borne much of the burden of the taxes, given their larger marketable surplus, while within the urban sector the premium would have had a proportionately larger effect on the incomes of poorer households. Trairatvorukul finds that a 20 percent increase in rice prices would have decreased the incidence of poverty among rice growers by about 2 percentage points and increased it among nonrice growers by about the same amount, leaving very little net change. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 103 The effect on wages of the premium-induced increase in the price of rice is difficult to estimate. Trairatvorukul attempts an estimate for rural wages. In a general equilibrium study, Amaranand and Grais (1984) find that whereas in the short run, without the wage effect, wage earners would have lost substantially from the elimination of the export taxes, the increased demand for labor for rice production (which is labor intensive) would have offset that impact and after six years, casual workers would have seen their real incomes rise slightly. Brave New World (1973-75) Thai Politics in Late 1973 On October 14, 1973, after a week of student-led demonstrations followed by a day of rioting, the King dismissed Field Marshall Thanom Kittikachorn, the latest of the military prime ministers who had headed Thai governments since 1948, and appointed in his place Professor Sanya Thammasakdi, a respected legal scholar and university rector as well as chairman of the Privy Council, and hence close to the King. The events that led to the change left the armed forces divided and demoralized and, at least for a time, unable to exercise power. There was a heady atmosphere in Bangkok after the change. There were many strikes even though strikes had been illegal since 1958. Farmers too began to demonstrate, demanding some measure of land reform and solutions to the debt problem. Meanwhile, domestic prices, especially of rice, were rising rapidly as a result of the worldwide commodity boom and general inflation at the time. While the students, workers, and farmers were in the streets, the establishment controlled the legislative assembly. The constitution then in force had been promulgated by the previous regime after its coup in 1971. The members of the military who had occupied the legislative assembly when Thanom's government fell had resigned en masse in response to public opinion. To replace them, the King had convened a National Convention of about 2,436 individuals from all walks of life. The convention was charged with electing members to the legislative assembly, which in turn was to draft a new constitution (Thailand's tenth in 43 years). The assembly that emerged was dominated by bureaucrats, both active and retired, and a number of former politicians. Such establishment figures were basically unsympathetic to what the students were pressing in the streets, but they could not ignore the need for some reforms. They enacted some worthwhile reform legislation in labor relations and land tenure, even though the good intentions embodied in these laws were later not always matched by rigorous implementation. While many of those active in politics at that time sensed that the student uprising of 1973 had irreversibly changed the rules of the political game, they could not foresee where the new political equilibrium would be or what the political and administrative modus operandi would be like. From a long-term perspective, 1974 was a particularly bad year to legislate administrative reforms, even though at the time there appeared to be a need to "strike while the iron was hot." Rice Markets, 1972-74 The world rice market entered 1972 still with a glut brought on by the rapid production expansion between 1967/68 and 1970/71,1 associated with the Green Revolution throughout Asia. The year 1972 is in fact remembered for a unique weather pattern that severely affected agriculture worldwide. Total world rice trade in 1971 was about 8.7 million tons. Between 1971/72 and 1972/73, output fell 4.1 million tons in South Asia, and 2.5 million tons in Southeast Asia. Rice output is normally more stable than that of other cereals, but the thin world market 1. All worldwide data cited here are from U.S. Department of Agriculture, Reference Tables on Rice Supply-Utilization for Individual Countries. Washington, D.C.: September 1982. 104 Ammar Siamwalla is less able to cope with sudden surges in demand. Thus the increase in the price of rice, after mid-1972, was greater than those in the prices of other cereals. In April 1974, the world rice price peaked at 4.6 times its level of June 1972 (see Figure 7.3).2 Figure 7.3 Export Price, Domestic Price of Rice, and Premium Rate (January 1972-December 1974) (thousand baht/ton; paddy basis) 8 7 6 Export price fo.b. Bangkok , ' " . . 5 4 Bangkok wholesale price 3 , ' / (domestic price) 3 2 1 _-Prenmium rate 0 Jan 1972 Jan 1973 Jan 1974 Sources: 1) Export price f.o.b. Bangkok: Board of Trade of Thailand. 2) Bangkok wholesale pnce: Departrnent of Internal Trade, Ministry of Commerce. 3) Premium rate: Departnent of Foreign Trade, Ministry of Commerce. Compiled by Thailand Development Research Institute. Part of the reason why the world rice price increase was so rapid was the action taken by Thailand. Record exports of 2.1 million tons in 1972 had left stocks very low at the beginning of the 1972/73 crop year (Figure 7.4), and rice production in that year turned out to be 9.7 percent below that of a year earlier. An unwise decision early in 1973 to sell 600,000 tons of rice to Indonesia eventually led to a ban on all exports not contracted for by June. In the second quarter of 1973, Thailand's reserve requirements were rapidly increased to feed the cheap-rice distribution system which was heading for a crisis (see Siamwalla, 1975). The premium rates were not increased until August 1973, but since that increase coincided with a period when there was an export ban, the impact on government revenue was at first minimal (Figure 7.5). 2. Data in Figure 7.3 are on a "paddy" basis, i.e., a weighted average of the prices (or premium rates) of the different grades of rice, the weights being chosen to represent the typical milling outturn of a ton of paddy. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 205 Figure 7.4 Export Quantity of Rice (January 1972-December 1974) (Thousand tons) 240 220 200 180 160 140 120 100 80 60 40 20 0 Jan 1972 Jan 1973 Jan 1974 Source: Bank of Thailand. Figure 7.5 Rice Premium Revenue (January 1971-December 1974) (Million baht) 600 500 400 300 2007 100 0 Jan 1972 Jan 1973 Jan 1974 Source: Thailand, Ministry of Finance. The 1973/74 crop was a record one, and as a result the export ban was lifted, but because overseas demand remained strong the premium rate was maintained. Beginning in January 1974, revenue from the premium soared, reaching about 3 million baht, or about 7 percent of total government revenue, for 1974 as a whole. It was this very rapid increase in the premium revenue that provided the context for proposals for reforming the premium system. We now know that the high prices of 1973/74 were merely a blip against the general downward trend of food prices. That was not the general opinion at the time, although discussions in Thailand tended to be somewhat more realistic 106 Ammar Siamwalla than was then current in international fora: those who thought prices likely to fall were generally given a respectful hearing and attempts were made to incorporate their concerns into the design of policy. The Genesis of the Farmers' Aid Fund3 Though the Farmers' Aid Fund Act was a product of the Legislative Assembly that was appointed after the fall of the Thanom government, the basic idea behind this legislation had earlier origins. In 1965, Dr. Thalerng Thamrong-Nawasawat, then an official at the National Economic Development Board (NEDB) but later to become a key official of the Ministry of Agriculture, proposed that the premium be gradually eliminated over a period of 10 to 15 years, and that, as an interim measure, the revenues from the premium be placed into a specially earmarked fund to undertake projects that directly aided rice farmers from whom the tax was collected. This approach may be termed the compensatory view of the Fund. Against this the Ministry of Commerce had hoped to earmark the funds for a much more specific purpose, namely to support prices in years when prices were low. Although the idea was current, it was first discussed in print in an article published in the Board of Trade Journal4 and later reprinted in Khao Panit, a daily published by the Ministry in February 1974. The article pointed out that the prevailing high prices of rice would sink to low levels again when other countries began to build up their productive capacities. The premium revenue-which the author estimated would rise to 3 billion baht or 10 percent of the budget-was therefore a windfall gain for the government, leading the author to suggest that the revenue thus obtained should be set aside in a price stabilization fund (Pratchayapinian), to be drawn on to support rice prices when they fell. The two motives-compensation and stabilization-were not always clearly distinguished. Thus in August 1973, the Minister of Commerce reported that the Cabinet had agreed that the revenue from the premium would be used to buy fertilizers to boost rice production in that season. His statement was echoed by Prime Minister Thanom a few weeks later. Thanom went on to state that the revenue from the rice premium in 1974 would be set aside for a Rice Farmers' Aid Fund. This money would be used to finance "agricultural development and the building of rice storage facilities." No information exists as to where in the bureaucracy this proposal originated, although many of its features suggest that it was probably the Ministry of Agriculture. With the fall of the Thanom government, the proposal was resubmitted to the Prime Minister's Council of Economic Advisors (CEA) in the form of draft legislation to set up the Fund. Given the unpopularity of the previous regime, there was no mention of the previous proposals. The CEA was at that time chaired by Dr. Puey Ungpakom, an economist of great distinction. Among its members were Dr. Thalerng, by then the Deputy Undersecretary at the Ministry of Agriculture, who brought the proposed legislation to the Council. The Council discussed the narrower alternative of using the Farmers' Aid Fund specifically for price stabilization purposes, but preferred the broader approach which allowed the resources to be used for general agricultural development as well as price stabilization. Part of the reason for the preference given to the looser, compensatory, approach may have been that the proponents of the stabilization fund did not have a concrete plan of operations for the Fund (Thanapornpun, 1985). It is questionable, however, whether a more concrete proposal would have led to a different outcome. Written mostly by bureaucrats who do not like to be tied down to a concrete plan, Thai laws are generally vague about such matters. The broader approach was consonant with general practice. 3. Much of the following section has benefitted from Thanapornpun (1985). I am indebted to Rungsan Thanapornpun for allowing me to make use of his material. 4. The Board of Trade is the main association of traders with close ties to the Ministry of Commerce. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 107 There was one lone voice of dissent in the Council. In a prescient statement, one of the council members, a retired Director-General of the Comptroller-General's Department, pointed out that earmarked extra-budgetary funds are not easily controlled and can be easily abused. After discussion by the Council, the bill wound its way through the Cabinet and to the Legislative Assembly. Proposing the bill on behalf of the government was again Dr. Thalerng, by then Deputy Minister of Agriculture. Debate in the Assembly was brief-all three readings were passed in two days. Again, doubts were expressed about the excessive breadth of the Fund's objectives, exposing it to possible abuse by the Cabinet, which would make the final decisions on the disposal of the Fund's resources. A number of amendments were tabled but they attempted to increase the farmers' representation on the committee overseeing the Fund. Also noteworthy is the fact that no precise data were presented at any point in the deliberations about the revenues from the premium even for the current or the following year, except in very broad terms. Dr. Thalerng suggested a figure of 2-2.5 billion baht in 1974. No estimates for future years were given. Nor was there any estimate of the costs of the various possible programs that would be made possible by the Act. The Provisions of the Farmers' Aid Fund Act 1974 The Act had two main parts: the first (Articles 6 through 10) dealing with the revenue side, i.e. the power to set and vary the premium rate, and the second (Articles 12 through 20) with the expenditure side. The linking Article 11 mandated the transfer of the premium revenue to the Fund. On the revenue side, the Act mostly regularized what had been previous practice, though with some minor changes. First, for the tax to be levied on a specific commodity, the Ministers of Agriculture and of Commerce were required jointly to obtain Cabinet approval and then make the announcement (Article 7). Once the announcement had been made, the Commerce Minister could set and vary the rate, although Cabinet approval of rate changes was still required (Article 8). Article 7, which set out the rules for assigning which commodity was to be subject to the premium, had a major weakness. It specified the conditions under which a commodity could become taxable-when there was a danger of a domestic shortage or when a high world price was leading to excessive profits among the exporters. But it did not specify how a commodity could be declared ineligible, and hence did not provide a mechanism for ceasing the taxation. Thus throughout the period 1983-85, rice continued to be subject to export taxation, at a time when the conditions specified in this article had long ceased to hold. From the point of view of the Ministry of Commerce, Article 8 was a somewhat retrograde step, curtailing its previous power to vary the premium rates without Cabinet approval. But between 1974, when the Act was passed, and 1982, the Ministry had another policy instrument up its sleeve: it had sole discretion over the size of the rice reserve requirement. Between 1975 and 1982, there were 11 changes in the premium rate as against 22 changes in the rice reserve requirement. Article 9 confirmed the power of the Commerce Ministry to adjust the premium rates levied on individual G-to-G shipments. The Ministry therefore had access to the extra profits to be made on this trade. These profits had been the traditional extra-budgetary resource available to this Ministry, just as the Farmers' Aid Fund was to become a similar resource for the Ministry of Agriculture. Turning now to the portion of the Act that specified the uses to which the Fund was to be put, Article 11 stated unequivocally that the Fund was not to be placed under the general revenues of the government, but was to be used for three purposes set out in Article 18: * promotion of agricultural production through activities such as input subsidization, credit provision and land reform; * support of agricultural prices; * emergency action on environmental dangers that might affect agriculture. 108 Ammar Siamwalla Allocations of the Fund's resources were to be proposed by a Farmers' Aid Committee (FAC) but approved by the Cabinet (Article 11). The Committee was composed of 11 specified bureaucrats5 and not more than eight other individuals6 appointed by the Cabinet (Article 12). It is hard to fathom what the framers of the Act desired of the legislation. The need to be all things to all the contending factions within the Legislative Assembly in practice allowed a very broad scope of expenditures from the Fund. The possibility of its abuse for patronage purposes was raised during the debate on the bill, and the solution was to give the authority to initiate expenditure proposals to FAC, which consisted only of bureaucrats and appointed farmers' representatives. FAC, however, turned out to be totally inadequate to the task, as will be seen below. The Implementation of the Act Rice Politics After 1975 With the promulgation of the constitution drafted by the legislative assembly in 1974, and after the ensuing elections, Prime Minister Sanya Thammasakdi handed over the reins of government to elected politicians. Their rule lasted 21 months after which the military returned, on the back of a bloody suppression of student demonstrators in October 1976. A gradual political loosening up occurred with elections held in 1978 under yet another constitution which allows for parliamentary elections, yet reserves substantial powers for the military. The resulting system of government since then has been a de facto coalition among the military, their allies in the civilian bureaucracy, and the elected politicians: the relative strengths of the components rising and falling somewhat according to short-term events and support (or withdrawal of support) by the monarchy. Throughout these changes, the civilian bureaucracy as a whole remained largely unscathed. Though individual officials were sometimes greatly affected, particularly after the October 1976 coup d'etat, but the bureaucracy functioned basically as it did before the October 1973 revolution. Generally, its power waxed and waned alongside that of the military. Essentially, the dominant political group since 1976-a combination of the monarchy, the military, and the civilian bureaucracy-exercises its power through control of a few key ministries: defense, interior and finance, as well as the prime ministership. The agriculture and commerce portfolios, which have the key influence over rice policies, were generally in the hands of the elected politicians from various political parties. They faced severe constraints on their power. All they could do was to set up a patronage system to yield them and their supporters the financial wherewithal to pay for their political activities. In the case of rice policies, the key actors were the rice mills and the rice exporters. The former, in, particular, were extremely important, because their proprietors were important patrons in the rural areas. Any aspiring member of parliament had to establish strong links with them to be able to garner in the votes as well as the campaign financing. Their input into the design of price support programs has therefore been critical. The Kukrit Pramoj government (February 1975-April 1976) set up in November 1975 a scheme to "guarantee" paddy prices at 2,500 baht per ton for the 1975/76 crop. The guarantee was to be implemented by enforcement of price controls. (Whether the scheme was practicable or not was never put to the test: in January 1976, two months after the scheme was launched, the Kukrit 5. They are the Permanent Secretaries of Agriculture (chairman), Finance, Commerce, Interior and Industry, the Secretary-General of the National Economic and Social Development Board, the Directors- General of Foreign Trade, Internal Trade, Comptroller-General and Agricultural Extension Departments and the Managing Director of the Bank for Agricultural Cooperatives. 6. Among whom one is to represent agricultural cooperatives, one farmers' group and three are to be "other farmers." The Genesis and Aftermrath of the Farmers' Aid Fund Act, Thailand, 1974 109 government dissolved the parliament and called new elections; Kukrit himself continued as a caretaker prime minister.) The guarantee scheme was meant to appeal directly to rural voters. The elements of patronage that were to loom large in later price support operations had not yet appeared.7 Rice mills were strongly opposed to this program. The elan with which the program was presented established something new and permanent in Thai politics: the yearly political rice cycle (Siamwalla and Wongtrangan). Since then, all governments have had to pay at least lip service to the idea of "supporting" rice prices for the farmers. This marked quite a drastic shift in the tenor of political debate from that in the 1960s and the early 1970s, when advocacy of higher rice prices would have been considered politically fatal, given the effects on urban consumers. It is in this respect that the Farmers' Aid Fund Act was perhaps unintentionally tuned into the political climate after 1973. By providing funding for the patronage system, it made the price support program possible. By locating the Fund's administration within the Ministry of Agriculture, it shifted the policy bias away from the preoccupations of the Ministry of Commerce. The Act can thus be viewed as a facilitator that allowed the macropolitical changes that favored farmers to be translated into policies that appeared to help them. Rice Price Support Programs, 1975-83 Price support programs in various forms had been in effect since 1965, but in earlier years, when the declared support price was below the market price, the programs were ineffectual. Neither were price supports very effective in 1969-72, when the market price began to drop below the support price, because the money available was never large enough to buy more than 200,000 tons a year. The average annual production of paddy between 1969-1972 was 13.4 million tons, with about half of it marketed. During this period, the premium was reduced but never eliminated (Siamwalla, 1975). The problems that plagued the program in 1969-72 (inadequate personnel, inadequate money, refusal to acknowledge the connection between low rice prices and the export tax) continued to nag the programs that were implemented after 1975, the only difference being that the problems and the failure were on a larger scale. The price support programs since 1975 lacked continuity of method and organization, and they fluctuated in size (Table 7.2). Typically, the volume of paddy procured would be large when parliamentary democracy was the operating mode of government and when Kukrit's Social Action Party was in charge of economic policy. Typically, there would be pressure for the government to embark on a support program a few months before or almost at harvest time of the wet season crop (December-January). If in that year the government was serious about implementing the programs, then around this time the support price, the method of support and the funding would be decided upon by the government. If the Minister of Agriculture was the advocate and chosen implementer of the program, then he would let the Marketing Organization for Farmers (MOF)8 be the agency responsible for the support program. MOF's task was to purchase paddy from farmers at the support price. If, on the other hand, the Minister of Commerce was to be responsible, then he would choose the Public Warehouse Organization (PWO),9 whose task was to procure rice from the rice mills at 7. The patronage during that year was handed out through the Tambon Development Fund, whereby every tambon in the country (a unit of local government with about 10-20,000 population) is given a grant of about $25,000 to be used for development purposes. 8. MOF is an agency set up on the same day the Farmers' Aid Fund Act 1974 was enacted. Although technically a public enterprise, it is affiliated to the Ministry of Agriculture and Cooperatives and very much its creature. 9. PWO is an older agency than MOF. As regards rice procurement, it plays the same role vis a vis the Ministry of Commerce as MOF does vis a vis the Ministry of Agriculture. Table 7.2 Funds Allocated for Price Support Programs 1974/75-1982/83 ------------------------ MOF --------------------------- -------------------------------------PVO- Revolving Tonnage Revolving Tonnage fund Used procured fund Used procured Season Source (mn.B) (mn. B) (1,000 tons) Source (mn. B) (mn. B) (1,000 tons) Wet 1974/75 FAF 200 216 } - - IF 249 Dry 1975 FAF 305 430 } Wet 1975/76 FAF 0 84 } }F 37 Dry 1976 FAF 0 2 } - Wet 1976/77 FAF 30 26 } n.a. n.a. 36.44 } 3F 15 ) 30.8 Dry 1977 FAF 0 0 } - - - Wet 1977/78 FAF 0 2 } o ) 4 Dry 1978 FAF 0 8 } Wet 1978/79 FAF 900 375 - } 178 Dry 1979 FAF 400 22 } - - Wet 1979/80 FAF 1,230 3,304 } } 1,146 Dry 1980 - - - } Wet 1980/81 Banks1 5,000 - 1,158.1 (white rice) Dry 1981 - Wet 1981/82 FAF 750 606 1 Banks2 6,000 - 104.3 }F 180 (white rice) Dry 1982 - - - } Wet 1982/83 FAF & 2,400 1,600 } BOT3 I 510 - - Dry 1983 - - - Wet 1983/84 FAF 400 - - - - - 1. 1/4 from BOT, others from BAAC 500 mn.B, Krungthai 530 mn. B, and FAF 85 mn.B. 2. 1/4 from BOT, 3/4 from 16 commercial banks. 3. 1,900 mn.B from FAF and 500 mn.B from BOT. Source: For MOF, up to 1979, from MOF, quoted in Thanapornpun (1980), Table 19 ands 20. For PWO, 1980-81, from BOT. For FAF, 1982- 83, from DIT, Ministry of Commerce. 112 Ammar Siamwalla the support price. In one year (1980/81) both ministers wanted to be involved, as they belonged to different parties and both parties wanted to win some points with their clients. In that year, therefore, both MOF and PWO were involved, the one being put in charge of supporting the milled rice price and the other the paddy price. The source of funds for supporting prices was generally the Farmers' Aid Fund, which would allocate the revolving funds for each operation. An alternative source, particularly when the Commerce Ministry or PWO was involved, was the profits from G-to-G trade. (PWO obtained the funds for its program in 1980/81 and 1981/82 by selling notes to the commercial banks, as discussed below.) These funds were sometimes rolled over a few times, as the government did not hold stocks but would in most cases release the rice back into the market very soon after procuring it. The government's main goal in supporting prices was to enable it to purchase rice (or paddy) at higher than market prices. Since the rice was to be resold at market prices, or alternatively, since it could have been obtained more cheaply from the free market, if used to fill G-to-G orders, the programs created a substantial amount of economic rent to those sellers lucky enough to sell to the government. If carefully administered, the programs could have ensured that the recipients of this rent would be some, if not all, farmers. As it was, neither MOF nor PWO was up to the task. Both organizations were small, MOF having about 400-500 employees and PWO about 800-900 employees. Consequently, they were unprepared for the administrative burden required for this task, at least in those years when the amount to be procured was large (say, more than 100,000 tons). Their ineffectiveness was exacerbated by the very programs they were to implement, because as they became the major conduits for patronage money, the top positions came to be occupied by political appointees. With each change in the government, there were changes in these positions. Typically, the rice mills were co-opted into the support programs. Since they alone had the skills and the storage capacity to act as agents for MOF or PWO in these programs, it is not surprising that they were the ones to skim off much of the economic rent that the programs generated. In the MOF-run programs, the design was for the farmers to sell the paddy directly to MOF at the support price, but the procurement operations were in fact conducted from the offices of the rice mills. In a careful study of the distribution of benefits and costs of the rice market intervention by the MOF in 1983, Pinthong (1984) estimated that, of the economic rent thus generated, 54 percent accrued to millers and exporters; 27 percent to government officials and political parties; 6 percent to farm leaders; and 13 percent to farmers themselves. In the case of the PWO-run programs, the poor design was even more evident. PWO accepted tenders from the rice mills to supply it with rice. The government's aim was to raise prices to the support level, which was higher than the market price, and there was an inherent contradiction between this objective and an instrument primarily designed to minimize the government's purchase price. The contradiction could only be resolved if the mills rigged their bids so that the winning one was close to the support level. This they obligingly did (Pinthong and Tinnakom, 1984). The last major price support program for rice was implemented in 1983. The reason it came to an end was that at one point the politicians overreached themselves. Instead of being content to use the Farmers' Aid Fund as they had done up to that point, they had the PWO borrow about 4.5 billion baht (about $200 million) from the commercial banks in 1981/82 to finance the support program. When the short-term notes fell due three years later, the PWO could not pay. Eventually the repayment was rescheduled over the next 10 years with the money coming from the seemingly limitless Farmers' Aid Fund. After this disastrous experience, the program in 1984 and succeeding years was much more modest in scale and became in effect a way of buying off farmers' protests. The administration of the program passed to the Ministry of the Interior which was allocated a revolving fund of 200 million baht (about US$8 million). As it is the main local administrative arm of the government, the Ministry of the Interior is better able to detect early signs of trouble in the The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 113 provinces. When this occurs, it will go out to douse the flames by buying off the local leadership, usually the rice millers who stand to gain the most from government intervention. Decision-Making with Respect to the Use of the Farmers' Aid Fund The figures in Table 7.2 show the central role of the Farmers' Aid Fund in providing the resources for price support programs whose objective could only be political patronage. It must be remembered that the support programs were proceeding at the same time as all sorts of imposts were put on exports by the same government. Indeed, part of the revenue from these imposts was providing the money for the Fund itself. Many who participated in the discussions that led to the 1974 legislation expected that the revenue collected from the premium would be kept until those years when world prices had fallen so low that the premium could not be collected, and that the accumulated Fund would then be used for price support. How could a law, passed by bureaucrats fully conversant with the patronage propensities of their political masters, allow such a diversion of the Fund? The Farmers' Aid Committee (FAC), set up by the Act to propose to the Cabinet the uses to which the Fund was to be put, failed to acquire any technical capability of its own to guide itself or the Cabinet. It never prepared a coherent budget for allocating the Fund's resources in any given year, let alone over a number of years, as would have been necessary to achieve the stabilization objective that had been written into law. Without any coherent position of its own, and without even any clear procedure of approval and follow-up, the Fund became a target for proposals of a sort that would have little chance of being approved by the Bureau of the Budget. Among the projects approved by FAC were an investment in a joint livestock venture with the Government of Singapore (that is now in difficulties), a project to prevent the extinction of sea turtles, and a disastrous project to sink deep tubewells in an area where farmers have already invested in a large number of shallow tubewells. The bulk of the money, however, was spent on price support and fertilizer subsidy programslO(Table 7.3). These programs were granted the funds in a uniform manner. The agencies concerned would have their ministers propose their projects in the Cabinet, who would then order FAC to disburse the funds. The ministry which made the most use of the Farmers' Aid Fund was the Ministry of Agriculture itself: it obtained 88.5 percent of the Fund between 1975 and 1982, because its Permanent Secretary (Dr. Thalerng, for most of the last 10 years) chaired the FAC. The only price support programs financed by the Fund were those operated by MOF and the Ministry of Agriculture. Those that were operated by the PWO and the Commerce Ministry had to obtain resources elsewhere although in the end, as we have seen, the accumulated losses of the PWO had to be paid for out of the Fund. The failure of the price support programs over the last decade had two interrelated causes: the failure to build any institutions that would operate in a professional manner and the lack of a clear rationale for these support operations. First, regarding the institutions, it is true that the Farmers' Aid Fund provided the capital to set up MOF, which was set up on the same day that the Farmers' Aid Fund Act was signed into law. Twelve years after its establishment, MOF remained organizationally weak, its personnel untrained and its accounts a shambles. The situation of PWO, an older organization, was even worse. The failure to build up institutions was itself a result of the second failure, to build up any rationale for the price support operations. These schemes were proposed on the naive argument that if the government bought rice alongside private traders, prices would rise both because the 10. The fertilizer subsidy programs involved MOF acquiring fertilizers and then reselling at a loss. Since MOF would normally acquire only about 20 percent of the total fertilizer needs of the farmers, a two-priced market emerged and generated economic rent similar to that in the paddy market. In one yar it had trouble collecting the fertilizer debts from the farmers, and ended up establishing a paddy price support program to collect these debts. 114 Ammar Siamwalla Table 7.3 Disbursements of Farmers' Aid Fund by Types of Activity, Fiscal Year 1974/75- 1982/83a (percent) Price support, of which 61.1 Rice 41.7 Sugarcane 10.8 Others 8.6 Livestock and fisheries production programs 1.5 Crop production support, of which 32.2 Fertilizer subsidies 22.9 Others 9.3 General marketing support 1.3 Disaster relief 3.8 Others 0.1 Total 100.0 a. The Thai fiscal year runs from October to September. Source: Farmers' Aid Fund, compiled by Thanapornriln (1984). government itself was paying a better price than the free market and because of the artificial demand thus created. The problem was that the government never acquired resources to purchase a sufficient amount of rice and withdraw it from the market into storage1l so that the price would rise. Because of the government's desire to spread the program as widely as possible, the money obtained for it had to be turned over many times, which could only be done by releasing the rice into the market as quickly as it was purchased. To the extent that any permanent withdrawal took place, it was to fill the G-to-G orders, but then the government could have just as effectively stepped up its G-to-G sales and filled the orders with purchases from the free market (in Bangkok) without any need to engage in costly support operations at the farm level. 11. Some officials justify the government programs by claiming that the temporary storage engaged in by the government would raise prices in the immediate post-harvest period. However, inasmuch as the procurement program is publicly known, millers and traders would include it in their information set when making decisions to buy and sell. Consequently, the (temporary) public stock would merely replace private stocks. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 115 Figure 7.6 Rice Export Taxes, 1981-1986 (Thousand baht/ton; paddy basis) 2.0 1.8 - Rice reserve 1.6- 1.4- 1.2 1.0 0.8 0.4 Export tax 0.2 - Premium Jan 81 Jan 82 Jan 83 Jan 84 Jan 85 Jan 86 Source 1) Rice Reserve Ratio: Department of Internal Trade, Ministry of Commerce. 2) Export Tax Rate: Ministry of Finance. 3) Premium Rate: Department of Foreign Trade, Ministry of Commerce. Compiled by Thailand Development Research Institute. All these arguments were pointed out by critics from outside the government, but were never raised by FAC. In fact, the main forum for arguing the merits and demerits of the support programs was the Rice Policy Committee, an interministerial group chaired by the Prime Minister or the Cabinet itself. The FAC was merely ordered by the Cabinet to approve the funding for the programs. Rice pricing in Thailand will always be a political matter. Whereas in most countries, the political arguments would center on the level of the support price, leaving decisions on the means to achieve that level to technical committees of civil servants, in Thailand the level of prices is proposed by technicians, usually from the Office of Agricultural Economics, and only cursorily discussed by the ministers, while the methods by which the price would be supported would be the subject of intense political debate, a debate in which technical and economic data were almost entirely absent. This shows clearly that the engine of patronage has been driving the whole decision-making process. Neither FAC, nor MOF, nor PWO possessed sufficient strength, intellectual or political, to obstruct the support programs until they finally destroyed themselves. The Shrinking of the Farmers' Aid Fund (1982-85) The softness of the world rice market from 1981 onward meant that eventually the Thai government had to reduce all the export taxes on rice including the premium, hitherto the main resource of the Farmers Aid Fund. This it did over a period of four years (Figure 7.6) to the accompaniment of much debate. The resistance to the reduction of the premium was only partly motivated by the need to maintain the Farmers' Aid Fund. Arguments for retaining the premium, simultaneously with a 116 Ammar Siamwalia price support program, were based on an alternative theory of how the rice market works. Though this theory was never fully elaborated by its proponents, the following explanation has been assembled from a number of their statements. The starting assumptions are that Thai rice exporters price their rice offers to foreign buyers on a cost-plus basis, and that without any control from the government, Thai rice exports would compete their margin down to zero or almost zero. Under these circumstances a reduction of the premium or other export taxes would lead to a reduction of the export price that Thailand received for its rice, a reduction moreover that fully matched the tax reduction. Another way of putting the same theory is that the burden of the export tax is fully shifted forward to foreign buyers. Critics of this theory face the problem that the theory is extremely simple and appeals to practical people. To counter it, one has to rely on theoretical arguments, running the risk of being branded a textbook economist-not altogether an inaccurate description, given the level of the argument. The critics' problem was compounded by the lack of good estimates of the elasticity either of foreign demand for Thai rice, or of the exportable supply. Consequently, the only response that they could come up with was a reductio ad absurdum: if the burden of the premium was fully shifted forward, why not jack it up to astronomical levels? This debating point did not catch on in the various committees where the battles were fought. The reductions in the premium that did occur were granted grudgingly and as a consequence of failures in alternative methods of price support. Thus the reduction in 1983/84 and 1984/85 took place in the wake of the disastrous performance of the programs run by the PWO in 1982/83. The final elimination at the beginning of 1986 took place as a result of the failure of a program launched by Kosol Krairisksh, then Minister of Commerce, in November 1985. The Kosol program was in its own way also a reductio ad absurdum of the theory alluded to above. The Minister argued that the only way to raise the whole rice price structure, domestic and overseas, was to raise the farmgate price. This he proposed to do by using the price control law. He argued that the higher costs to the mills would be passed on to the exporters who would then pass on the higher prices to foreign buyers. To ensure this last consequence, he proposed to set up also an exporters' cartel to curb what was felt to be excessive competition among them. The Kosol program was, in its own way, well thought out. It even took note of the possibility that raising prices might reduce demand, for it had plans to increase domestic stocks of rice. Unfortunately, since the government had little money or, more accurately, was unwilling to put up much money after the PWO disaster, the Minister's recourse was to force exporters to hold rice stocks as a requirement for obtaining export licenses. He also managed to induce the Bank of Thailand to open a low-interest window to get the rice mills to stock more rice. However well-thought out it might have been, the plan failed. The idea that the Thai government could control the price in the vast number of transactions that normally occur in any given year was not very credible to farmers, rice mills, traders or exporters. When the price of rice refused to rise-it in fact dropped sharply, the Minister was forced to resign in January 1986, two months after he launched his program. His successor swept away the program and the remaining premium as well. It would be a mistake to regard the elimination of the rice premium on January 14, 1986-the first time in four decades when there was no levy of any kind on rice exports-as a triumph of "good economics over bad." The extraordinarily low real prices of rice prevailing in the world market were an important factor. Notwithstanding the rhetoric and the theories indulged in by government officials, particularly at the Ministry of Commerce, the traditional main aim of the government's export management had been to stabilize domestic prices, and the elimination of the taxes in 1986 was in keeping with tradition. The reduction of the revenues of the Fund (Table 7.4) as a consequence of the reduction of the premium led to a corresponding fall in its role and that of its creature, MOF, now a discredited institution. By the end of 1986, a bill was being proposed by the government to set up a National Agricultural Council. This council, which will have a full-time staff, will take charge of the premium policy (should it be revived) and of the Farmers' Aid Fund. The Genesis and Aftermath of the Farmers' Aid Fund Act, Thailand, 1974 117 Table 7.4 Receipts of Farmers' Aid Fund (current prices, million baht) Transfers Repayment Fiscal from Rice Sugar Other of loan year other funds premium premium Interest income Total advanced Total 1974/75 100.28 741.27 612.63 4.68 0.00 1,458.86 0.00 1,458.86 1975/76 0.00 346.38 91.19 14.48 0.00 452.05 311.16 763.21 1976/77 0.00 847.91 -14.63 20.63 0.00 853.91 446.11 1,300.02 1977/78 0.00 1,025.41 4.69 66.59 0.51 1,097.20 410.85 1,508.05 1978/79 0.00 1,301.16 -3.05 79.32 0.42 1,377.85 952.98 2,330.83 1979/80 75.00 1,695.43 9.62 73.46 0.00 1,853.51 426.46 2,279.97 1980/81 0.00 1,329.56 167.36 37.38 77.12 1,610.42 1,451.46 3,061.88 1981/82 81.81 939.66 620.82 243.19 128.45 2,013.93 1,441.98 3,455.91 1982/83 73.00 662.80 35.12 162.36 30.84 964.12 1,448.50 2,412.62 1983/84 46.15 587.10a n.a. 143.89 84.20 861.34 1,975.03 2,836.37 1984/85 8.65 471.19b 0.00 305.55 54.88 840.27 1,089.64 1,929.91 1985/86 52.46 179.78 0.00 155.08 75.24 462.56 2,163.76 2,626.32 a. Includes sugar premium. b. Includes kenaf premium. Source: For 1974/75-1982/83 from Thanapornpun (1984). For later years, Comptroller-General's Department, Ministry of Finance. Conclusion What lessons can be drawn from the tortuous course of Thai rice policies since the passage of the Farmers' Aid Fund Act? The law had problems at three levels. At the most basic level, the framers of the Act were unclear about the objectives it was to serve. At the time the bill was being debated within the bureaucracy and in the Legislative Assembly (which at that time happened to be an extension of the bureaucracy), two very different approaches to the use of the Fund were mooted. One was that the burden of the export tax was borne mainly by farmers, primarily rice farmers. In some sense the resources so obtained should therefore be returned to them as compensation, and the deployment of a specially earmarked fund then seemed an appropriate solution. The second approach viewed the Fund as a price stabilization device involving the taxation of exports in high-price years and subsidization in low-price years. What emerged was a fund which was all things to all men. At the second level, the vagueness of the objectives would have led to little damage had the framers of the Act realized the dangers of an extrabudgetary fund. All bureaucracies have developed elaborate mechanisms to control the appropriation and disbursement of central funds. Although many resulting rules and regulations may appear cumbersome and even picayune, the inherent rationality of the control system must not be overlooked. A fund that bypasses the controls altogether is open to a great deal of abuse, unless its use is also subject to very strict guidelines-nowhere to be seen in the Farmers' Aid Fund Act. The framers of the Act 118 Ammar Siamwalla seem to have relied on the fact that the Fund would be overseen by FAC, which was to be picked almost entirely by bureaucrats, who would prevent abuses. But since these bureaucrats hailed from different ministries and had conflicting aims, they became easy pickings for the ministers. Here we come to the third level of problems with the Act. Given the lack of a clear set of objectives and of a control mechanism written into the law itself, its passage should have been followed by administrative action to set up a proper staff that would examine in depth the rationale of the law, work out the financial implications of the different proposals and plan the use of the funds received in a coherent manner. This was never done. Funds were siphoned off for patronage purposes by politicians for the simple reason that no alternative claimant had a sensible approach to the whole issue. FAF thereby became truly a slush fund. BIBLIOGRAPHY Amaranand, Piyasvasti, and Wafik Grais. 1984. Macroeconomic and Distributional Implications of Sectoral Policy Interventions: An Application to Thailand. World Bank Staff Working Paper 627, Washington, D.C.. Arrow, Kenneth J. 1974. The Limits of Organization. New York. Forrester, John. 1984. "Bounded Rationality and the Politics of Muddling Through." Public Administration Review, January-February. Gawthrop, Louis G. 1983. "Organizing for Change." Annals of the American Academy of Social and Political Science 466 (March). Grindle, Merilee. 1980. Politics and Policy Implementation in the Third World. Princeton. Grindle, Merilee, John M. Cohen, and S. Tjip Walker. 1985. "Foreign Aid and Conditions Precedent: Political and Bureaucratic Dimensions." World Development 13(12). Hall, Peter A. 1983. "Policy Innovation and the Structure of the State: the Politics- Administration Nexus in France and Britain," Annals of the American Academy of Social and Political Science 466 (March). Lamb, Geoffrey B. 1987. Managing Economic Policy Change: Institutional Dimensions. World Bank Discussion Paper 14. Washington, D.C. Leman, Christopher K., and Robert H. Nelson. 1981. "Ten Commandments for Policy Economists." Journal of Policy Analysis and Management 1(1). Lindblom, Charles. 1965. The Intelligence of Democracy. New York. Meesook, Oey Astra. 1979. Income, Consumption and Poverty in Thailand, 1962/63 to 1975/76. World Bank Staff Working Paper 364. Washington, D.C. Nelson, Joan M. 1985. "Notes on the Political Economy of Adjustment." Washington D.C.: Economic Development Institute, World Bank. Pinthong, Chirmsak. 1984. "The Distribution of Benefits and Burden of the Rice Market Intervention by the Marketing Organization for Farmers, 1982/83." Warasan Thammasat. Vol. 2 (June) (in Thai). 119 120 Bibliography Pinthorg, Chirmsak, and Prance Tinnakorn. 1984. The Evaluation of :Ri., Price Policies in the Crop Year 1980,/81. Research Report Series 49. Faculty of Economics, Thammasat University (in Thai). Rcischauer, Robert D. 1984. "Getting, Using, and Misusing Economic information.' In Allen Schick, ed., Making Economic Policy in Congress. Washington, D.C. Schultze, Charles L. 1968. The Politics and Economics of Public Spending. Washington, D.C. Siamwalla, Ammar. 1975. "A Historv of Rice Price Policies in Thailand." In Prateep Sondysuwan, ed., Finance, Trade and Economic Development in Thailand: Essays in Honour of Khunying Suparb Yossundara. Bangkok: Sompong Press. Siamwalla, Ammar, and Stephen Haykin. 1983. The World Rice Market: Structure, Conduct and Performance. Research Report 39. Washington, D.C.: International Food Policy Research Institute. Siamwalla, Ammar, and Suthad Setboonsarng. 1987. "Agricultural Pricing Policies in Thailand, 1960-1984." Siamwalla, Ammar, and Kanok Wongtrangan. 1985. "Institutional Basis of Thai Rice Price Policies." Background paper prepared for the World Bank. Th7ailand: Pricing and Marketing Policy for Intensification of Rice Agriculture. Washington, D.C. Thanapornpun, Rungsan. 1984. "The Role of the Farmers' Aid Fund between 1975 and 1983." Warasan Setthasat Porithat vol. 7, no. 1 (January-June):67-134 (in Thai). Verdier, James M. 1994. 'Advising Congressional Decision-makers: Guidelines for Economists." Journal of Policy Analysis and Management 3(3). Wildavsky, Aaron. 1979. Speaking Truth to Power: the Art and Craft of Policy Analysis. Boston, Mass. Wong, Chung Ming. 1978. "A Model for Evaluating the Effects of Thai Government Taxation of Rice Exports on Trade and Welfare.'/ American Journal of Agricultural Economics, vol. 60, no. 1 (February):65-73. The World Bank I * *. _ 6 _ ' * Z~~~~~~~~~~~~~~~' I I _IIII _ . S* .I S , *a I! ! _ ~~~~~~* m - _ , _ =~~~~~~~~ S dl m I I' m. I I El - ls l - !* * * i I *~ l. - . s _! 3|. I_I *~~~~~~~~~~~~~~~~~~~IB 0-813194-