Linking macroeconomic and agricultural policies for adjustment with growth : the Colombian experience

-l ---? 2. Linking Macroeconomic and Agricultural Policies for Adjustment with Growth The Colombian Experience Vinod Thomas A WORLD BANK PUBLICATION Linking Macroeconomic and Agricultural Policies for Adjustment with Growth THE COLOMBIAN EXPERIENCE A World Bank Publication Linking Macroeconomic and Agricultural Policies for Adjustment with Growth THE COLOMBIAN EXPERIENCE Vinod Tbomas with contributions from Sebastian Edwards John Nash Jorge Garca-Garda Jost B. Sokol Ai Chin Wee Mateen Thobani Published for The World Bank THE JOHNS HOPKINS UNIVERSITY PRESS Baltimore and London Copyright @ 1985 by 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 The Johns Hopkins University Press Baltimore, Maryland 21211, U.S.A. The World Bank does not accept responsibility for the views expressed herein, which are those of the author and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research sup- ported by the Bank; they do not necessard represent official policy of the Bank. First printing November 1985 Library of Congress Cataloging-in-Publication Data Thomas, Vinod, 1949- Linking macroeconomic and agricultural policies for adjustment with growth. "A World Bank publication." Includes index. 1. Colombia-Economic policy-Addresses, essays, lectures. 2. Agriculture and state-Colombia- Addresses, essays, lectures. 3. Colombia-Commercial policy- Addresses, essays, lectures. 4. Price policy-Colombia-Addresses, essays, lectures. 5. Coffee trade-Colombia-Addresses, essays, lectures. I. Title. HC197.T38 1985 338.9861 85-45105 iSBN 0-8018-3121-0 Foreword DURING THE 1980s many developing countries have faced increasingly complex issues of adjustment and growth. The difficulties can be attributed to both external and internal factors. One set of problems relates to the varying fortunes of exports and imports of the developing countries in world markets; uncertainties concerning the availability of external capital have produced additional difficulties. Meanwhile, domestic policies have not always been adequate to stimulate growth in the econ- omy and to provide stability in the balance of payments and fiscal accounts. This book deals with economic policies for growth and adjustment during pe- riods of rapidly changing circumstances and links them to developments at the sec- toral level. It is based on the experience of Colombia, which, after a long period of successful performance, has been experiencing a downturn in economic growth and increased deficits in the external and fiscal balances. The study analyzes factors that have contributed to the country's long-term development. Turning to the more recent events, it addresses policy measures that are needed to adjust the economy, bring about a resumption of growth, and generate more employment. In the exploration of growth prospects, the focus of the investigation is on op- tions for revitalizing the agricultural sector, which in the past has been a major source of long-term growth. A variety of policy alternatives in the areas of coffee diversification, price supports, price stabilization, input subsidies, and investment strategy are evaluated. In addition, trade-related issues of adjustment which go be- yond the typical concerns of any single sector are emphasized. Such a macroeco- nomic framework has been necessitated by the close links in Colombia among agri- culture, trade, and growth. The performance of coffee exports, in particular, has strongly influenced the course of macroeconomic events. In turn, macroeconomic policies-the exchange rate, import controls, fiscal and monetary measures-have substantially affected progress in agriculture, both directly and indirectly. The broader focus that is brought to bear on sectoral analysis is a noteworthy aspect of this work, which could be useful in the study of sectoral strategies in other countries with similarly complex economic problems. It highlights the importance of coffee and agriculture in driving economic developments and the behavior of sectoral dynamics in an economywide context. A second feature of this volume, V Vi FOREWORD one that is of particular concern to policymakers, is the review of long-term sector- specific options for growth in the context of pressing economic choices: in the present case, economic stabilization measures such as exchange rate adjustments and fiscal restraint are evaluated jointly with pricing and investment decisions at the sectoral level. Finally, a variety of technical analyses are employed that could be applied in other studies of a similar nature. In the area of macroeconomic policy, analyses are provided of the effects of a major commodity-and a leading sector- on the rest of the economy, and the effects of key policy variables such as the exchange rate. Price policy analysis includes measurement of the benefits and costs of various price intervention options. Finally, the volume provides some of the data used in the estimations. During the final work on this study in 1984 and subsequently, the government has taken a variety of policy measures-some of which are recorded here-with the goal of achieving adjustment while protecting economic growth and employment generation. The reforms have by no means been easy to adopt, and they reflect the vision of the policymakers in making politically difficult decisions in order to sus- tain the country's medium-term performance and prospects. The World Bank has provided its support for this economic program, both through policy analysis and through a major policy loan designed to strengthen the country's external trade. There is evidence of a restoration of stability in the economy so far and of the maintenance of the international financial community's confidence and involve- ment in the country during this difficult period. The equally important task of sustaining the policy improvements and restoring rapid development through the medium term remains. The World Bank has maintained a fruitful relationship with Colombia for more than thirty years. The first World Bank economic mission to a developing country visited Colombia in 1949. The first World Bank Economic Report in 1952 and the first publication of a country report in book form in 1972 also concerned Colom- bia. The present volume is the result of the World Bank's more recent economic and sector work in the country. It has grown out of a 198 3 study in Colombia, led by Vinod Thomas, and the author's subsequent findings. Its publication reflects the continuing involvement of this institution in the analysis of the country's economic policies, and it represents a timely contribution to discussions of its developmental achievements and prospects. A. DAViD KNOX Regional Vice President Latin America and Caribbean Office The World Bank June 1985 Contents Preface xi Abbreviations and Acronyms xiii Currency Equivalents xvii PART ONE. MACROECONOMIC POLICY AND AGRICULTURE 1. An Overview of Trade and Agricultural Policies 3 Long-Term Performance and Its Sources 5 Trade in Agriculture 7 Recent Developments and Some Explanations 10 Macroeconomic Policy and Inflation 15 The Medium-Term Outlook 19 A Policy Overview 20 Notes 22 2. Trade Policy and Export Promotion 25 Factors That Affect Policy Choice 25 Past Policies and Exports 26 Competitiveness and the Real Exchange Rate 30 Stimulating Noncoffee Exports 33 Addressing the Appreciation of the Peso 36 Export Promotion Efforts 38 Policy Conclusions 39 Notes 41 3. Import Policy for Growth and Stability 43 Long-Term Policy Directions 44 Long-Term Development of Imports 45 Recent Developments 48 Policy Implications of Recent Developments 53 Policy Conclusions 55 Notes 56 vii Viii CONI ENT S PART TWO. AGRICULTURAL PRICE POLICY 4. Price Interventions, Competitiveness, and Incentives 61 Government Interventions in Agriculture 62 Protection and Efficiency in Agriculture 66 Production Incentives for Major Commodities 71 Trends in Relative Prices of Agricultural Products 73 Policy Conclusions 76 Notes 78 5. Price Stabilization in Agriculture 80 Some Basic Considerations 80 Addressing Year-to-Year Instability 83 Seasonal Price Fluctuations 87 Policy Conclusions 97 Notes 99 6. Coffee Policy under Changing Prices and Technology 101 Colombia's Coffee Economy 101 Adjustments in the Coffee Economy 107 Coffee Price Policy 110 Technology and Yields 113 Coffee Diversification 114 Policy Conclusions 117 Notes 118 PART THREE. PRODUCTION POLICY IN AGRICULTURE 7. Investment in Agriculture 123 Public Sector Expenditures 123 Private Expenditures 130 Recent Government Efforts 133 Policy Conclusions 134 Notes 135 8. Agricultural Technology, Input Policy, and Marketing 136 Trends and Problems in Input Adoption 136 Research, Technology, and Infrastructure 140 Renewable Natural Resources 142 Rural Labor and Income 143 Fertilizer Pricing 144 Credit Policy 145 Agricultural Marketing 149 Policy Conclusions 152 Notes 154 9. Conclusion 156 Macroeconomic Policy, Performance, and Agriculture 157 Coffee Policy and Noncoffee Production 159 Sectoral Incentives 160 CONTENTS ix APPENDIXES A. The Interaction of Coffee, Money, and Inflation in Colombia 165 Sebastian Edwards Changes in the Price of Coffee and Competitiveness 165 Coffee, Money, Inflation, and Competitiveness 169 Empirical Results 173 Conclusion 177 Notes 177 B. The Exchange Rate and Noncoffee Exports 179 Sebastian Edwards A Simple Model to Determine Colombia's Noncoffee Exports 179 Estimation of the Model 181 Notes 185 C. The Rate of Devaluation and the Nominal Interest Rate 186 Sebastian Edwards Interest Rate, Rate of Devaluation, and Money 186 Estimation 188 Forecasting the Interest Rate 192 Conclusion 193 Notes 194 D. The Stability and Predictability of Prices, Producers' Income, and Profitability 195 John Nash Data Sources 195 Method of Calculation of the Indexes 196 E. The Welfare Cost of Price Stabilization 201 John Nash Note 203 F. The Organization and Management of the Coffee Economy 204 John Nash Coffee Marketing and Export 205 Export Taxes and Contributions 208 The Tax System and the Producers' Price 211 The Recipients of Coffee Taxes 216 Notes 216 STATISTICAL APPENDIX 217 Index 245 Preface COLOMBIA HAS EXPERIENCED a rapid and sustained economic development during the past three decades. In the early 1980s, however, the country's economic perfor- mance slowed. During 1983-84, when this study was completed, Colombia's ex- ternal accounts were facing significant difficulties. Faced with growing balance of payments problems, the government has initiated an adjustment program that em- phasizes export development and fiscal and monetary restraint. With a substantial depreciation of the real exchange rate, a selective opening up of imports is envis- aged. Some of the problems of the external sector during the first half of the 1980s could be attributed to a slowdown in the expansion of noncoffee production and exports combined with a downswing in coffee prospects. Equally important, how- ever, were the unfavorable trends in the capital markets, which were responding to the economic and debt woes of Latin America. Thus, even while Colombia was achieving a significant adjustment in its current account of the balance of payments, the pressure on the capital account continued to hurt the external sector balances. Given continuation of policies for adjustment with growth and attention to genera- tion of employment, a strengthening in the balance of payments and a revitalization of development are to be expected. While international factors are crucial to an understanding of the evolving situa- tion, this study is focused on domestic measures managed by policymakers. Even though the near-term issues of stabilization are of great importance, the book also establishes some of the requirements of medium-term performance. Emphasizing the need for sectoral revitalization, the work reviews growth options in agriculture. These issues, however, cannot be divorced from the trade-related macroeconomic concerns, not only because of the pressing nature of the latter but also because of their close relationship with agriculture. The book is in four parts. Part one is a discussion of trade-related macroeconomic policies, with special attention given to their connection with agriculture. Chapter 1 begins with an overview of issues and the effects of macroeconomic policy on the agricultural sector. The next chapter is a review of export policy focused on man- agement of export incentives, while in chapter 3 import policy alternatives are con- Xi Xii PR EFA CE sidered. Part two is an assessment of price policies in agriculture, including price support and import restrictions (chapter 4); price stabilization issues (chapter 5); and policies regarding coffee production and pricing, and issues related to agricultural diversification (chapter 6). In part 3 nonpricing issues that influence the production environment in agriculture are examined, including public and private sector invest- ment in agriculture (chapter 7) and direct policies aimed at increasing agricultural productivity (chapter 8). A final chapter draws out the main conclusions of the work. The appendixes contain technical analyses to support some of the conclu- sions. This work has gained from the contribution of several people to whom I am most grateful. Enrique Lerdau, Miguel Schloss, and Guy Pfeffermann gave critical guidance at a very early and formative stage of this work. Roberto Junguito Bon- net, minister of agriculture and subsequently minister of finance of Colombia, pro- vided an orientation to the study that has proved to be extremely valuable. The study has drawn on the contributions of Alberto Ararat, Deepak Bhattasali, Ana Rita Cirdenas, Laurens Hoppenbrouwer, Kei Kawabata, and German Rioseco. This project benefited from the support of several government and private offi- cials in Colombia, particularly Edgar Guti rrez-Castro, former minister of finance: Gustavo Castro Guerrero, minister of agriculture and later minister of develop- ment; Jorge Ospina Sardi, director of planning; Hugo Palacios Mejia, general man- ager of the Central Bank and later minister of finance, Jorge Crdenas Guti&rrez and Hernin Uribe Arango, general manager and deputy manager, respectively, of the Coffee Federation; and Carlos Ossa Escobar, president of the Sociedad de Agri- cultores de Colombia. I am grateful for the advice and suggestions given by Nicho- las Carter, Melvin Goldman, Juan Carlos Jaramillo, Fernando Montes, Diego Pi- zano, German Valenzuela, and Hugo Valdez. I would like to acknowledge the comments of Takamasa Akiyama, David Argyle, Mauricio Cabrera, Ricardo Can- delo, Ramon Lopez, Jayati Datta Mitra, Enrique Ospina, Gabriel Montes, Jose Olivares, Jairo Ramirez, Franz Schorosch, Alvaro Silva, Naveen Verghese, and the reviewers who commented on the manuscript for the World Bank, Raymond R. Frost and J. Peter Neary. David Howell Jones carefully edited the book. Myrna Hanson provided valuable assistance throughout the preparation of the study. In the absence of the proverbial understanding wife and children, I would like to thank my parents and sister for their recognition of the contribution made by my prolonged bachelorhood to the completion of this work. VINOD THOMAS Abbreviations and Acronyms ABOCOL Abonos Colombianos S.A. (a Colombian private sector fertil- izer company) ACOSEMILLA Asociaci6n Colombiana de Productores de Semillas (Colom- bian Association of Seed Producers) AGD Almacenes Gencrales de Dep6sito (General Storage Facility) ALADI Asociacion Latino Americana de Desarrollo y Integraci6n (Latin American Association for Development and Integra- tion) Almagrario Almacen Agrario (Agricultural Storage) Almapopular Almacen Popular (Popular Storage) ASOCANA Asociaci6n de Cultivadores de Cafia de Az6car de Colombia (Sugar Mills Association of Colombia) ASOCOLFLORES Asociaci6n Colombiana de Productores de Flores (Colombian Association of Flower Producers) Banco Ganadero Livestock Bank Bolsa Bolsa Nacional Agropecuaria, S.A. (National Stock Exchange for Agriculture) BOR Banco de la Republica (Central Bank) BP Bono de Prenda (subsidized storage credit) Caja, or Caja de Cr dito Agrario, Industrial, y Minero (Agricultural, Caja Agraria Industrial and Mining Credit Bank) CAT Certificado de Abono Tributario (tax credit certificate) CD Certificado de Dep6sito (term certificate of deposit) CECORA Central de Cooperativas de la Reforma Agraria, Ltda. (Union of Agrarian Reform Cooperatives, Ltd.) CERT Certificado de Reembolso Tributario (tax reimbursement certif- icate) cOFIAGRO Corporaci6n Financiera de Fomento Agropecuario y de Expor- taci6nes (Financial Corporation for Agricultural and Export Development) COLPUERTOS Puertos de Colombia (Colombian Port Authority) xiii xiv ABBREVIATIONS AND ACRONYMS CONPES Consejo Nacional de Politica Econ6mica y Social (National Economic and Social Policy Council) CORABASTOS Corporaci6n de Abastos de Bogota (Central Wholesale Market of Bogot6) CVC Corporaci6n Aut6noma Regional del Valle del Rio Cauca (Au- tonomous Regional Corporation of Rio Cauca Valley) DANE Departamento Administrativo Nacional de Estadistica (Na- tional Department of Statistics) DNP Departamento Nacional de Planeaci6n (National Planning De- partment) DRI (IRDP) Programa de Desarrollo Rural Integrado (Integrated Rural De- velopment Program) EEC European Economic Community EMCOPER Empresa Comercializadora de Productos Perecederos (Enter- prise for Marketing of Perishable Products) ERP Effective Rate of Protection FAO Food and Agriculture Organization FEDEARROZ Federaci6n Nacional de Arroceros (National Rice Producers' Federation) FEDEGAN Federaci6n Colombiana de Ganaderos (Colombian Cattle Owners' Federation) FEDEPALMA Federaci6n Nacional de Cultivadores de Palma Africana (Na- tional Federation of African Palm Growers) FEDEPAPA Federaci6n Colombiana de Productores de Papa (Colombian Producers of Potato) FEDERACAFE Federaci6n Nacional de Cafeteros de Colombia (National Fed- eration of Colombian Coffee Growers) FEDERALGODON Federaci6n Nacional de Algodoneros (National Cotton Pro- ducers' Federation) FEDESARROLLO Fundaci6n para la Educaci6n Superior y el Desarrollo (Founda- tion for Higher Education and Development) FENALCO Federaci6n Nacional de Cooperativas (National Federation of Cooperatives) FFAP Fondo Financiero Agropecuario (Agricultural Financial Fund) GATr General Agreement on Tariffs and Trade HIMAT Instituto de Hidrologia, Meteorologia, y Adecuaci6nes de Tie- rras (Institute for Hydrology, Meteorology and Land Im- provement) ICA Instituto Colombiano Agropecuario (Colombian Agricultural Institute) ICA International Coffee Agreement ico International Coffee Organization IDB Inter-American Development Bank ABBREVIATIONS AND ACRONYMS XV IDEMA Instituto de Mercadeo Agropecuario (Agricultural Marketing Institute) IFS International Financial Statistics 11CA Instituto Interamericano de Ciencias Agricolas (Inter-American Institute for Agricultural Sciences) INCOMEX Instituto de Comercio Exterior (Foreign Trade Institute) INcORA InstitutO Colombiano de la Reforma Agraria (Colombian Insti- tute for Agrarian Reform) INDERENA Instituto Nacional de los Recursos Naturales Renovables y del Amblente (National Institute for Renewable Natural Re- sources and the Environment) LP Licencia Previa (prior licensing) MUv Manufactured unit value index NCF National Coffee Fund NER Nominal exchange rate n.f.s. Nonfactor services OECD Organisation for Economic Co-operation and Development OPSA Oficina de Planeaci6n del Sector Agropecuario (Agricultural Sector Planning Office) PAN Plan Nacional de Alimentaci6n y Nutrici6n (National Food and Nutrition Plan) PLANIA Plan Nacional de Investigaci6n Agropecuario (National Agri- cultural Research Plan) PLANIF Plan Nacional de Investigaci6n Forestal (National Forestry Re- search Plan) Plan Vallejo Import duty drawback for exporters PPP Purchasing power parity PROCARA Productores de Caia (Cane Producers) PRODESARROLLO Programa de Diversificaci6n y Desarrollo de Zonas Cafeteras (Program of Development and Diversification of Coffee Areas) PROEXPO Fondo de Promoci6n de Exportaciones (Export Promotion Fund; also denotes Export Promotion Agency) REER Real effective exchange rate RER Real exchange rate SAC Sociedad de Agricultores de Colombia (Colombian Farmers' Association) SEA Special Exchange Account SENA Servicio Nacional de Aprendizaje (National Apprenticeship Ser- vice) Situado Fiscal Revenue earmarking USDA United States Department of Agriculture Currency Equivalents Currency unit-Colombian peso (ColS) Exchange rate as of January 1, 1985: USSI = Col$113.890 Col$I = US$0.00878 Dollar amounts given are in U.S. dollars unless specified otherwise. Average exchange rate 1981 1982 1983 1984 USSI CoIS54.491 US$1 ColS64.102 USS1 = CoIS78 861 USS1 Col$100.992 CoISI US$0-0184 CoISI = USSO-0156 Col$1 = US50S0127 CoISI US$0,0099 xvii ø 4< vz 1 An Overview of Trade and Agricultural Policies THIS BOOK addresses issues in economic policymaking in Colombia, a country con- fronting rapidly changing external circumstances. A central issue is the slowdown in the performance of the Colombian economy during the first half of the 1980s, after a long period of rapid expansion. The economic record has been closely associated with success in the agricultural sector. Agricultural and other exports have been the mainstay of favorable outcomes in the external accounts in the past and a significant source of economic growth. At the same time, a proximate cause of the recent difficulties has been the setback in the external sector, particularly that related to the performance of exports. For these reasons, a primary objective of this work is to consider ways and means of stabilizing the external sector of the country and revi- talizing growth, in particular by enlarging the net contribution of agriculture to foreign exchange. Since the latter half of the 1960s Colombia has managed the economy with a well-integrated package of exchange-rate, fiscal, and export-incentive policies that have succeeded in promoting the development and growth of its exports. Through the 1970s the performance of industry was more striking than that of agriculture, partly because the former started from a smaller base. Nevertheless, agriculture maintained an average growth rate of about 4 percent, and the development of agricultural exports other than coffee was impressive. The economy steadily diver- sified into one that was more urban and more resilient to external shocks. The policy package, however, began to unravel during the latter half of the 1970s, even as the country was enjoying a coffee boom. The expansion in total noncoffee exports began to slow down, although it was hidden by the coffee boom during 1976-80. By the early 1980s, however, this was accompanied by a down- turn in real coffee earnings, contributing to increasing deficits in the current ac- count of the balance of payments. During 1982-83 this deficit amounted to more than 7 percent of gross domestic product (GDP). As in the mid 1960s, problems began to appear in the external sector, although at this time the situation was aggra- 3 4 MACROECONOMIC POLICY AND AGRICULTURE vated by growing capital-market constraints. The underlying domestic economic situation, however, remained far stronger than it had been twenty years earlier. Noncoffee exports from agriculture and the rest of the economy had established a high plateau of performance, and the real price of coffee was higher. A way to correct the problems in the external sector would be to reestablish che previous basis for export development and growth. Restoration of agricultural in- centives would be essential to the revitalization of this important sector. At the macroeconomic level, however, new constraints need to be addressed. The govern- ment budget has grown rapidly during the 1980s; the surplus that existed during most of the 1970s has changed to deficit, and considerable efforts have been re- quired to achieve a better balance. Wage-rate and interest-rate expectations have also been related to exchange-rate policy, making improvements in the real ex- change rate more complex. Nevertheless, recent experience has shown positive results in reversing these trends through careful macroeconomic management. More remains to be done. In this Colombian context, macroeconomic policies and sectoral initiatives in agriculture for achieving the needed overall adjustmient while sustaining growth will be examined. In recent work the connection between macroeconomic and agricul- tural developments has been noted.' The present study extends the analysis through an explicit focus on this interrelationship in Colombia and also provides a framework suitable for other country studies. Macroeconomic developments in Colombia have been strongly influenced by the fortunes of agriculture, particularly by its external performance. In turn, the record in agriculture has been substantially affected by macroeconomic policies. In view of these links, a part of this work is devoted to an analysis of trade-related macroeconomic policy, which has a signifi- cant bearing on agricultural performance. After this broader perspective has been established, the focus will be narrowed to a study of sectoral policies that affect agriculture directly; then even more specific financial issues will be addressed. The sectoral initiatives that will be investigated include price interventions, particularly price support and import controls for import-competing commodities, storage and price-stabilization measures for cereals, tax policies in the case of coffee, and credit subsidies that affect most products. Taken together, the various pieces of analysis will seek to shed light on possible ways of achieving stability and accelerated growth in the external sector, particu- larly as they involve incentives to produce and export agricultural commodities. The domestic prices of traded commodities in relation to prices of other commodi- ties are determined not only by international price trends but also by domestic policies toward the exchange rate, production costs, and, more generally, the rate of inflation within a country. Inducements to produce and to export are also de- rived from various export-promotion and import policies. In addition, a variety of policies that are confined to agriculture can alter output and input prices and in turn affect farm incentives. Finally, considerations of microeconomic production that OVERVIEW 5 involve yields and farm budgets influence the production environment and affect actual performance. Long-Term Performance and Its Sources The Colombian economy has shown rapid and sustained growth in output and employment since the mid 1950s. Even from the mid 1970s to the early 1980s, while most other countries in the region suffered significant declines in growth, Colombia's GDP rose at a trend rate of 4-5 percent a year. Juxtaposed with an annual population growth rate of about 2 percent, this produced a respectable long- term growth in per capita output. For much of the 1970s, the country enjoyed a substantial balance of payments surplus, and prudent management of the external sector contributed significantly to the confidence in Colombia's creditworthiness shown by international financial markets. On the domestic front, the country real- ized an extraordinary boom in employment and improvements in real wages during most of the 1970s, which accompanied the rapid expansion of labor-intensive pro- duction and exports. Although fortuitous external circumstances have been helpful, it is also true that a favorable policy environment that has prevailed over the long term has been instrumental in producing these results.' The contribution of agriculture to growth has been sizable. Agriculture accounts for between a fifth and a quarter of GDP, some two-thirds of export earnings, and a third of total employment in the economy. The sector continues to influence major macroeconomic trends, not only through its contribution to foreign-exchange reve- nues and import capacity, but also directly through its effect on employment and incomes in both agricultural and nonagricultural activities. An important segment of the sector that deals primarily with food production is predominantly traditional in its production processes. Vigorous economic growth in Colombia has been positively related to the devel- opment of exports. The country's success in increasing exports during the period 1967-75 was based on an outward-looking policy that was in part the result of significant domestic inducements to export promotion. Activism in exporting di- minished significantly after the mid 1970s, partly on account of the increase in foreign-exchange earnings that accompanied the commodity price boom of 1976- 79. Rapid increases in external demand for Colombian agricultural products were the most significant development of this period, with the demand for coffee and illegal drugs leading the rest of exports. The rate of economic growth fell between 1981 and 1983. Colombia's external sector has been in difficulty in part because of the economic problems of its neigh- bors and the debt problems of Latin America. Although an important source of long-term growth, agriculture suffered a sharp downswing, attributable to weak export demand and low international prices and to domestic developments. 6 MACROECONOMIC POLICY AND AGRICULTURE Growth in noncoffee exports, agricultural and others, decelerated during 1975-8 3, first as the relative producer prices in the domestic markets for these commodities declined and later as international export conditions became less and less favorable. Sectoral Contributions to Growth From the late 1960s through much of the 1970s growth in GDP averaged a sizable 6 percent, although it decelerated during the second half of the decade. An even sharper turning point came at the beginning of the 1980s, when the growth rate in GDP began to dip (see table 1-1). During the long term agriculture has grown at a rate about 1.5 percentage points less than the average for the whole economy. Although long-term growth rates in manufacturing and services have been higher, the contribution of agriculture to growth has been just as great in view of the larger share of agriculture in GDP, and the sector has been the mainstay of economic per- formance since the mid 1970s. During 1960-82 agriculture contributed, on the average, about a fifth of the increase in GDP. Manufacturing grew at an unaccustomed rate during 1967-75; the contribution of the manufacturing sector to the increment in GDP during that period was more than 2 3 percent, whereas its contribution was much smaller during the following period. The vitality of the sector has weakened significantly since the mid 1970s for a number of reasons: an appreciated (lower) real exchange rate for exports in the Table 1-1. Total and Sectoral Growth Rates, 1960-82 (percent) Selected Total Period Agriculture' Manufacturing servicesI GDP 1960-67 2.9 5.4 5.6 4.6 1967-75 4 5 8 7.7 6.4 1975-80 4 5 5.2 6 5 5.5 1975-82 3.8 3 4 4.8 4.6 1960-82 4.1 60 6.4 5 Contribution to growth' 21 1 22 8 19.5 100 0 Note: In making these calculations we relied on data from the Central Bank, which are a longer series than those of the Departamento de Administrativo Nacional de Estadistica (DANE, the National Depart- ment of Statistics), which has been revising the national accounts, the revised data from DANS are pre- sented in appendix tables SA-1-4. The direction and orders of magnitude are the same in both sets of data. Growth rates are those of GDP at factor cost, calculated by the least-squares method a. Includes fishing, hunting, and forestry. b Commerce, banking, finance, and insurance. c Sectoral contributions to growth are measured as sectoral growth rates weighted by the corre- sponding shares of the sector in GDP. These do not add to 100 because some sectors are excluded Source. Author's calculations based on Central Bank data. OVERVIEW 7 face of the coffee bonanza; the slowdown in domestic economic activity; lagging technological progress in the highly protected domestic industrial sector; and, more recently, the downturn in the world economy. During 1980-82 manufacturing declined in real terms, accounting for a good part of the domestic recession. Ser- vices that support the productive sectors sustained a fairly steady and high rate of growth, accompanying high growth rates either in manufacturing or in agriculture. Through the long term, this sector's contribution to the increment in GDP has been about the same as that of agriculture or manufacturing. Following the late 1970s, however, as both agriculture and manufacturing suffered setbacks, the growth in services also dipped.' Components of Total Demand The significant contribution of increases in exports to aggregate demand, particu- larly in industries, has been noted in earlier work.' Outward-looking strategies have contributed positively to the record of agriculture. The period 1970-75 saw a growth rate of about 5 percent a year in noncoffee agriculture, while it was less than 4 percent in the years following 1975. There was a distinct switch in emphasis from external demand-driven growth to domestic demand-based expansion be- tween one period and another, as shown in table 1-2. The contribution of exports to total demand fell sharply, while that from import substitution turned negative between the two periods. The slowdown in domestic activity during the early 1980s was in part attributable to the downswing in world economic expansion. Additionally, the inward-looking nature of development in agriculture during the latter half of the 1970s compared to that in the previous half has also been associ- ated with a slowdown in noncoffee growth. Although Colombia is only moderately open by international standards-accord- ing to a criterion of the share of trade in total output-a finer disaggregation of economic activity makes it immediately apparent that for most of the period under consideration trade has indeed been an important source of growth in the econ- omy. In addition to fortuitous external circumstances, the establishment of an out- ward-looking policy environment during the period 1967-75 was instrumental in creating the basis for a dynamic response to external market conditions. As a result, values of total exports and imports have increased significantly in real terms since the mid 1960s, producing a modest increase in their shares of GDP in the long term.' Trade in Agriculture As a whole, agriculture has had a strong orientation toward trade, in large mea- sure because of the importance of coffee in international trade. During 1982-83 agricultural exports represented some 65 percent of total exports, which roughly corresponds to the behavior of a country with the per capita income of Colombia.' Table 1-2. Shares in Growth of Components of Agricultural Aggregate Demand, 1970-75 and 1976-81 (percent) External demand Domestic demand Intermediate Fixed Changes consumption Intermediate Final capital in Sub- for Sub- Import Total Period consumption consumption formation stocks total Exports exports total substitution agriculture 1970-75 43.0 24.5 2.3 -1.2 68.6 7.4 16.8 24.2 7 2 100.0 1976-81 43 0 32.6 0.8 10.6 87.0 3.4 15 9 19.3 -6 3 1000 Note: The method used to estimate the average contributions of the components of total demand during a given period is explained by Gabriel Montes and Ricardo Candelo in "El crecimiento industrial y la generaci6n de empleo en Colombia," Revista deplaneaci4ny desarrollo, vol. 12, nos. 1 and 2 (January-June 1981), p. 87. Source: Calculations of the Departamento Nacional de Plancamento (DNe, the National Planning Department), using DANE data. OVERVIEW 9 Exports have been a far more significant component of GDP in agriculture, with or without coffee, than imports, and the net foreign-exchange position of agriculture has been strongly positive in the long run.' During 1970-83, the share of agricul- tural commodity exports averaged roughly a third of agricultural GDP. Excluding coffee, this proportion-that is, noncoffee agricultural exports over noncoffee agri- cultural GDP-falls to about 9 percent, which is about I percentage point higher than the share for nonagriculture (see tables SA-5-7). During the same period agri- cultural imports constituted only 3 percent of sectoral GDP but about 18 percent in the rest of the economy. An alternative approach to measuring the trade orientation of agriculture is dem- onstrated in a recent review in which an attempt is made to estimate the share in agriculture of tradables-that is, exportables and importables-defined as the total value added in domestic production of categories that are or have been exported and imported.' The purpose of this estimate is to show the full size of value added of a category-in addition to the share actually traded-that would be affected by an export or import intervention. Including coffee, total exportable and importable agricultural commodities are estimated to constitute 70 percent of sectoral out- put-60 percent exportables and 10 percent importables, primarily food items. The remaining 30 percent is made up principally of food items for domestic consump- tion. Excluding coffee, agricultural tradables are estimated to constitute 40-45 per- cent of noncoffee agricultural output. In contrast, the rest of the economy is characterized by a larger proportion of nontradables: a proportionately larger part of nonagricultural production is derived from generally nontradable sectors such as transport, communication, commerce, banking, and public services. The conclusion is that on the average agriculture con- tains a larger tradable component than the rest of the economy, and in that sense trade policies can influence agriculture more on the average than they can affect the rest of the economy. Furthermore, because of the greater share of exports than imports in agricultural GDP, incentives to agricultural exports-given a satisfactory demand environment-would have a more significant effect on the balance of pay- ments than a comparable protection given for import substitution in agriculture. During 1970-8 3 as a whole, agricultural exports grew at an estimated (by least squares) rate of nearly 5 percent a year in constant prices, coffee growing at a rate slightly above this average and other exports slightly below. During this period, flowers, bananas, fresh meat, hides and skins, and tobacco have grown at impressive rates, albeit from small bases. It is noteworthy, however, that after a growth rate of 5 percent a year in constant prices during 1970-75, noncoffee agro-based exports grew at a rate of only 1.7 percent during the period 1975-83, and even with its own recent decline in growth, coffee has been the mainstay of export growth. Agri- cultural imports have grown steadily, although at a rate below the average for the rest of the economy: in 1970-83 agricultural imports grew at 5.5 percent, while the rate outside the sector was more than 6 percent. The larger part of the increases in agricultural imports have come from increased domestic consumption of im- 10 MACROECONOMIC POLICY AND AGRICULTURE ported foods, which have not been quantitatively very significant in the balance of payments. The decline in the growth of noncoffee agricultural exports in recent years has been a more serious source of concern, although improvements were estimated for 1984 and early 1985. The position of agriculture as a net exporter is true regardless of whether coffee is included or not. Including coffee, the trade surplus during 1970-8 3 was nearly 30 percent of sectoral GDP. For noncoffee commodities, exports and imports have rep- resented fairly small proportions of domestic production, and during 1970-83 the trade surplus of the noncoffee portion was nearly 5 percent of value added of the same category. The country has remained a net exporter of food products also, steadily registering a small trade surplus. With this trade surplus, food production has been slightly above net domestic supply in all years since 1975. The net supply of food is estimated to have grown by roughly 4 percent a year in constant pesos. The gross value of food output is about 65 percent of the agricultural total, and it is a highly diversified output, the growth of which can contribute enormously to the domestic food supply. Fears of undue reliance on international markets for food and expectations of benefits from heavy protection of domestic food production or from export controls are somewhat misplaced, in view of the trade surplus and the small share of agricultural imports in GoP and in total imports. Recent Developments and Some Explanations In addition to the international recession-and particularly that in Latin Amer- ica-internal developments have also contributed to emergence of the recent prob- lems mentioned at the outset of this study. To understand this process better. it is helpful to disaggregate agriculture into two categories-coffee and the rest. This makes it possible to see that the country has exhibited the symptoms of the boom- ing-sector syndrome observed elsewhere in the world, where the rapid growth of a few primary exports and the resultant inflow of foreign exchange have been cou- pled with a real appreciation of the exchange rate, defined as Colombian pesos per unit of foreign currency, causing significant changes in the deployment of labor and the use of resources that have hurt other productive sectors by drawing resources into the booming sectors.' In Colombia, the coffee and drug export windfalls of the second half of the 1970s produced these adverse effects on noncoffee agricul- ture and the rest of the exportables and importables, and the effect persisted even after the end of the commodity boom period.' External Factors Exports to Mexico, Argentina, Brazil, Ecuador, Costa Rica, and Chile, which had expanded significantly during 1970-1982, declined after 1982. Speculation against the Colombian peso following the 1983 and 1984 Venezuelan devaluations OVERVIEW 11 and the ensuing tight import restrictions in Venezuela exacerbated Colombia's problems. During the first quarter of 1984, the Colombian peso remained appreci- ated in relation to the currencies of a number of Latin American countries, particu- larly to the Venezuelan bolivar. Venezuela used to account for about 25 percent of Colombian exports of goods and services and for much of the transfers received from abroad. In 1984 foreign-exchange receipts from these sources were expected to be less than a fifth of what they were in 1982. The external debt problems of other Latin American countries have been a proximate cause of the pressure on the capital account also. The availability of external financing from commercial banks has shrunk, and the terms and conditions of the loans have hardened. In general, domestic performance has been positively associated with growth in world production and trade. During the period 1965-73 worldwide export vol- umes grew at an estimated 8.7 percent; they grew at only 4.9 percent during 1973- 80. The downturn in the world economy during 1980-82, when growth in indus- trial countries averaged barely 1 percent a year, by dampening international demand for Colombian exports, undoubtedly hurt Colombia's performance. In general the country's terms of trade have been positively related to domestic perfor- mance. After the mid 1960s Colombia's terms of trade began gradually to improve. Having fallen between 1974 and 1975, they improved after 1975 because of sharply rising coffee prices, which compensated for the decline in real prices of noncoffee exports during 1977-79. After 1979 noncoffee export prices began to recover in real terms, partly offsetting the decline in coffee prices. The overall terms of trade in 198 3 were above the level of 1975 (see table 1-3). Excluding coffee, in fact, the terms of trade showed an improvement over those of 1975. For noncoffee agriculture alone, however, they had not regained the 1974-75 level (see chap- ter 4). Coffee exports are determined in part by special arrangements under the Interna- tional Coffee Agreement (ICA), while domestic incentives for coffee production and exports are provided in a manner quite distinct from those for other commodities (see chapter 6). The value of coffee exports rose during 1978-80 and fell during 1980-82, following the ups (1976-78) and downs (after 1979) in world coffee prices. Although domestic prices of coffee were considerably stabilized, the higher producer prices in 1976-78 induced a dramatic increase in production. The higher prices were the main reason for increased export values in 1976-77. The record export sales in 1978-80, on the other hand, were the result both of higher prices and of greater volumes. During 1981-8 3, export volumes declined from the 1978- 80 levels and stabilized at levels that were nevertheless higher than the 1970-75 average. The decline in export values during 1981-83 were also on account of the decline in prices. Noncoffee exports in real terms-that is, total value adjusted by a price index- did not do well during 1979-8 3, after a modest increase in the second half of the 1970s, although they were estimated to have improved in 1984 and early 1985. The fall in an index of international real prices of these commodities between 1977 12 MACROECONOMIC POLICY AND AGRICULTURE Table 1-3. Real Prices of Exports, Imports, and the Terms of Trade, 1970-83 (1975 = 100) Noncoffee All Terms of Year Coffee a eXportSb exports trade 1970 1248 77.1 101.3 99.0 1971 102.6 80.9 91 6 92 5 1972 107.2 80.3 94.0 97 3 1973 120.2 84.2 102.2 103.6 1974 105.8 121.1 112.4 108 1 1975 100.0 100.0 100.0 100.0 1976 190.3 102.6 144.4 140.3 1977 263 8 129.4 193.2 194.6 1978 176.7 104.0 140.1 145.3 1979 154.1 89.1 122.0 129.8 1980 137.4 108.5 124.0 132.0 1981 97.8 123.0 110.4 111.3 1982 109 9 115.9 112.9 107.8 1983 104.8 124.2 114.5 110.4 Note: For additional details of agricultural prices, both external and domestic, see chapter 4 These export prices have been deflated by a July 1985 World Bank estimate of a unit-value, U.S -dollar price index of manufactured exports from developed to developing countries. a. New York price of Colombian coffee, from IFs yearbook, line 76e b. Based on indexes of export unit values from irs; the noncoffee export index was calculaced from price indexes of coffee and all other exports, weighted by their respective shares. c. Export price index divided by import price index for Colombia, as in iFs. Sovrces: International Financial Statistics (iis) (Washington, D C : International Monetary Fund, an- nual publication); World Bank estimates. and 1979, particularly in relation to coffee, may have contributed to the poor ex- port performance. Quantities of these noncoffee commodities exported may have slackened in the 1980s in response to previously depressed international prices. External prices, however, cannot fully explain the declining export trend following an improvement in world prices and Colombia's overall terms of trade. Within agriculture alone, a fall in external prices has been a more serious impediment than in the rest of the economy. In general, however, domestic price developments and nonprice factors must also be accounted for in explaining recent trends. Evolution of Internal Relative Prices Prices of exports and imports in relation to those of nontraded commodities in the domestic market reflect the degree of competitiveness of the traded goods. Be- tween 1975 and 1977, when the relative price of coffee was rising, the internal relative prices of both noncoffee exports and imports with respect to nontradables declined, and this reduction continued in 1978 (see table 1-4). Between 1979 and OVERVIEW 13 Table 1-4. Internal Relative Prices of Exports and Imports, 19 70-83 All Coffee Noneoffee Year exports exports' exports Imports 1970 88.9 111.6 68.3 82.6 1971 79.3 92 9 67.8 80 1 1972 86.3 104.6 73.3 78 7 1973 98.8 121.4 83 8 80.9 1974 104.7 102.9 106.0 97.3 1975 100.0 100.0 100.0 1000 1976 120 6 165.1 90.6 91.9 1977 135.4 235.6 83.6 82.2 1978 111.2 151.7 78.9 77.0 1979 97.0 110.9 83.1 78.5 1980 102.4 111.9 93.6 80.2 1981 85.0 80.2 89.2 77.5 1982 80.8 83.6 79,0 71.5 1983 79.8 80 1 79 6 70.1 Note. Relative prices, as shown in this table, are the ratio of the implicit prices of exports and imports to implicit GDP deflators. a The export price of coffee divided by the oDP deflator. This is not the real internal price of coffee shown in table 6-6. Source. Computed from DAE data by Jorge Garcia-Garcia, "Aspects of Agricultural Development in Colombia" (BogotA, April 1983, processed). 1980 this tendency was reversed, with the relative price of coffee dropping and the relative price of noncoffee exports and imports increasing. During 1981-83, the relative prices of coffee and noncoffee traded goods fell. These price developments would be expected to erode the incentives to produce noncoffee exportables and import-competing commodities, which is what has happened in recent years. By raising disposable incomes and spending, the increase in the world coffee price increased the prices of nontraded products not only in relation to the price of cof- fee, but also in relation to the prices of products in other traded categories.'I In view of the large component of traded products in agriculture, the effect of declin- ing relative prices was felt particularly in this sector (see table 1-5). The relatively high rate of domestic inflation, coupled with a low rate of depreciation of the peso during the period 1976-82, further reduced the degree of competitiveness among traded goods other than coffee. The higher world price of coffee in 1976-79 gener- ated a dramatic increase in international reserves, higher rates of growth of high- powered money, and a tendency toward a higher rate of inflation. The government responded by implementing a stabilization policy and sterilizing-that is, neutraliz- ing-a part of the additional reserves and thus controlling other sources of inflation. Toward the end of 1979, most import restrictions were eased significantly (see chapter 3). Meanwhile, for stabilization, the rate of depreciation of the crawling 14 MACROECONOMIC POLICY AND AGRICULTURE Table 1-5. Internal Relative Prices of Agricultural Exports and Imports, 1970-82 Agricultural Agricultural Agricultural Year exports exports! imports, 1970 134.1 183.5 74.0 1971 117.1 120.5 77.9 1972 1309 135.4 78.9 1973 127 8 139 8 109.6 1974 107.6 1054 121,3 1975 1000 1000 100.0 1976 94.3 147.4 87.8 1977 99.3 212.6 93 4 1978 90.2 148.4 58.3 1979 93 4 113.7 81.5 1980 81.4 107.0 89 2 1981 85.4 84 5 83.7 1982 96.6 89.2 79.7 a. In relation to nonagricultural exports; agriculture includes unprocessed coffee, other crops, and livestock b. In relation to nonbroad agricultural exports, "broad agriculture" includes-in addition to unpro- cessed coffee, other crops, and livestock-the sectors in which coffee and sugar are processed c. In relation to nonbroad agricultural imports; this column refers to imported agricultural commodi- ties within the broad agriculture sector Source: Same as for table 1-4. peg was reduced," and the real exchange rate (pesos per unit of foreign currency) was allowed to decline-that is, appreciate-significantly, contributing to a reduc- tion in the competitiveness of categories of traded products other than coffee. The decline in producer prices in agriculture in relation to prices in the rest of the economy since the mid 1970s has hurt noncoffee agriculture. In addition, price interventions and other sectoral developments have contributed to the performance of the sector, as will be elaborated upon later. The government has intervened in the sector in an effort to provide special inducements to agricultural production, principally through import controls, domestic price supports for selected commodi- ties, and credit subsidies. Import policies affect only a small proportion of these importables and have thus been largely ineffective and often inefficient. Further- more, additional import controls constitute an indirect tax on exports, further erod- ing incentives in the part of agriculture in which exportable commodities are pro- duced. The period of falling agricultural incentives, 1975-83, also coincided with declining government efforts in agricultural investment (see chapter 7). The private sector has taken the lead in investment, although a declining government contribu- tion in such critical areas as research, extension, and development of infrastructure has been detrimental to sectoral development. Meanwhile, agricultural production costs have mounted, and yields of only certain products have increased signifi- cantly. Investments in a number of areas-research, extension, irrigation, input sup- ply-together with improvements in the provision of credit and in marketing can OVERVIEW 15 make a significant contribution to increasing yields and reducing production costs (see chapter 8). The macroeconomic adjustment can help improve agricultural in- centives, as evidence for 1984 and 1985 shows (see chapter 9). Macroeconomic Policy and Inflation Macroeconomic policy interacts with agriculture in good measure through its effects on inflation, the real exchange rate, and incentives to export and import. Particular macroeconomic policies may originate within the agricultural sector it- self. Coffee is one source of such a linkage in Colombia. Coffee Prices and Inflation The increase in the price of coffee produced higher disposable incomes and an increase in the demand for both tradables and nontradables. Since the price of trad- ables other than coffee is, to a significant extent, determined by the world price and the exchange rate, this income effect caused the relative price of nontradables to rise (see appendix A). More important, the increase in the price of coffee tended to generate a balance of payments surplus and an accumulation of international re- serves. Since the increase in international reserves was not fully neutralized, the monetary base also increased. Additional spending induced by the higher coffee earnings increased the demand for credit and probably increased velocity, and in the short run inflation tended to accelerate. A fairly close statistical relation be- tween inflation and monetary growth has been documented." Another avenue through which the higher coffee price affected the real exchange rate was discussed earlier. The monetary authority, worried about a higher rate of inflation, reacted to a higher price of coffee by trying to use exchange rate policy for stabilization and reducing the rate of nominal depreciation of the crawling peg. As confirmed by the statistical analysis in appendix A, the higher coffee price was associated with an increase in the monetary base, which was related positively to a rise in the rate of domestic inflation and a fall-that is, an appreciation-in the real exchange rate. Additionally, the fiscal deficit of the central government rose from 1.2 percent Of GDP in 1979 to about 4 percent in 1982 and 1983; a containment was estimated for 1984-85. (Colombia's fiscal year is identical with the calendar year.) These inflationary developments, combined with the inadequate exchange rate adjustment, meant a reduction in the relative prices of noncoffee tradable goods and a consequent loss in their competitiveness. Fiscal Deficit and Inflation With the decline in real prices of coffee in the I 980s, the trends in the interna- tional reserve position and the size of the monetary base were reversed. These ten- dencies should be expected to bring down the rate of inflation and, other things 16 MACROECONOMIC POLICY AND AGRICULTURE being equal, depreciate the real exchange rate. During the early 1980s, however, the growing fiscal deficit became a source of the upward pressure on domestic prices. Through most of the 1970s, the central government's cash balances were positive or modestly negative. A significant turnaround occurred during 1980-82, when current expenditures rose steadily as a result of real wage increases, ear- marked revenues, and automatic transfers to decentralized public agencies, while current revenues hardly improved in real terms; in 1980-82 there were also rapid increases in investment. By 1982 the overall cash deficit of the central administra- tion reached Col$ 110 billion-that is, thousand million-or more than 4 percent of GDP (see table 1-6). The corresponding deficit in the consolidated public sector ac- counts was twice as high, although accurate data for this aspect are not available. Fiscal policy needs to curb expenditures in line with the generation of revenue, while protecting high-priority investments. The government has been increasing-albeit from relatively low levels-expendi- tures that need to be financed by borrowing from private savings and from the creation of money. A part of the government deficit is financed by the creation of money and not met by a concomitant increase in aggregate supply." Since 1980 there has been a sharp decline in the contribution of the growth of international reserves to the money base, while deficit financing has assumed increasing impor- tance and has contributed to inflation. The rising deficit has also meant a diversion of part of investable funds from private savings toward current government expen- ditures, thus reducing the availability of such funds for the rest of the economy. Finally, the fiscal deficit has also reduced the degree of freedom of the government with respect to management of deficits in the balance of payments, given the need Table 1-6. Cash Deficit of the Central Government, 1978-84 (billion pesos) Item 1978 1979 1980 1981 1982 1983 1984' Current revenueb 80.1 94.6 131.1 155.0 197.3 236 2 297 3 Current expenditure 57 2 83 0 121.8 154.2 211.5 277.6 358.9 Current account 22.9 11.6 9.3 0.8 - 14.2 -41.4 -61.6 Investment 21.0 26.0 42.0 61.2 96.1 81,8 101.8 Overall deficit -1.9 14.4 32.6 60.4 110.3 123.2 163.4 Net financing - 1.9 14.4 32.6 60.4 110.3 123.2 163 4 Net external financing -2.2 5.4 16.8 19.1 14.9 2.5 19 6 Net domestic financing 0 3 9.0 15 8 41.3 954 120.7 143 8 Monetary financing 1.7 7.5 22.6 47.4 103.8 107.3 149.3 Domestic credit and other -2 0 1.5 -6.8 -6.1 -84 13.4 - 5 5 Note: In this study the word billion is used to mean one thousand million. a Estimated. b. Excludes nontax proceeds from the special-exchange account and receipts from certificates Source: Calculated by Laurens M. Hoppenbrouwer on the basis of data from the Colombian treasury. OVERVIEW 17 to finance a part of the budget deficit from external financing by reducing foreign reserves. Appendix A shows the positive empirical relation between the government defi- cit and the rate of inflation. Such a positive association, however, may be difficult to discern by merely considering particular years without separating out the simul- taneous effects of other important variables. It is possible in a given year for the deficit to increase while the rate of inflation declines. A recession can diminish tax revenues, thus increasing the deficit, while simultaneously putting a lid on the rate of inflation. In 1983, the reserves dropped drastically, holding the expansion in the rate of growth of the money supply and inflation in check without any decrease in the fiscal deficit. These observations, however, do not negate the partial positive effect of deficits on inflation. Another implication of the foregoing is that under recessionary circumstances, an increase in the deficit may not increase inflation in the very short term. As the economy recovers, however, the inflationary effect becomes evident, other things remaining the same. The results depend on monetary policy as well. The inflation- ary effect of higher deficits can be offset by a tight monetary policy, but such a policy could lead to higher interest rates, which might arrest economic growth. During 198 3-84 the real monetary base was reduced, mainly on account of falling reserves, which in the face of a rising budget deficit exercised an upward pressure on the interest rate. The determination of interest rate is also heavily influenced by the expected depreciation of the peso (see chapter 2). Real Exchange Rate and Competitiveness The data in table 1-7 support the more formal empirical analysis in appendix A, which establishes the statistical link between the price of coffee, the rate of growth of high-powered money, and inflation in Colombia. Together, the findings confirm some of the characteristics of the Dutch-disease type model: other things being equal, a higher price of coffee has led to a higher rate of inflation and an apprecia- tion of the real exchange rate.'6 A growing fiscal deficit has aggravated the inflation- ary tendency, further lowering the real exchange rate. According to table 1-7, during most of 1970-78 and 1981-83, the real price of coffee and the real exchange rate showed the above-mentioned relation. During 1978-81, however, the opposite seems to have been the case. It should be empha- sized, nevertheless, that during the second half of the 19 70s, real coffee prices re- mained at very high levels-even during years in which they declined-contributing to the appreciation of the real rate of exchange. Three factors seem to be notewor- thy. First, the exchange-rate effect of coffee prices involves some lags-that is, the real rate continued to appreciate even after the price of coffee began in 1978 to fall from its extraordinarily high levels. Second, coffee revenues, not just prices, are relevant; revenues continued to rise through 1980, contributing to appreciation of the peso. Finally, during 198 1-82, the real price of coffee hardly changed, but the 18 MACROECONOMIC POLICY AND AGRICULTURE Table 1-7. Selected Macroeconomic Variables, 1970-84 Central Rate of govern- Real growth ment cash Average price of deficit Rate of rate of Real of Terms of money as per- infla- devalua- exchange coffee trade' baseb centage tion d tione rate Year (1975=100) (1975=100) (percent) of GDP' (percent) (percent) (1975 = 100) 1970 124.8 99.0 19.8 n.a. 6.7 6.5 81 1 1971 102.6 92.5 8.8 n.a. 11.6 8.1 83 1 1972 107.2 97 3 25.4 n.a. 13.8 9.7 86.4 1973 120.2 103.6 31.1 1.2 22 0 8.1 89.3 1974 105.8 108.1 18.8 1.5 25.2 10.3 94.7 1975 100.0 100.0 31.7 0.5 23.6 18.7 100.0 1976 190 3 140.3 41.6 -06 19.9 12.2 97 3 1977 263.8 194.6 40.1 -0.5 34.7 6.0 83.4 1978 176 2 145.3 35.2 -0.2 16 7 6.3 83 5 1979 154.1 129 8 30.4 1 2 24.9 8 8 81.8 1980 137.4 132.0 28.8 2.1 27.2 11.1 82 0 1981 97.4 111 3 21.8 3.1 28.1 15.3 78 1 1982 109.9 107.8 17.7 4.4 24.6 17.6 73.4 1983 104.8 1104 13.5 4.1 19.8 23.0 75 2 19848 111.5 115.0 18.3 4.5 16.4 27.9 81 8 n a Not available a. From table 1-3. b. Currency in the hands of the public plus reserves held by commercial banks in nominal terms. c. Before amortization. d. Percentage change in period averages of the consumer price index. Blue collar workers. e. Period averages are from FS. f. Measured against a trade-weighted basket, as explained in table 2-3. g. Preliminary estimate Sources. Banco de la Republica, IFS, International Monetary Fund, DANE, and World Bank data. peso continued to appreciate. This was to a significant extent the result of a grow- ing fiscal deficit. The net result was to hurt the performance of noncoffee tradables. On the other hand, there has not been any significant output response from the relative prices of nontradables, which improved. The coffee boom hid the negative effect on the noncoffee-tradable economy. With a fall in noncoffee prices in the I 980s, a lower rate of inflation and an increase-that is, a depreciation-in the real exchange rate ought to be expected. The decline in the rate of inflation was slow until 1982, however; the continuing increase in the fiscal deficit combined with increases in real wages kept increases in aggregate demand higher than increases in aggregate supply. OVERVIEW 19 Real Exchange Rate and Import Policy While export development was depressed, imports showed a healthy growth during 1980-82. The share of imports-goods and nonfactor services-in GDP during 1980-82 was higher than its level during the early 1970s. The balance of payments effect was becoming steadily more negative, as evidenced by the growing resource gap during this period. The appreciated exchange rate has hurt the com- petitiveness of import-competing industries and has contributed to balance of pay- ments difficulties. As will be seen in chapter 3, imports are to a degree a policy variable, because the government influences the level of imports through licensing, tariffs, and some pro- hibitions. With the problems in the external sector, the controls increased during 198 3-84, partly to offset the overvaluation (appreciation) of the exchange rate and partly to protect foreign exchange reserves directly. By the same token, when im- port restrictions have been effective, they have been observed to contribute to infla- tion in Colombia and to constrain domestic production." As the balance of pay- ments stabilizes and adjustments in the exchange rate are improved, therefore, gradual liberalization of the import regime will be desirable. The Medium-Term Outlook During 1980-8 3 GDP grew at an average of less than 1.5 percent a year." Non- coffee exports fell in constant prices from 1980 through 1983." Combined with the fall in coffee revenues, this meant a sizable decline in commodity exports in constant prices during 1980-8 3 (see table 1-7). Commodity imports, in the mean- time, increased in constant prices in 1981 and 1982; this increase, combined with the slump in exports, produced deficits in the trade and resource balances during 198 1-8 3-the first in seven years. The economy remained modestly active during 198 3-84. A significant outcome was the drop in inflation to 19.8 percent in 1983 and an estimated 16.4 percent in 1984, although this drop was partly the result of declines in reserves and in the growth of the monetary base. In four major cities, the average unemployment rate in 198 3 was estimated to be 11.8 percent; it had been 9.2 percent in 1982, and it is estimated to have increased further, to 13.5 percent, in 1984. GDP in real terms is estimated to have increased by less than 1 percent in 198 3 and about 3 percent in 1984. Exports and imports fell in 1983, as can be seen from table 1-8. A large deficit in the current account-only marginally lower than the deficit in the pre- vious year-was estimated for 1983. Net capital inflow declined significantly in 1983, causing a drop of 37 percent in international reserves. On the basis of preliminary data the current account deficit was estimated to have been reduced significantly in 1984-85. In the absence of a significant turnaround in capital inflows, however, the balance of payments position remained difficult in 20 MACROECONOMIC POLICY AND AGRICULTURE Table 1-8. The Balance of Payments Situation, 1970, 1975, and 1978-84 (million U.S. dollars) 1970 1975 1978 1980 1981 1982 1983 19841 Exportsb 1,000 2,165 4,039 5,747 4,678 4,785 4,050 4.542 Importsb 1,149 2,030 4.131 5,494 6,078 6,738 5,814 5,113 Current account balance -339 -98 357 104 -1,722 -2,886 -2,826 -1,870 Percentage Of GDP -4.8 -0.7 1.5 0.3 -4.7 -7.4 -7 3 -5.1 Net year- end reserves 152 547 2,482 5,416 5,630 4,891 3,079 1,796 Equivalent months of importsb 1 6 3 2 7.2 11.8 11.1 8.7 6.5 4.2 a. Preliminary estimate. b. Goods and nonfactor services. c. Estimates made by the Banco de la Republica, Bogota, and the World Bank Sources: Banco de la Republica and World Bank estimates. 1984. Any substantial improvement in world coffee prices would obviously pro- duce an improvement in the external accounts. An increase in net capital inflows remains essential in order to stabilize the reserve position. According to a base-case or moderate projection, assuming world economic recovery and good domestic policies-some of which are discussed in this study-it is envisaged that, with faster depreciation of the real exchange rate combined with a vigorous export drive backed by efforts to strengthen aggregate supply, rapid economic growth could be resumed in the mid 1980s.20 The balance of payments is expected to stabilize and improve significantly during the latter half of the 1980s, particularly as substantial exports of oil and coal come on stream. Macroeconomic adjustments were initiated in 1984-85, and early indications show that the central government cash deficit as a proportion of GDP Is being re- duced. With a significant increase in the rate of the crawling peg and reduction in the domestic inflation rate, the real exchange rate is estimated to have depreciated substantially. These adjustment measures are to be deepened and complemented by some additional capital inflows, with borrowing policies kept within prudent limits. Once the near-term difficulties have been overcome through these developments, the medium-term outlook is expected to remain favorable for growth, employ- ment, and the balance of payments. A Policy Overview Macroeconomic management in Colombia over the long term has successfully responded to emerging realities within the confines of a democratic process of poli- OVERVIEW 21 cymaking. Policy decisions, however, have been found to be especially difficult when coffee prices fluctuate sharply. This arises from the uncertainty attached to assessments of the duration of price increases and, consequently, the degree of ad- justment in macroeconomic variables that is required. During the coffee price boom, alternative means of stabilization were pursued to varying degrees and with varying effectiveness. To some extent, the increases in reserves were neutralized by the monetary policies adopted, some increase in imports, and appreciation of the real exchange rate. A smaller appreciation of the peso might have been workable if it had been sup- ported by a larger inflow of imports to absorb the increase in domestic demand and if borrowing had not increased sharply, as it did during 1978-82. It should be noted, however, that a major liberalization in the face of an already appreciated peso could exacerbate adjustment problems of import-competing domestic indus- tries. Temporary protection for noncoffee exports in the form of special export incentives might be necessary for a short period and was, in fact, actually extended to a limited degree. Once the price boom has been diagnosed as transitory, how- ever, a reversal of appreciation of the peso-even if the exchange rate might have been at an adequate level during the boom-would normally be the right approach to take. In this respect, the adjustment of the Colombian economy-in reducing inflation, improving the real exchange rate, and shifting resources into production of products other than coffee-could perhaps have been more timely. The difference between domestic and external inflation during 1975-84 was large; domestic prices, measured at the official exchange rate, rose by some 100 percent, while the increase in one index of external prices was about 50 percent. Partly as a result, the producer prices of noncoffee tradables-which, as mentioned earlier, are strongly influenced by international prices-have fallen in relation to the price of domestic goods and services in this period. Since the share of tradables in agricultural output is greater than in the rest of the economy, this decline in relative producer prices has been especially adverse for the sector. Unfortunately, the shift of incentives in favor of nontradables has not produced any significant output re- sponse from this domestic sector as a whole, so there has been little offset to the losses in production and employment, which are the result of slower growth in the tradable goods sector. Colombia now faces real coffee prices that are roughly comparable to those that existed during the mid 1970s but lower than the average real price during the first half of the 1970s. Colombian terms of trade today correspond roughly to the level of those of the mid 1970s. The direction of macroeconomic policy at present con- sists, among other things, of a depreciation of the peso to reach an equilibrium level as soon as possible. At the same time, a reduction of domestic inflation-which could otherwise increase as a result of accelerated depreciation of the peso-is sought through fiscal and other measures, such as a gradual opening up of imports as exports respond to greater incentives. A target of current policy is to bring the fiscal deficit of the central government down from the level of over 4 percent of GDP 22 MACROECONOMIC POLICY AND AGRICULTURE in 1982 and 1983 to less than 2 percent in 1985 and to reduce it thereafter. Corre- sponding decreases in the deficit of the consolidated public sector are also expected. These steps aimed at reducing inflation and improving the real exchange rate would in general be favorable to agriculture, reducing the pressure to provide special price incentives directly to the sector. The government would also have to execute a more forceful strategy of return to export promotion to absorb any significant shift in the domestic supply of agricul- tural products. It should include an adequate general incentive system for noncoffee exports to regain competitiveness, a gradual reduction in the level and dispersion of protection across sectors, and an assurance to exporters of rapid access to foreign exchange and the imports needed to produce exports. A balance of payments pro- jection consistent with a recovery of growth through the 1980s indicates also the need for some increased net capital inflows and investment in the near term. Ac- cordingly, the country needs to continue to speed disbursements of existing loans and put together a program for some additional external financing. This study suggests that both the world economic recession and an adverse shift in relative producer prices in domestic markets have been significant in explaining the performance of noncoffee exports, both agricultural products and others.21 What is also needed is a product-by-product and country-by-country examination of external demand and of institutional and other arrangements required to break into new markets, supported by efforts to improve yield, quality, processing and marketing. An important aspect of agricultural adjustment revolves around the problem of reducing excess stocks of coffee-unless there is a dramatic increase in the demand facing Colombia-and the need to restrain the internal real price of coffee and to induce a shift of resources into noncoffee activities. A good part of diversification is likely to be in the output of additional food, domestic expenditure on which contributes nearly 40 percent of the consumer budget. The demand for noncoffee agricultural products-food and nonfood combined-is projected to grow sufficiently to absorb a significantly higher output level-by about 3 percent (domestic and external combined) during the 1980s under one set of "moderate" assumptions of a world recovery and improvements in Colombia's competitive- ness.22 Efforts to reduce production costs and increase yields (see chapters 7 and 8) will permit a better alignment of additional supplies with future demand. Notes 1. Vinod Thomas and others, "Colombian Agriculture: Selected Issues and Some Directions for Strategy," Report no. 4275-CO (Washington, D.C.: World Bank, Latin America and the Caribbean Country Programs, Colombia Division, 1983). 2. In addition to the coffee boom, there was also rapid growth in drug trade during the 1970s, although assessment of it remains difficult. In Roberto Junguito and Carlos Caballero, La otra economia," Coyuntura economica, vol 8 (December 1978), pp. 103-39, drug exports were estimated to have grown from about 5300 million to $850 million between 1974 and 1977 (dollar amounts are in OVERVIEW 23 U.S. dollars unless another currency is specified). This growth would be equivalent to an additional 0.4 percent average annual growth in GDP, or an increase of nearly 10 percent in the growth rate, during that period. 3. The recession in Venezuela and increasing import restriction in that and other neighboring coun- tries during the early 1980s seriously hurt Colombian exports. 4. A more detailed discussion of the economy is given in World Bank, Colombia: Economic Develop- ment and Policy under Changing Conditions (Washington, D C , 1984). It should be noted that some of the calculations involving national accounts data obtained from the Central Bank in that study and the present one differ somewhat from the newly published data of the Departamento Administrativo Na- cional de Estadistica (DANE, the National Department of Statistics). Greater use of the data base of DANE could be made in future work. 5. See Gabriel Montes and Ricardo Candelo, "El crecimiento industrial y la generacion de empleo en Colombia Entre la substituci6n de importaciones y la promoci6n de exportaci6nes," Revista deplanea- ci6n y desarrollo, vol. 12, nos. 1 and 2, (January-June 1981), p. 87; and Francisco Thoumi, "Interna- tional Trade Strategies, Employment and Income Distribution in Colombia," in Trade and Employment in Developing Countries, ed. by Anne 0 Krueger and others, vol 3 (Chicago: University of Chicago Press, 1981). 6. Details of this and other trade-related observations can be found in tables SA-1-10. 7 See Roberto Junguito and Diego Pizano, 'Primary Products in Latin America," in Lattn America and the New International Order, ed. R Ffrench-Davis and Ernest Tironi (London: Macmillan, 1982) 8. The main agricultural export commodities, other than coffee, are cotton, sugar, bananas, flowers, leaf tobacco, livestock products, fish, and rice. Agricultural imports are defined here as outputs of agri- cultural origin, and they exclude processed foods and beverages as well as agricultural inputs such as fertilizer 9. See Jorge Garcia-Garcia, "Aspects of Agricultural Development in Colombia" (Bogota, April 1983, processed). Exportables are defined as coffee, rice, sugarcane, leaf tobacco, beans, bananas, ses- ame, cotton fiber, flowers, cattle, and sugar. Importables consist of cereals other than rice, green peas, other vegetables, other fruits, soybeans, cocoa, and milk products. Domestic commodities are brown sugar, tubers, tomatoes and other vegetables, plantains, oilseeds, noncotton fibers, hogs, sheep, horses, poultry, and eggs 10. A decline in the number of Colombian pesos paid per unit of foreign currency is defined as a nominal appreciation of the peso. After adjustments have been made for differential movements in price levels in Colombia and externally, however, a measure of real movements in the exchange rate is ob- tained. 11. Although there are parallels, the monetization of export receipts and the fiscal effects of illegal exports of drugs would be quite different from those of coffee. Of course, the domestic social and sociological aspects of coffee and drug exports are worlds apart. 12. The measures of relative prices of exports and imports would by and large be representative of the trends for exportables and importables, respectively, as well Therefore, often a finding concerning the competitiveness of traded goods is generalized for tradables in this book. 13. See "Evaluaci6n de Ia estrategia de exportaciones nuevas, 1979-1982" (Bogots: Fondo de Pro- moci6n de Exportaciones [PROEXPO, the export promotion agency], 1982). 14. James Hanson, "Inflation and Imported Input-Prices in Some Inflationary Latin American Econ- omies" (Washington, D C.: World Bank, Industry Department, 1982, processed); Laughlin Currie, "La demanda de dinero y la velocidad ingreso de la moneda en Colombia, 1960-80," Desarrolloy Sociedad, no. 6 (July 1981) 15. Juan Carlos Jaramillo and Armando Montenegro, "Cuenta especial de cambios: Descripcion y analisis de su evoluci6n reciente," Ensayos sobre politica econ6mica, no.2 (September 1982), pp 109-86; and Jose Antonio Ocampo and Guillermo Perry, "La reforma fiscal, 1982-8 3," Coyuntura econdmica, vol. 13, no. 1 (March 1983), pp. 215-64. 16. A Dutch-disease type of model is formally explained in appendix A. See W. Max Corden and 24 MACROECONOMIC POLICY AND AGRICULTURE J. Peter Neary, "Booming Sector and De-Industrialization in a Small Open Economy," Economic Jour- nal, vol. 92 (December 1982), pp. 825-48; Arnold C. Harberger, "Dutch Disease: How Much Sick- ness, How Much Boon?" Resources and Energy, vol. 5 (March 1983), pp. 1-20, and Sebastian Edwards and Masanao Aoki, "Oil Export Boom and Dutch-Disease: A Dynamic Analysis," Resources and Energy, vol. 8 (September 1985). 17. See Hanson, "Inflation and Imported Input-Prices." 18. See, for a discussion, Edgar Gutierrez Castro, "Presentacion ante el Congreso Nacional sobre la Emergencia Econ6mica" (Bogota, 1983, processed); see also table SA-9. 19 See table SA-7. If dollar receipts are deflated by the declining international price index for manu- factured goods in 1981 and 1982, however, in constant dollar terms no significant change is indicated. 20. Underlying the base case are, among other things, assumptions of good OECD growth and a set of good domestic policies concerning the exchange rate, monetary, fiscal, and financial matters, and man- agement of external debt. 2 1. Even with faster world recovery, the success of Colombia's efforts to increase exports might be predicated on its ability to take into account the exchange-rate adjustments of its competitors (see chap- ter 2). 22. Thomas and others, "Colombian Agriculture." 2 Trade Policy and Export Promotion TRADE POLICIES that have a significant influence on export performance are the subject of this chapter. The superior performance of outward-oriented economies has been related both to competitive exchange rates and to other incentives, such as access to duty-free imported inputs for exporting firms. Successful outward-looking economies have been known for their policies to ensure similar incentives to pro- duction for export and home markets. Equally important have been quality con- trol, prompt delivery, and flexibility in composition of products in response to vari- ations in foreign demand. Furthermore, where import protection is not heavy, domestic production has had to compete with imports either for domestic sales or eventually for exports.' Export incentives in Colombia usually involve exchange rate management, ex- port loans, subsidies, and special facilities, as well as import policy. Import consider- ations having been set aside for chapter 3, a close examination will be provided of the way these incentives have varied with the passage of time. An important ele- ment of this analysis is the exchange rate, which since 1967 has been based on a crawling peg. In this chapter the way the exchange rate has varied in order to stimu- late exports (1967-75), the way it has been used for stabilization purposes (1976- 80), and the possibilities there are in the 1980s for restoring its function of export promotion will be reviewed. Factors That Affect Policy Choice In evaluating exchange rate policy, it is essential to distinguish between short- term fluctuations around long-term equilibrium levels and persistent disequilibrium levels. While short-term variations can be problematic, persistent deviation from a long-term or fundamental equilibrium is of greater concern. Purchasing power par- ity (PPP) suggests that changes in the nominal exchange rate should be in accordance with differences between domestic and external inflation, but it does not account 25 26 MACROECONOMIC POLICY AND AGRICULTURE for equilibrium departures from PPP when the price of an important commodity, such as coffee, changes significantly in more than a transitory way. PPP also does not indicate the base exchange rate level for comparison. Measures of changing compet- itiveness as guides for the appropriate exchange rate are also arbitrary, because they require choosing base years when the country's costs and prices are considered to be in reasonable balance and when a basic equilibrium may be considered to have existed. One alternative is to examine normal years, when the real exchange rates permitted the matching of balances in the current account with trends in capital flows.2 The rate would have to provide sufficient incentives to productive sectors to allow a reasonable rate of growth in a framework of balance of payments viability given external conditions and domestic fiscal and monetary conditions. Wide fluctuations in the country's external terms of trade have made it difficult to determine and maintain a long-term equilibrium exchange rate in Colombia. Most of these changes have been the result of variations in world prices of coffee- coffee representing some 56 percent of goods exported during the period 1970- 82-which in turn have affected the level of the exchange rate the authorities con- sider sustainable for noncoffee commodities. During the 1950s and through the mid 1960s multiple exchange rates were used, with rates for noncoffee exports- sometimes separate rates for imports and for petroleum and other products-at times floating and often pegged at a level different from that for coffee. Instability and unpredictability of world coffee prices have also discouraged policymakers in the past from seeking long-term equilibrium rates, using import controls instead and export subsidies for commodities other than coffee. Since 1967 multiple rates have been abandoned, and the approach to an equilibrium rate has been more successful, although not always without difficulties. With the trade reforms initiated in 1967, the peso depreciated significantly, reaching an equilibrium by the mid 1970s. The balance of payments was in reason- able equilibrium from 1973 to 1975. Commodity prices, especially those of coffee, were not abnormal in that period, and Colombia was still self-sufficient in petro- leum, so there was no significant oil price effect. In fact, the terms of trade were about the same during 1973-75 as in 1983 (see table 2-1). These considerations support the use of the mid 1970s as a base for measuring the real exchange rate, although this will clearly need to be a flexible policy, to be revised upon the appear- ance of new evidence. Past Policies and Exports During the 1950s and through the mid 1960s, the peso was chronically overval- ued from the point of view of competitiveness of noncoffee exports, effectively constraining export diversification.' Although isolated attempts at stimulating ex- ports had been made before, it was not until 1967 that a coherent set of measures TRADE POLICY AND EXPORT PROMOTION 27 Table 2-1. Indicators of Performance in the External Sector, 1970-84 Current account Net balancel' capital Terms inflow bc of Million (million Change trade, U.S. Percent U.S. in Year (1975 = 100) dollars of GDP dollars) reserves 1970 99.0 -339 -48 337 -2 1971 92 5 -484 -6.4 409 -75 1972 97.3 -214 -2 5 406 192 1973 103.6 -80 -0.8 305 225 1974 108 1 -384 - 3 0 20 -364 1975 100.0 -98 -0 7 237 139 1976 140 3 210 1 4 352 562 1977 1946 449 2.3 403 852 1978 145.3 357 1 5 253 610 1979 129 8 562 2.0 675 1,237 1980 132.0 104 0 3 1,206 1,310 1981 111.3 - 1,722 -4.7 1,936 214 1982 107.8 -2,885 -7.4 2,146 -739 1983 110.4 -2.826 -7.3 1,014 -1,812 1984, 115.0 - 1,870 -5.1 587 - 1,283 a. Unit export price divided by unit import price, from IFs data. b. From table SA-8. c. The difference between the current account balance and the change in reserves, which means that errors and omissions are included. d. A minus sign signifies a drop in reserves. e. Preliminary estimate. Sources Banco de la Republica, IMF, and World Bank data. was undertaken to promote nontraditional exports-that is, exports other than cof- fee and petroleum. The 1967 Trade Reforms The most important element of the new policies was the introduction of a crawl- ing peg exchange rate system.' A package of export incentives was also introduced, including fiscal incentives-Certificados de Abono Tributario (cATS); concessionary credits for export-related activities from the Export Promotion Fund (PROEXPO credit); and an expanded and more effective import-export regime (Plan Vallejo). The new policies represented an attempt to compensate for the distortions in rela- tive prices generated by the import-substitution effort. Together with the favorable development of world trade, this shift in policy emphasis brought about impressive results. 28 MACROECONOMIC POLICY AND AGRICULTURE Between 1968 and 1974 manufactured exports, in current prices, increased from S58 million to more than $390 million, and their share in total exports rose from 8 percent to 28 percent.' Industrial value added increased 7 percent a year in real terms, and the growth in manufacturing employment of 8.5 percent a year reached unprecedented levels, largely as a result of the labor-intensive nature of the leading export subsectors-textiles, apparel, footwear, and leather products. This growth of manufactured goods and other minor exports also succeeded in easing the foreign exchange problems that had plagued the economy in earlier years. About a third of Colombia's export growth during 1967-75 was attributed prin- cipally to the sharp expansion in world trade of manufactured goods, and the re- maining two-thirds to the increased competitiveness of Colombian industry in in- ternational markets. This has in turn been linked to adjustments in the real exchange rate to reflect the effective value of the CAT export incentive, the effective value of subsidized PROEXPO credits, and the conditions of the import duty draw- back scheme (Plan Vallejo). A variety of products took advantage of these incen- tives, and several new commodities entered the export market. A number of im- pressive results can be cited: Colombia's textile exports grew faster than those of Taiwan, Hong Kong, and Singapore during 1970-75; exports of cut flowers, be- ginning from negligible levels in the mid 1960s, climbed dramatically to reach $19 million in 1975 and 5112 million in 1982; by 1978 Colombia had become the world's leading exporter of children's books, exporting $58 million worth, up from $2 million in 1970. Several developments through the 1970s might not have been predicted at the beginning of that decade: cheese, bananas, frozen shrimp, footwear, refrigerators and stoves, work gloves, and false teeth are some of the products exports of which had reached impressive levels by 1980. These examples suggest that a favorable export climate can produce remarkable results, not all of which can be anticipated. Exports to the Andean Group have been roughly 15-16 percent in recent years, and this share has not changed significantly since the mid 1970s. Protection of agricultural imports in other countries is a potentially limiting factor for Colombian exports of certain commodities to certain countries or groups of countries. In the aggregate, however, there appears to be head room to increase Colombian exports; the issue of export markets, however, needs to be examined further. Nonprimary exports benefited more from the exchange regime established dur- ing 1966-70 than did products based on natural resources, but both were encour- aged to take advantage of a rapidly growing world market. This approach was continued in the 1970-74 development plan. The resultant growth reflected the use of excess capacity created during the period of substantial investment, 1962- 66, and a judicious use of expansionary exchange-rate and aggregate-demand poli- cies. Growth also reflected an improvement in allocation of resources, however. For example, no new and expensive import-competing projects were initiated. Growth in agriculture, particularly of export-oriented commodities, accelerated while output of less efficiently produced crops-such as wheat-fell sharply. These TRADE POLICY AND EXPORT PROMOTION 29 production effects, particularly the improvement in agricultural performance, also contributed to reduction of inflation. Developments during the 1970s Most of the thrust of the 1967 reforms was maintained in subsequent years, but their effectiveness has varied with changes in world conditions and occasional de- lays in policy responses. In the second half of the 1970s, an externally generated increase in the money supply gave rise to higher rates of inflation: world coffee prices soared and unrecorded exports and net capital inflows grew rapidly, while recorded imports as a percentage of GDP did not expand and aggregate real purchas- ing power rose significantly. In January 1977 the administration initiated a stabili- zation package, which consisted of sharp rises in reserve requirements, including those for foreign borrowing, the reimposition of prior deposits for imports to re- duce the money base, the use of exchange certificates to delay the monetization of export receipts, a further tightening of fiscal policy, and a reduction in the rate of depreciation of the peso; efforts to contain inflation thus met with qualified success. A more rapid increase in selected imports than actually occurred could have, in addition to absorbing the rising liquidity, allowed the country to grow faster and enjoy efficiency benefits of liberalization, although the timing of such a policy dur- ing a coffee boom is not without problems. Smuggled imports, by all accounts, did increase. Recent Developments and Outlook While international coffee prices began to decline from their post-World War II peaks of 1977, the government's budget deficit began to climb steadily in the sec- ond half of the 1970s, contributing to inflation. Public investment and expendi- tures had been reduced by the stabilization efforts of the 1974-78 administration, and the new administration (1978-82) sought to reverse this trend. Both current and capital expenditures grew in 1979 and 1980, but these contributed to an in- crease in the fiscal deficit, in the absence of a matching tax effort. A large part of this deficit was financed by the use of the Special Exchange Account, a part of it repre- senting the creation of money. The Central Bank's deficit-financing operations were estimated to account for more than 90 percent of the expansion in the money base during 1981, a sharp change from the preceding five years, when public sector borrowing had been negligible. Once the accumulation of reserves had been re- versed in the 1980s, the rising fiscal deficit became a reason for a high rate of inflation. A growing government deficit squeezes the private sector out of the do- mestic credit market, with particularly serious effects given the present difficult circumstances of the international financial markets. The recession hurt worldwide exports, which grew 1.5 percent in 1980, stag- nated in 1981, and declined some 2 percent in 1982. Only a few developing coun- 30 MACROECONOMIC POLICY AND AGRICULTURE tries managed to increase exports, mainly of manufactured goods, during this pe- riod by improving productivity and quality.' Colombia's exports also declined in constant prices during 1980-82. Agricultural exports other than coffee fell from $643 million in 1980 to $590 million in 1981, $507 million in 1982, and an estimated $420 million in 198 3 (see table SA-8). A recovery of the world economy and a depreciation of the exchange rate and export promotion efforts would be vital to Colombia's export performance, as estimates of better performance in 1984 and early 1985 already suggest. Competitiveness and the Real Exchange Rate Alternative measures of competitiveness can be used in an attempt to capture changes in a country's production costs-evaluated at a suitable nominal exchange rate (NER)-in comparison to international changes. In this section estimates of the way domestic and international inflation have differed in the course of time will be presented; that the measures presented would capture changes in production costs only in a broad sense has been recognized. In addition, estimates of the real ex- change rate (RER) are also presented. The RER captures the effect of changes in the nominal exchange rate in offsetting the difference between domestic and external inflation. Colombia's crawling peg is defined by the rate of exchange between the Colom- bian peso and the U.S. dollar. Traditionally, in calculations of the RER comparisons of the peso and the dollar have been used, adjusting the nominal rate by the differ- ence between domestic and U.S. inflation. Such comparisons, however, would still have to be modified to the extent that the peso, like some of the other principal currencies, floats in relation to the dollar. Depending on the relation between the dollar and these currencies, a particular change in the RER, as indicated by the rate of exchange between the peso and the dollar, may or may not be reflected by a similar change of the peso in relation to these other currencies. In particular, when the dollar has appreciated against other currencies-as in 1984-the peso has also, other things being equal, tended to appreciate in real terms against these currencies. A government policy to depreciate the peso-dollar rate in this case may not depreci- ate the peso in relation to a basket of the relevant currencies. The exchange rate movements in real terms against these currencies can be accomplished by construct- ing an index of the RER with respect to a basket of currencies of the trading partners, taking into account their rates of inflation (see table 2-3). Estimates of the Real Exchange Rate Between 1970 and 1975, the rise of Colombian prices in U.S. dollars was signifi- cantly slower than the rise of external prices. The opposite has been true since 1975, as shown in table 2-2. The RER indexes in table 2-3 use estimated 1975-84 noncoffee trade weights-that is, export and import-for fifteen most important TRADE POLICY AND EXPORT PROMOTION 3l Table 2-2. A Comparison of Colombian Inflation with International Price Movements, 1970-84 (1975 = 100) Colombian Colombian price price index in Official index in Colombian rate of US. International Year pesosa excbangeb dollars' price index' 1970 44.2 59.5 74.3 55.6 1971 47 8 64.4 74 2 58.5 1972 54.2 70.8 76.6 63.7 1973 65.4 76.4 85.6 73.9 1974 81.4 84.5 96.3 90.0 1975 100.0 1000 100.0 100.0 1976 120.2 112.3 107.0 101.4 1977 160.0 119.1 134.3 111.4 1978 188.4 126.5 148.9 128 3 1979 235.0 137 9 170.4 145.3 1980 297.4 153.1 194 3 159.3 1981 379.1 176.4 214.9 160.2 1982 472.1 207.4 227.6 155 5 1983 565.6 255.3 221.5 153.6 19841 656 7 326.2 201.3 150.8 a. Consumer price index. average for the period b). Peso price of a dollar, average for the period. c. Column I divided by column 2. d Manufactured unit value index (muv) estimate, July 1985 e. Preliminary estimate. Sources. irs and World Bank data. partners.7 Two of the more traditional measures of the RER are also presented. Both measure the peso-dollar exchange rate, one using the U.S. inflation wholesale price index, the other using the U.S. consumer price index. The 1975-82 appreciation- that is, a decline in the RER-iS less serious if it is based on the latter dollar-peso measure of RER, as shown in the third and fourth columns of table 2-3. The mea- sure against the basket-the second column in table 2-3-however, shows a more serious level of appreciation, on account of the appreciation of the dollar against other major currencies. Effect of Special Incentives The provision of special incentives for exports-CAT and, more recently, Certifi- cado de Reembolso Tributario (CERT, tax reimbursement certificate) and PROEXPO credit-was increased following 1975 in order to offset a part of the appreciation of the RER for exports. During 1975-81 the RER against the U.S. dollar-including 32 MACROECONOMIC POLICY AND AGRICULTURE Table 2-3. Indexes of the Real Exchange Rate of the Peso, 1970-84 (1975 = 100) Against the U.S. dollar Against Wholesale Consumer a price price Year basket' indexes indexes 1970 81.1 108.3 98.2 1971 83.1 108.5 101.5 1972 86.4 105.1 101.4 1973 89.3 100.4 96.4 1974 94.7 96.7 94.9 1975 100.0 100.0 100.0 1976 97.3 95.5 98.7 1977 83.4 84.8 83.7 1978 83.5 82.6 81 3 1979 81.8 79.2 79.0 1980 82.0 80.8 78 7 1981 78.1 81.9 78.5 1982 73.4 78.2 78.7 1983 75.2 80.1 83.5 1984, 81.8 88.6 95.8 a. Estimated 1975-84 noncoffee trade weights for Colombia's fifteen most important trading part- ners were used. A foreign-country index of wpis was used to reflect external inflation, and the Colom- bian cpi was used to reflect domestic inflation. For 1983 and 1984, estimated weighted averages of official and parallel exchange rates and inflation rates for Venezuela were used. b. Preliminary estimate, based on incomplete data. Source: Computed from 11s data. changes in CAT-Is estimated to have declined (depreciated) by 19.8 percent, while the RER (excluding CAT) against the U.S. dollar fell 22 percent. In 1983 and 1984 CAT rates and the new CERT rates were significantly increased, as shown in table 2-4. Details of the rates for 1978, 198 1, and 1983 before the August increase are given in table SA-12; the August 1983 reform raised the CAT rate to 20 percent for 265 products, while in 1984 many of these rates were increased further through CERT. With the CAT increases applicable from 1978 to mid 1983 on average, about 7 percentage points of the appreciation may have been corrected from the point of view of the exporters.' Such calculations, however, do not account for changes in export incentives in other countries. Other limitations of considering additional special incentives will be noted below. Exchange Rate Depreciation and Special Incentives A policy of faster depreciation of the exchange rate differs from one of higher export subsidies in several respects. First, a nominal depreciation of the peso that succeeds in generating a real depreciation raises the prices of all tradables as a TRADE POLICY AND EXPORT PROMOTION 33 Table 2-4. Aggregate Weighted Average of Tax Reimbursement Certificate (CAT) and Tax Reimbursement Certificate (CERT) Rates, 1978, 1981, and 1983-84 (percent) Weighted average rate Year Total Agriculture' 1978 3.7 n.a. 1981 3.5 n.a 19831 11.3 9.8 1984' 12 3" 10.6 1984' 15 0 n.a. n.a. Not available. a. Percentage of export value, excluding coffee. b. Before the August 198 3 increase. c After the August 1983 Increase, effective in 1984; a rough estimate based on the projected total additional fiscal expenditure on CAT and the 1982 value of exports other than coffee d. Preliminary estimate. e. An estimate of the average rate of CERT, which in 1984, by Ley 48, replaced CAT. Source: Table SA- 12 group-exportables and importables-in relation to the prices of nontradables. An increase in the export incentives alone excludes the import-competing industries. To avoid the latter effect, import tariffs can be increased, but only at some welfare cost, as will be seen in chapter 3. Differential rates of export subsidies and import tariffs can produce substantial inefficiencies and welfare losses. Second, export in- centives promote only trade in merchandise, while an increase in the real exchange rate affects invisibles as well. Third, commercial policies-export subsidies and im- port tariffs-do not directly affect the peso value of foreign assets, whereas a depre- ciation of the peso would increase it. Finally, an expected faster rate of the crawl, which is the government policy, will tend to raise the nominal interest rate.' Com- mercial policies do not produce such effects on the interest rate. In practical terms, further increases in export subsidies may not be feasible in view of various trade agreements. These incentives also have significant negative consequences for public finance. Where the diagnosis points to an overvaluation of the exchange rate, the primary policy tool would need to be the exchange rate, and the aim of accompanying policies should be to offset potential side effects, such as those on external debt repayments and the interest rate. Once the real exchange rate has attained an equilibrium level, therefore, special incentives might gradually be replaced by a further depreciation of the peso. Stimulating Noncoffee Exports While noncoffee exports expanded 5 percent a year in constant pesos during 1970-75, they increased only 2 percent during 1975-82. In agriculture, the share 34 MACROECONOMIC POLICY AND AGRICULTURE of noncoffee exports gradually increased from 18 percent in 1970 to 24 percent in 1982. In dollar terms, the most important agricultural export items are bananas, flowers, cotton, sugar, and livestock products, which constitute about 25 percent, 2 1 percent, 17 percent, 14 percent, and 10 percent, respectively, of the value of noncoffee agricultural exports; less important exports are tobacco, rice, and fish. World Bank projections as of January 1985 indicate little improvement in coffee prices in constant dollar terms through the rest of the decade (see table 2-5). For the rest of the 1980s, banana prices are projected to show some decline. The prices of both sugar and cotton are projected to rise significantly from their depressed 1982 levels; substantial increases are expected in the price of sugar. The price of tobacco is projected to decline during the rest of the decade, while a substantial recovery is expected in the price of rice. Food prices are projected to remain well below the levels registered during the 1970s, particularly 1970-77. These projections suggest modest increases during the 1980s in the prices of the agricultural products now exported by Colombia. The country can continue to expand banana exports rapidly and can capture additional shares of the cotton and sugar markets. Significant increases in cattle and beef are not unreasonable to ex- pect, along with efforts to take advantage of the projected recovery in rice prices. The industry believes, with good reason, that a continued expansion in exports of flowers is possible despite the recent increase in U.S. import duties on this category. A tentative review of various agricultural commodities on a product-by-product basis suggests that while coffee exports in constant dollars may increase more than I percent a year during 1985-90, noncoffee agricultural exports could increase about Table 2-5. Price Projections for Important Agricultural Exports, 1976 and 1981-90 (1983 constant dollars per metric ton, unless otherwise indicated) Actual' Projection Export commodity 1976 1981 1982 1983 1984 1985 1990 Coffee" 477 271 305 290 324 311 291 Bananasb 390 385 370 429 377 420 295 Cotton 256 177 158 185 182 148 173 Sugar 386 359 178 187 117 74 315 Beef 240 237 236 244 231 225 246 Tobacco 2,138 2,253 2,381 2,245 2,035 2.303 2,182 Rice 386 463 289 277 257 230 339 Food' 110 89 70 74 75 64 82 a These are not actual prices of Colombian exports but World Bank estimated average prices of these commodities; they could therefore differ from Colombian prices quoted elsewhere in this book-for coffee, for example. b. Cents per kilogram. c. A weighted index, with 1977-79 as base. Source: World Bank estimates as of July 1985. TRADE POLICY AND EXPORT PROMOTION 35 5 percent a year. These projections, however, do not account for a faster rate of crawl in the exchange rate than at present, or for the development of nontraditional exports with a favorable export climate in the country.'0 Fruits and vegetables could become significant exports in the future, with improved processing, packag- ing, and marketing and with the establishment of lasting export relations abroad. The excellent environmental conditions for growth in Colombia and its large re- serves of tropical hardwood forests indicate that the country might have a compara- tive advantage in producing and exporting such tropical hardwood products as lum- ber and plywood. The low level of present exports in these categories and their decline during the past decade suggest that this possibility should be examined closely. An effort has been made here to separate out empirically the effects of some primary variables, including the real exchange rate on noncoffee exports, using quarterly data for 1970-81 (see appendix B).'' While world economic growth has a stronger long-term effect, the findings show that the effect of changes in the level of the real exchange rate on the volume of noncoffee exports is significant." It is also noteworthy, however, that real changes in the exchange rate are needed, which means that in order for a nominal depreciation to be effective, it should not be accompanied by commensurate increases in the domestic price level. Colombia's exports depend strongly on the commercial policies of its trading partners. The protectionist behavior of some industrial countries and neighboring Latin Ameri- can countries can critically affect the possibilities for expansion of noncoffee ex- ports. The results suggest that the long-term elasticity of the volume of noncoffee ex- ports in Colombia with respect to the real exchange rate lies between 1.2 and 1.7, which narrows down earlier estimates. The various estimates suggest different be- haviors of the lag structure of this elasticity. While some estimates indicate a higher short-run response, others suggest a more nearly even distribution of export re- sponse. To arrive at a more definite answer will require additional work. The long-term elasticity of the volume of Colombian exports with respect to the world level of activity is typically large, ranging from 5 to 9. Even though this is a wide range, it does suggest that to a large extent the recovery of Colombia's exter- nal sector will depend on the world economic situation, provided that movements in the Colombian exchange rate match those of other countries that compete for the same markets. These findings indicate that both domestic exchange rate policy and world economic activity will be important to export performance. Finally, the long-term elasticity of exports with respect to the domestic level of activity, holding external activity constant, is, as expected, negative. In most cases its coefficients are not significant, suggesting that changes in Colombian real income do not signifi- cantly affect the performance of noncoffee exports. Regression analysis indicates that a real depreciation of the peso of 10 percent against the U.S. dollar, holding constant the relation of the dollar to the basket, could be expected to increase noncoffee exports in real terms more than 10 percent 36 MACROECONOMIC POLICY AND AGRICULTURE above what they would otherwise be in the long term. A world economic recovery, without improvements in the Colombian exchange rate, on the other hand, may not ensure a revitalization of Colombian exports in the face of depreciation of the exchange rate by Brazil, Mexico, Chile, and other countries that produce similar exportables. Addressing the Appreciation of the Peso While some observers suggest that the appreciation should be corrected by accel- erating the rate of depreciation of the crawling peg, others have recommended a once-and-for-all devaluation." These alternative approaches can have quite differ- ent effects on factor prices, and they in turn can affect success in achieving a change in the real rate of exchange. In particular, the functions of real wages, expectations, intermediate goods, and the interest rate in determining the effect of changes in the NER on the RER need to be considered. The Role of Wages Even if all factor costs were held constant, a higher NER would increase the price of tradables in nominal domestic currency, and the cost of nontradables would be increased to the extent that they would contain tradable input components. The higher the share of these tradable components in nontradable production, the more domestic prices would rise and the less the effect of a higher NER would be in actu- ally raising the RER. More realistically, if prices of factors and inputs were allowed to adjust at least partially, the change in RER would generally be smaller. The impor- tance of wage adjustments would depend on the share of the wage bill in total production costs and in the type of wage negotiations already under way that affect the rate of wage increases in the face of a devaluation. Interest Rate Effects Underlying the analysis in appendix C is the recognition that Colombia's is a semiopen economy and that, consequently, both open economy models and closed economy models are inappropriate. Three alternative formulations for the determi- nation of the interest rate in a semiopen economy were developed and tested, using quarterly data for the period 1968-82. The results obtained are remarkably good and indicate that discrepancies between the domestic (nominal) interest rate and the world interest rate plus expected devaluation will be eliminated through time; that in one quarter, between a third and half of a unitary discrepancy between the do- mestic rate and the world rate plus the expected rate of devaluation will be cor- rected; that in six quarters an acceleration of the rate of devaluation of the crawling peg will be almost completely translated into an equivalent increase in the domestic TRADE POLICY AND EXPORT PROMOTION 37 rate of interest; and that an excess supply of (real) money will exercise significant negative pressures on the nominal interest rate-that is, that there is a liquidity effect. Appendix C suggests that an increase in the rate of depreciation of the crawling peg will be quickly translated into an equiproportionate increase in the domestic nominal rate of interest. If this faster rate of crawl, moreover, produces a desired higher real exchange rate, the domestic rate of inflation will increase by less than the acceleration of the peg and will produce an increase in the real interest rate. During the period of the expected acceleration of the crawl, the interest-rate effect can hurt, in particular, nontradable industries such as the construction sector, in addition to industries with high ratios of debt to equity. With respect to the tradable goods sectors, the higher rate of the crawl will produce effects that will then work in the opposite direction. To the extent that a higher crawl results in a higher real ex- change rate, on the other hand, the level of activity in these tradable sectors would increase, possibly dominating any depressing effect from the higher interest rate. The domestic nominal interest rate is observed to move broadly in line with the world interest rate plus the expected devaluation in Colombia (see appendix C). An increase in the rate of the crawl may therefore be expected to produce a higher nominal rate of interest, affecting the cost of capital-that is, the rental rate of capital-and exerting upward pressure on the prices of nontradable goods; to that extent, there will be a tendency for the acceleration of the crawling peg to be less effective. Some observers have taken the view that the rising interest rates will generate a higher rate of inflation through the cost-push mechanism: such an effect is possible in the face of the persistent stickiness of certain costs and prices. An empirical investigation of this issue as it pertains to Colombia needs to be made. The main aspect of this interest rate problem is that a higher rate of the crawl can affect the expected rate of depreciation and produce higher interest rates during the period of acceleration of the peg. It is possible that a large one-step devaluation, if perceived by the public to be large enough, can remove expectations of higher rates of devaluation, leaving the domestic interest rate unaffected. If the public is not convinced by the magnitude of such a one-step devaluation, however-and it usu- ally has not been in Latin American countries-there may be no such mitigating effect. Effect of Monetary Disequilibrium The analysis presented so far has been focused on cost aspects, ignoring demand considerations. The real effect of a nominal devaluation, however, will also depend on, among other things, the behavior of monetary policy. In particular it is ex- pected that an excess supply of money will be translated into an excess demand for nontraded goods, which will then exert an upward pressure on their prices. In that sense the effect of a nominal devaluation on the RER could also be offset by an easy monetary policy that would produce an excess supply of money and an increase in 38 MACROECONOMIC POLICY AND AGRICULTURE the nominal price of nontradables. In order for a devaluation to affect the RER, special attention must be given to both wages and monetary behavior. Export Promotion Efforts Although coffee dominates exports, Colombia sells a wide range of commodities abroad. Manufactured products constitute about twice the value of agrobased com- modities, excluding coffee, and the share of these two categories combined in total exports has gradually increased. Colombia exports to a large group of countries. The North American share of Colombian sales declined from about 37 percent in 1970 to 22 percent in 1982, while the European share declined marginally from 49 percent to 48 percent during the same period; these shares have been increasing since 1982, however. Developing countries have been buying ever larger amounts, and by 1982 Central America and the Caribbean claimed 6.5 percent and Asia, including Japan, 4.4 percent of Colombia's exports. The Function of PROEXPO PROEXPO, the Export Promotion Agency, following policy guidelines of the Na- tional Council of Economic and Social Policy (CONPES), is in charge of export pro- motion activities. An autonomous institution subsidiary to the Central Bank, PROEXPO is able to maintain independence from most other public institutions; its primary function is the implementation of export promotion policies, particularly those concerned with the provision of export credits. The central executing agency dealing with most trade-related policies, however, is the Foreign Trade Institute (INCOMEX), which directs the execution of government policies concerning imports and exports, through instruments such as the provision of licenses for importing and exporting. Apart from reimbursements received from its past credits earned in the course of sixteen years of operation, the budget of PROEXPO benefits from a 5 percent import tariff. Its total annual budget is roughly S800 million. About 92 percent of its expenditures consist of the provision of credit for exporters and investments in a variety of activities. A number of services-identification of exporters, exportables, and export markets and the provision of technical assistance-are also offered. In the administration of PROEXPO better definition of functions and areas of re- sponsibility and improved internal coordination of activities are needed. The agency could make better use of the commercial attaches abroad and reduce its excessive dependence on the Central Bank. Statistical information needs strength- ening as well. In general, promotional activities need greater attention; some efforts could be channeled away from pure financing of exports by the government and into export promotion activities. Support of exporters needs to be stepped up through more studies of export potential, a strengthening of relations with the ex- TRADE POLICY AND EXPORT PROMOTION 39 port community, better coordination of Colombia's commercial offices, and a more active share in the development of new export lines by PROEXPO. On the policy side, a gradual elimination of import restrictions, which now constrain the supply of inputs needed for exports, and reduction of effective protection across the board will be needed to sustain rapid export growth. The increase in import restric- tions is likely to hurt export activities to some extent despite the import duty draw- back provided by Plan Vallejo. In 198 3 about half of noncoffee exports did not or could not take advantage of this scheme. A reorientation of the free trade zones toward export promotion and the attraction of foreign investment in export indus- tries would be helpful initiatives. Foreirgn Investment Historically, the development of natural resources has not had much to do with growth despite the country's rich agricultural and mineral base. This lagging devel- opment was caused in part by the lack of processing industries, in part by high transport costs, and in part by the authorities' unwillingness to support investment projects of either private or foreign participation in these areas fully. Traditionally, the country has had a lukewarm attitude toward foreign investment. In addition, Colombia was self-sufficient in petroleum until 1976, reducing the need to come to grips with foreign investment in this area. The oil crisis and the need to import petroleum during the second half of the 1970s changed that situation substantially, and Colombia is now seeking development of its hydrocarbon and petrochemical resources, as well as of coal and nonfuel minerals, with significant foreign participa- tion. Recent government policy actions, moreover, particularly those that would allow greater profit remittances across the board and gradually dismantle regulation of private sector borrowing from abroad, signal a policy shift toward allowance of greater foreign participation in the private sector. Policy Conclusions Colombia's crawling peg has by and large been successful in the long term, al- though the real exchange rate remained significantly appreciated during 198 3-84. Using 1975 as the base period, the RER index measured against a trade-weighted currency basket stood at 75 in 1983. During 1983-84, the government pursued a policy of accelerated crawl of the exchange rate against the U.S. dollar. Given the appreciation of the U.S. dollar against other major currencies, however, a part of this potential gain has been offset. Nevertheless, a significant real depreciation was achieved in 1984, and this has accelerated in 1985. The government's policy is now to eliminate the overvaluation of the peso by means of an accelerated crawl. Although the foregoing calculation implies a signifi- 40 MACROECONOMIC POLICY AND AGRICULTURE cant lag to be adjusted for, the real gain in 1984 was nevertheless estimated to be the largest in a single year since 1967, when the crawling peg was established. While the nominal rate of the crawl could be increased even further, its effect on the real rate would depend on the way domestic inflation would be affected and on the behavior of external inflation: the NER is a policy variable, but the RER Is only partially so. It is expected that in 1985 more depreciation of the RER than that achieved in 1984 will occur. In the future, while a depreciation of the U.S. dollar against the major currencies of the world would assist in improving the RER rapidly with respect to the basket, it might nevertheless be important to pay special attention to the real rate of exchange between the peso and the dollar as well: a good part of the exports are denominated in U.S. dollars and the U.S. market may be crucial for the future growth of agricul- tural and industrial exports from Colombia. In determining future adjustments of the peso, it may be advisable to account also for a gradual phasing out of special export incentives. A direct comparison with exchange rate movements of Colom- bia's export competitors may also be necessary. In principle, a nominal depreciation will raise the RER only if there is a price variable such as nominal wage that is not tied to the nominal exchange rate, if the nominal money supply is not fully adjusted to accommodate higher prices, or if there are accompanying reductions in some component of aggregate demand such as government expenditures. Nominal wages are neither rigid nor completely flexi- ble in Colombia. This means that real wages can be flexible, and a nominal depreci- ation of the peso can, in principle, produce a real depreciation. Negotiations in 198 3 brought about an increase of 22 percent in the urban areas and 28 percent by mid 1984 in the rural areas in the nominal minimum wage for 1984, significantly higher than the 1983 rate of inflation and the average for 1984 and ahead of the improvement in productivity, at least as far as the rural wage is concerned. Wages will need to be set with greater cognizance of a possible inflationary effect, unless they are closely tied to increases in productivity. Additionally, since the prices of nontradables are moderately flexible, a reduction in the fiscal deficit would be es- sential. Such fine tuning to affect the nominal prices of nontradables, while by no means easy, appears to be feasible in Colombia, as shown to some extent by the 1984-85 experience. With the domestic nominal interest rate estimated to move broadly in line with the world interest rate plus the expected depreciation of the peso, an increase in the rate of the crawl might tend to produce a higher nominal rate of interest domesti- cally. This interest rate complication is a factor that has supported proposals for a one-step devaluation, which, if perceived by the public to be large enough, can remove expectations of higher rates of devaluation, leaving the domestic interest rate unaffected. The government's present preference is for an acceleration of the crawl based in part on its good track record in Colombia and the recent success in depreciating the peso in real terms with this method. TRADE POLICY AND EXPORT PROMOTION 41 World economic conditions and the RER have significant statistical effects on non- coffee exports. Exports from agriculture have been hit in recent years by low inter- national prices and surplus stocks. Furthermore, penetrating into agricultural mar- kets and crossing trade barriers with a number of important commodities is problematic. At the same time, the country needs to return to more aggressive export promotion efforts. Additional sector work is proposed to examine con- straints to domestic policy, institutional and legal frameworks for the rapid increase of exports, and market conditions and impediments to entering new areas on a product-by-product and country-by-country basis. Also needed are a gradual elimi- nation of import restrictions, a reduction of effective protection across sectors, and, as the government has initiated, improvements in the actual functioning and effi- ciency of Plan Vallejo. Notes 1. On trade policy and performance, see David Morawetz, Why the Emperor's New Clothes Are Not Made in Colomba (New York: Oxford University Press, 1981). 2. Examining normal years means excluding years in which there were unusual commodity prices, oil stocks, or other such special circumstances. Clearly basic changes in the domestic or world economy- such as significant variations in domestic productivity, the availability of a new resource at home, or permanent changes in the terms of trade of a country-can change the equilibrium level. See, for exam- ple, John Williamson. The Exchange Rate System (Washington, D.C . Institute for International Eco- nomics, September 1983). 3. C. F. Diaz-Alejandro, Foreign Trade Regtmes and Economic Development. Colombia (New York Columbia University Press for the National Bureau of Economic Research, 1976). 4. A detailed discussion of the exchange rate and other trade reforms is provided in Jan Peter Wogart and others, "Colombia: Manufacturing Sector Development and Changes in Foreign Trade and Finan- cial Policies," Report no. 4093-CO (Washington, D.C. World Bank, Latin America and the Caribbean Projects Department, Industrial Development and Finance 2, 1983); and Eduardo Wiesner, "Devaluaci6n y mecanismo de ajuste en Colombia," in Politica economia externa Colomba (Bogotia, 1978, processed) S. See Wogart and others, Colombia Manufacturing Sector Development." 6. The spectacular performance of some of the East Asian economies in the 1980s is worth mention- ing. Growth of GDP in Korea and Singapore ranged from 6 to 10 percent a year during the 1980s, and similar results were obtained in Malaysia and Thailand. Exports from Korea, Thailand, and Singapore have been increasing substantially in real terms during the 1980s. 7. For an earlier study of the subject, see Fernando Montes, "Principales determinantes de comporta- miento de las cuenta corriente durante la decada," in Ensayos sobrepolitica econ6mica, no. 2 (September 1982). 8. That is, (1.113 - 1.037)/1 037 = 0 07 3 For other studies see Jorge Ospina Sardi and Mauricio Carrizosa Serrana, "Evoluci6n y perspectiva del certificado de abono tributano (CAT)," in Revista na- cional de agricultura, no. 856 (November 1981); and FEDESARROi.LO, Coyuntura economtca, December 1983. 9. Given the option of investing externally, a broad equivalence has been observed between the Co- lombian interest rate, adjusted for the expected rate of depreciation of the peso, and U S. rates. 42 MACROECONOMIC POLICY AND AGRICULTURE 10. Korea has shown great flexibility in increasing its exports and capturing additional exports mar- kets, even when external conditions have been unfavorable. Between 1979 and 198 3, Korean exports are estimated to have increased some 50 percent in constant dollars. 11. On the relation between exports and the exchange rate, see, for example, Jos& Antonio Ocampo, "Politica econ6mica bajo condiciones cambiantes del sector external," in Ensayos sobre politica economica, no. 2 (September 1982), pp. 7-65; Mauricio Carrizosa, "El futuro de [a balanza comercial," in La economica colombiana en la decada de las ocbenta (Bogota: FEDESARROLLO, 1979); and Juan Jose Echavarria, "La evolucion de las exportaciones colombianas y sus determinantes," Revista del Banco de la Republica (Bogota, 1980). 12 For the statistical analysis a real exchange rate between the dollar and the peso was used as the relevant explanatory variable, so that the result could be compared with results obtained in earlier stud- ies. If an exchange rate defined in relation to a basket of currencies is used, however, the results are similar. 13. For the former suggestion, see Jose Antonio Ocampo, "En defensa de la continuidad del regimen cambiario," Coyuntura economica, vol.13, no. I (March 1983); for the latter, see Armando Montenegro, "La sobrevaluaci6n del peso" (BogotA: CEDE, University of Los Andes, April 1983, processed) 3 Import Policy for Growth and Stability THIS CHAPTER concerns changes in Colombian import regimes, focusing on the recent increases in import restrictions and associated economic problems. Import restrictions-consisting mainly of import licensing and tariffs-have periodically been increased in order to protect domestic producers in agriculture and other sec- tors, to prevent a deterioration of the balance of payments, or for both these pur- poses. The use of tariff increases as a tool for financing public sector deficits has also influenced import policy: tariffs have been increased to finance the deficits of public sector agencies. On the other hand, additional import restrictions have raised the level of domestic prices, have caused inefficiencies in allocation of resources, and have caused consumers to suffer losses in welfare. In recognition of these problems, a policy of opening up imports has been periodically followed, particularly during periods of high growth. Following a period of moderate liberalization after 1967, however, there was a drastic increase in restrictions after the third quarter of 1982, raising the level of import controls beyond that which existed before the opening up of 1979. Begin- ning in early 1983 import licensing and exchange controls were significantly in- creased, primarily to reduce the loss of international reserves. Even before the 1982-84 measures, roughly a third of imports were subject to prior license, and World Bank estimates of tariff rates showed average levels on items, excluding ex- emptions, of about 26 percent. A significant decline in imports in 1983-84 hurt production and exports in agriculture and in nonagricultural sectors.' On the other hand, these measures partly offset the disadvantages to import-competing sectors from appreciation of the peso. At this writing import policy has been undergoing further changes. The govern- ment is in the process of adopting a more outward-looking strategy than that repre- sented by the changes made in 1982-84. Selective opening up of imports has been initiated, with emphasis first on making available greater quantities of imported inputs for export production. The full effects of such a shift in policy are yet to be assessed. 43 44 MACROECONOMIC POLICY AND AGRICULTURE Long-Term Policy Directions Major policy tools have been quantitative restrictions consisting of licensing, prior deposit requirements, and import tariffs.' The rate of effective protection has been determined not only by the level and coverage of nominal tariffs and other import taxes and the extent and structure of quantitative restrictions within the import license regime, but also by the controls under which foreign exchange is allocated to authorized imports and by the real exchange rate of the peso, which has varied substantially. Severe import controls used in the 1950s and throughout much of the 1960s were attempts to foster industrialization by giving strong protection to the domestic production of manufactured goods. In 1966 many of the tariff barriers and quanti- tative controls on imported inputs and investment goods required to produce ex- port goods were reduced or removed. Unfortunately, these measures were not ac- companied by improvements in the real exchange rate or the real interest rate, and as a result the balance of payments was hurt and controls were eventually reim- posed. Under the two development plans of 1966-70 and 1970-74, a vigorous policy of export promotion, which eventually permitted a reduction in import con- trols through most of the 1970s, was pursued. The gradual move toward opening the economy favored employment by encouraging labor-intensive exports.3 Opening Up of Imports in the 1970s By the mid 1970s a process of liberalizing and simplifying the import regime had been initiated. Since then, its implementation and effects have varied widely in response to changing economic conditions and constraints imposed by arrange- ments under the Andean Pact. During 1967-82 as a whole, tariffs were gradually reduced, and a large number of items were moved from the prohibited and prior- license lists to the free imports list. Furthermore, between 1967 and 1982 varying importance was attached to the deposit required prior to receipt of the import permit: the requirement was eliminated in early 1976, reimposed in mid 1976, increased in 1977 and 1979, and reduced in early 1982.4 The participation by Colombia in the Andean Pact may have limited its ability to attract foreign investment and, to a lesser extent, to reduce protection. The Andean Pact involves some limits on foreign investment except in major energy and mining projects. There has been foreign investment in the petroleum sector and a few scientific mining projects. On the side of trade restrictions, Colombia's tariffs in many goods had fallen to roughly the average of the Andean Pact's common mini- mum tariff by 1979. Further reductions in protection were inhibited somewhat by the pact. On the other hand, Colombia did benefit from a rapid expansion of manu- factured exports to Venezuela and Ecuador following initiation of the pact, al- IMPORT POLICY FOR GROWTH AND STABILITY 45 though much of this growth can be attributed to the petroleum boom in the other two countries. Effect of Import Substitution In general, import substitution has stimulated the economy when output was depressed below normal levels. Import substitution came into conflict with export promotion, however, particularly as production capacity became more fully used and both sectors began to compete for the same scarce resources and bid up costs and prices. In sum, a general policy for import substitution is responsible for a variety of costs. A reduction in imports, while benefiting a specific import-compet- ing industry, increased costs and prices in other industries that use the import as an input.' On balance, import substitution tended to increase prices in the aggregate and to slow the growth of output in the long run. Nontraditional, non-resource- based exports, moreover, have been more labor-intensive in Colombia than import- competing industries. Thus, with growth returning to a normal level along a long- term trend, continued dependence on an import-substitution strategy reduced both the long-run demand for labor and the growth rate more than would a develop- ment strategy of concentrating new investment and resources in those industries in which the country has a comparative advantage in exporting or import substitu- tion, as was done in the period 1967-74. Long-Term Development of Imports Between 1970 and 1982, Colombian imports of merchandise increased more than sixfold, from $800 million to 55.4 billion. The acceleration was especially pronounced after the mid 19 70s (see table SA-9). The growth in imports during the 1970s was significantly greater than that registered during the 1960s; imports ex- panded 5.7 percent a year during 1970-80, but only 2.7 percent during 1960-70. As a proportion of GDP, legal imports of goods and n.f.s. grew during 1970-82 (see table SA-1). Not included in these official statistics, however, is a substantial amount of contraband. Smuggled imports of textiles in some recent years are esti- mated to have been as much as 20-25 percent of the value of local production; import smuggling has been estimated to be equal to 10 percent of official imports.' The shares of the broad categories-consumer, intermediate, and capital goods- have not undergone any significant change since 1975. During the period 1980-8 3 the category of raw materials and intermediate goods constituted about 52 percent of total imports, followed by capital goods (36 percent) and consumer goods (12 percent), which are not significantly at variance with the proportions recorded in 1970 (see table 3-1). It is also estimated that industrial inputs represented some 35 percent and industrial equipment 21 percent of total imports in 1982, while nondu- 46 MACROECONOMIC POLICY AND AGRICULTURE Table 3-1. Imports, by Economic Category, 1970, 1975, and 1980-83 (percent) Category 1970 1975 1980 1981 1982 1983 Consumer goods 10.3 11.3 13.3 12.8 12.6 10.8 Raw materials and intermediate goods 43 7 52.2 52.7 52 0 50.6 51.2 Capital goods 44.0 36.1 34.0 35 2 36 8 38.0 Unclassified 2 0 0.4 . . Total1 100.0 100.0 100.0 100.0 100.0 100.0 Zero or negligible. a. Before balance of payments adjustment Source: Table SA-10. rable consumer goods represented only 6 percent. As will be seen, food and agricul- tural outputs and inputs have constituted relatively small proportions. Imports of agricultural commodities as a proportion of agricultural value added are rather small, although they rose from about 2.2 percent in 1970 to about 3.5 percent in 198 3. As a proportion of total imports, the share of agricultural imports was more than 5 percent in 1982-83. Imports of certain categories, however, are more important than these aggregate measures suggest: in 1981, roughly 15 per- cent of the gross value of domestic output of cereals and nearly 10 percent of that of vegetables were imported. Imports of wheat, corn, sorghum, and barley are esti- mated to have amounted to 467, 120, 82, and 50 thousand tons of imports in 1982. Roughly 90 percent of domestic consumption of wheat is met by imports. Aggregate estimates of agricultural imports, nevertheless, contradict widely held beliefs concerning their relative importance and their effect on the domestic econ- omy. Nominal tariffs on foods, beverages, vegetables, oil, and livestock products have ranged from 18 percent to 38 percent, which on the average are lower than the rates on nonagricultural commodities. More important, several potential and actual imports are controlled by the Instituto de Mercadeo Agropecuario (IDEMA, the Ag- ricultural Marketing Institute), which has a monopoly on certain agricultural im- ports, and as a result these products receive higher nominal rates of protection than the tariff rates would imply. As will be elaborated in chapter 4, a combination of import controls and price supports by IDEMA has helped to maintain the domestic prices of several agricultural commodities at levels significantly higher than interna- tional levels. IDEMA has been exempt from import tariffs, and it has enjoyed a mo- nopoly on licenses to import certain commodities. It is noteworthy that domestic support prices have been significantly higher than international prices of corn, wheat, and sorghum, in which the country does not seem to have a comparative advantage, while the opposite has been true of rice and barley, which could com- IMPORT POLICY FOR GROWTH AND STABILITY 47 pete in international markets. Domestic producer prices have been broadly in line with the support prices for these crops, notwithstanding year-to-year variations. High and rising agricultural production costs have been related to rural real wages and to the cost of agricultural inputs that incur high port-handling and domestic transport costs, in addition to the import tariffs and restrictions on certain inputs. While import tariffs on fertilizer and some other inputs have been low (3-7 per- cent), other port and transport costs raise domestic prices well above border prices (see chapter 8). Tariffs on transport vehicles have been high, while the incidence of tariffs on agricultural equipment has varied, depending on whether the same prod- ucts are produced domestically. In evaluating protection for importables, therefore, the effect of increased input costs must be accounted for (see chapter 4). In the aggregate, a drive to protect agriculture from imports is unlikely to have any significant quantitative effect on the balance of payments or on domestic pro- duction, and such an effort will become unnecessary as the real exchange rate is improved. Throughout the long term, the foreign exchange savings from any major import-substitution effort in behalf of principal agricultural imports is likely to be modest. Wheat, corn, sorghum, and soybeans are now imported in significant quantities, but, as shown in table 3-2, while world prices of these commodities are expected to increase from their depressed 1982 levels, they are not expected to exceed 1976 levels in real terms. Continued imports of these commodities at exist- ing levels ought therefore not to be viewed with alarm. One area for import substi- tution, however, might be forestry products. The trade deficit in all forestry prod- ucts in 1980 was estimated to be more than $108 million. Most of the deficit was accounted for by pulp and paper imports, which grew to $113 million in 1980 and $157 million in 1981. The possibility of increasing pulp and paper production might be explored under the proposed National Forestry Research Plan (see chap- ter 8) and the reforestation plan of the Federaci6n Nacional de Cafeteros de Co- lombia (FEDERACAFE, the Coffee Federation), under its diversification program (see chapter 6). Table 3-2. Actual and Projected Prices ofAgricultural Imports, 1976 and 1981-90 (1983 constant dollars per metric ton) Actual' Projected Commodity 1976 1981 1982 1983 1984 1985 1990 Wheat 226 188 165 170 168 168 153 Corn 170 125 108 136 138 120 113 Grain sorghum 159 121 107 129 121 117 109 Soybean oil 664 486 442 527 737 645 563 a. See table 2-5 for an explanation. Source. World Bank data as of July 1985 48 MACROECONOMIC POLICY AND AGRICULTURE Recent Developments The tendency to liberalize imports, observed since 1967, was reversed during 1982-83. In 1982, Resolution 39/82, dated September 7, moved a number of luxury items from the free-imports list to the prior-license list. Decree 3080/82, dated October 28, on the other hand, raised the level of import tariffs on most import items by about 20 percent. The reason for these measures was to protect the domestic industry, both by barring luxury imports and by granting a higher level of protection to domestically produced goods. These measures-together with the in- crease in export incentives-were intended to compensate in part for the apprecia- tion of the real exchange rate. On April 19, 1983, by Resolution 015/8 3, it was decided to move 684 addi- tional items from the free-imports list to the prior-license list. The main difference between this measure and earlier measures-Decree 3080/82 and Resolution 39/ 82-was that now items were moved from one list to the other not to protect the domestic industry but to reduce the level of total imports. In fact, the items with higher import value were generally the ones moved from the free-imports list to the prior-license list. (There were some exceptions-fertilizer, for example.) The gov- ernment hoped that these measures would arrest the depletion of international re- serves. On May 16, 198 3, Resolution 030/8 3 moved an additional hundred prod- ucts that were considered luxury items into prior licensing. By this time roughly 56 percent of the positions in the import list, excluding prohibitions, were under prior licensing. In October 198 3, another 103 items-most of them intermediate goods in production, such as chemicals-were added to the list. By 1984, more than 90 percent of the items were on prior licensing. In mid 198 3, a further tariff increase took place, raising the rates by about 10 percent over the existing levels, followed by a 2 percent increase, a subsequent 25 percent increase in 1984, and an 8 percent surcharge on most imports. The recent changes do not violate agreements under the General Agreement on Tariffs and Trade (GArr) or the Andean Pact. Import Tariffs Table 3-3 sets out the weighted average nominal tariffs for twenty-one sections during 1979-8 3 on the basis of the stated tariff levels applicable to ninety-nine chapters. These levels would, of course, be vastly lower if they were calculated on the basis of the implicit rates applicable to all importers, including all exemptions. In particular, the present calculations exclude certain other duties for PROEXPo and general purposes and do not take into account various exemptions relevant to gov- ernment imports, to imports from the Andean Group (about 6 percent of imports), to imports from the Asociaci6n Latino Americana de Desarrollo y Integraci6n IMPORT POLICY FOR GROWTH AND STABILITY 49 Table 3-3. Weighted Average Nominal Tariff 1979, 1981, and 1983 (percent) Section Product Chapters 1979 1981 1983 1. Live animals and related products 1-5 25.25 22.75 27.25 2. Vegetables and related goods 6-14 15.93 1409 16.75 3. Greases, oils, vegetable oils 15 20 32 18.17 21.03 4. Foods, beverages, alcoholic beverages, tobacco 16-24 35 77 32.31 38 45 5. Mineral products 25-27 12.29 11.21 13.28 6. Chemical products 28-38 17.91 16.41 19.62 7. Plastics, cellulose, rubber products 39-40 35 78 27.41 32.26 8. Leathers, furs, luggage 41-43 21.75 22.63 25 49 9. Timber, cork, vegetable coal 44-46 38.95 35.41 42.89 10. Paper and related products 47-49 30.55 25.37 28.38 11. Textiles 50-63 51.55 52.24 63.31 12. Shoes, hats, artificial flowers 64-67 53.09 53.15 63.78 13. Stoves, cement, pottery, glass 68-70 38.53 29.69 35.16 14. Pearls, precious stones, coins 71-72 33.71 34 71 41.66 15. Common metals 73-83 28.49 25.24 30.05 16. Machinery and electrical materials 84-85 26.51 22.83 26.80 17. Transport materials 86-89 45.96 38.50 45 99 18 Optimal materials, medical materials, music, and television 90-92 22 69 21.98 23 60 19. Weapons 93 49.13 49.80 59.70 20. Other products 94-98 44.92 38.73 45.72 21. Art objects and antiques 99 0.00 000 0.00 Sources. Constructed from data obtained from Gonzalo Giraldo Echeverri, "Estructura de la proteccion arancelaria y para-arancelaria en Colombia despues de las reformas de 1979," Revista de planeacidn y desarrollo, vol. II (May-August 1979), pp 7-47; the DNP; Arancel de aduanas legis; and DANE, Anuario de comercio exterior, various issues. (ALADI, the Latin American Association for Development and Integration), and to imports from other groups. A further weighting of the rates for the economy, excluding all exceptions, indi- cates that the average nominal tariff in the first quarter of 1979 was 27.62.7 This rate declined to 26.11 in the second quarter of 1981 before increasing to 29.02 in the first quarter of 198 3 and an estimated 32.00 in the third quarter of 198 3. This increase in the weighted average tariffs between 1981 and 1983 represented an increase of 4.6 percent in the effect of the tariff, defined as one plus the tariff rate, which-barring changes in quantitative restrictions and transport costs-determines the internal price of importable goods, together with changes in the exchange rate. It was preliminarily estimated that the basic rate increased to about 42.50 in 1984, implying a further increase in the effect of the tariff from 1981 by about 13 per- cent. An estimate of the average nominal tariff rate at the beginning of 1985 was 50 MACROECONOMIC POLICY AND AGRICULTURE roughly 50 percent, although the effective rate might have been only about a third of that. The last generalized tariff reduction was in mid 1979.9 In 198 3 the average tariffs were higher than in 1981 for ninety-six of the ninety-nine chapters of the tariff schedule. Also, for seventy-one of the ninety-nine chapters, nominal tariffs were higher in 1983 than in the first quarter of 1979, when the last major step toward liberalizing the import sector was taken. At this level of aggregation in table 3-3, tariffs increased in twenty of the twenty-one sections between 1981 and 1982; tariffs were higher in fourteen of the twenty-one sections, however, than they were in 1979. These tables show that even though the recent increase in the level of tariffs has been fairly general, affecting most sectors, the magnitude of the tariff increases were modest, although the 1984 changes were significant. Furthermore, the increase in protection is based fundamentally on the increase in the coverage of import licenses, rather than on tariff increases. EFFECTIVE PROTECTION. A number of studies have been concerned with the de- gree of protection granted to value added, or effective rates of protection (ERPS). The ERP can be thought of as the subsidy to value added, and takes into account distortions in the output price and in the prices of traded inputs. Even though there is information on the level and structure of effective protection for 1979 through 1981, there is little information readily available for the more recent period. Since data on the historical behavior of ERPS (up to 1981) in Colombia are available in a number of publications and reports, table 3-4 contains only the overall averages, standard deviations, and maximum and minimum rates of ERPs for 1979, 1980, and 1981 " The average rate of effective protection declined steadily until 1981 as a result of successive liberalization measures. In 1981, however, its dispersion, mea- sured by the standard deviation, was higher than in 1979 and 1980. PUBLIC SECTOR IMPORTS. Import tariffs have not applied to most public sector imports-roughly a quarter of all imports including petroleum-in Colombia; there is thus an element of discrimination between the treatment of the public and private Table 3-4. Rates of Effective Protection, 1979-81 (percent) Statistic 1979 1980 1981 Average 47 55 43 79 38 66 Standard deviation 43 42 40.31 55 26 Maximum rate 397.90 399.51 606.26 Minimum rate -48.65 -45.78 -60.84 Sources: Giraldo, "Estructura de la protecci6n", and the DNP. IMPORT POLICY FOR GROWTH AND STABILITY 51 sectors. This policy has also encouraged the public sector to use a mix of inputs that is heavily based on imports. In practice, however, public sector agencies must apply to the Industrial Division of the DNP for exemption from import duty, and about 30 percent of such applications are rejected, causing some conflicts within the public sector. In principle, according to the policymakers, it would be preferable to elimi- nate the provision that exempts the public sector from the payment of import tar- iffs. This step would also help to achieve uniformity in the protective structure. Import Licenses Import licenses have traditionally been an important tool for controlling the vol- ume and composition of imports. Between 1974 and mid 1982, as a result of the liberalization process, the number of items subject to prior import licenses declined steadily. In late 1982 and early 1983, however, this tendency was reversed, as many items were moved from the free-imports list into the prior-license list. The fact that a particular good is in the prior-license list means that any person or agency that wants to import a certain amount of that good has to apply to the Instituto de Comercio Exterior (INcOMEX, the Foreign Trade Institute) for a license. Because INCOMEX does not have recent experience in processing numerous applica- tions, it has taken considerable time for a decision to be reached regarding any given application. The criteria for deciding when to grant a license, to whom, and for what amount have not yet been determined, although some guidelines exist. Given the tight foreign exchange situation, the government has revived the practice of allocating a monthly quota of foreign exchange to INcOMEX, which then decides how to allocate it among various applicants. Table 3-5 presents information on the proportion of goods within each section of the tariff schedule that was subject to prior licenses in 1979, 1980, and the first quarter of 198 3. The principal change in the second quarter of 198 3, not shown in the table, is that most of the products in sections I through 4 have been moved into prior licensing. In addition, the October change has increased the prior-license cov- erage, particularly in chapters 29, 39, and 41. These figures have been constructed as weighted averages, the importance of each chapter having been weighted by relative imports to construct the averages. In the agricultural goods sections-sec- tion 1, live animals and related products; and section 2, vegetables and related goods-the number of items subject to prior licenses has increased significantly from the number in both 1980 and 1979. This is especially true of section 1, in which 92 percent of imports have been subject to prior licenses since April 1983. The coverage has also increased for processed foods, sections 3 and 4. The change is especially dramatic when the coverage of prior licenses for section 4-food, bever- ages, alcoholic beverages, and tobacco-in 1980 is compared with 1983. For the rest of the sections-that is, those that do not include agricultural prod- ucts or food-the change in the coverage of the prior license system since 1979 is less drastic. It is clear, however, that at present the volume of goods, weighted by 52 MACROECONOMIC POLICY AND AGRICULTURE Table 3-5. Imports, by Regime, 1979-83 a (percentage under licensing, weighted average) April Section Chapter 1979 1980 1983b 1 1-5 78 54 92 2 6-14 82 79 90 3 15 71 65 89 4 16-24 51 20 90 5 25-27 72 53 51 6 28-38 74 22 20 7 39-40 45 30 48 8 41-43 68 19 38 9 44-46 99 74 60 10 47-49 53 38 59 11 50-63 45 45 81 12 64-67 68 18 69 13 68-70 28 15 54 14 71-72 74 0 24 15 73-83 51 46 63 16 84-85 50 48 45 17 86-89 80 76 84 18 90-92 22 18 31 19 93 100 100 100 20 94-98 90 62 96 21 99 100 100 100 a. For a description of goods included in each section, see INCOMEX, Arancel de aduanas legis. b. Subsequent changes in May and October 1983 raised the percentage for chapters 1-24 to 100. Sources: Giraldo, "Estructura de la protecci6n"; Resolution 015/83; and INcOMEX data. their relative importance, that are subject to prior licenses has increased significantly since 1980. Additional information regardingthe distribution of imports by regime is given in table SA-15, which presents disaggregated data at the chapter level for 1979, 1980, and 1983. Other Restrictions on Trade In addition to tariffs and licenses, there are a number of taxes on imports that also have the effect of restricting trade. The following are the most important ones. According to Resolution 2/82 of the Monetary Board, the importation of a number of goods was subject to a prior deposit of 10 percent, payable before the nationalization of a commodity. This deposit did not yield any interest, was not negotiable, and could be used to pay for a portion of the import of that particular good. In some exceptional instances these deposits could be used to pay for imports of other goods (see Resolution 16/82 of the Monetary Board). These restrictions, IMPORT POLICY FOR GROWTH AND STABILITY 53 however, were eliminated in 1983 by Resolution 99, which also eliminated the maximum payment period for imports and prior deposit for transport cost. Subse- quently a minimum payment period for some imports has been introduced. According to article 229 of Decree 444 of 1967 all imports into Colombia are subject to a 5 percent tax over their c.i.f. value. The proceeds of this tax are used to finance PROEXPO. According to article 20 of Decree 688, 1967, all imports into Colombia are subject to an additional tax (currently 2 percent over the c.i.f. value). The proceeds of this tax are used to finance general purposes in the government budget. There is also a stamp tax (3.15 percent) on the f.o.b. value. Policy Implications of Recent Developments Toward the end of 1982 there was a clear movement away from the tendencies to open up the economy that had been observed since 1967. Even though until 198 3 the increase in the level of nominal tariffs was relatively modest, the coverage of the system of import licenses increased dramatically. Recent plans are to reverse this trend gradually in order to introduce greater efficiency into the structure of production. Increasing Trade Restrictions The increases in the level of protection during 1982-84-that is, higher tariffs, more restrictive licenses, or both-have compensated in part for the present over- valuation of the peso and have brought about a higher real effective exchange rate for import-competing goods than otherwise (see chapter 1); to that extent, the de- gree of competitiveness of this group of goods has increased. When tariffs and li- censes are used, however, as in this instance, there is a serious problem in that the recovery of the importable sector takes place partly at the expense of the exportable sector. The higher real effective exchange rate for importables means a lower real effective exchange rate for exporters. This policy of using commercial policy to promote importables is equivalent to imposition of a tax on exportables." Conse- quently, while tariffs tend to generate an increase in the level of activity of import- ables, they also tend to produce a decline in the level of activity of noncoffee ex- portables, especially nontraditional exports. Furthermore, the increase in protection tends to affect the overall level of activity of the economy. It has been documented that export-promotion economies tend to outperform import-substitution economies in growth, income distribution, and employment." To the extent that the result of the new commercial policies is an implicit tax on exports, a slowdown in overall growth could be expected. The au- thorities also increased export subsidies during 1982-84. If the increases in tariffs and export subsidies are sufficiently large, this combined policy could generate an effect on relative prices similar to that of a devaluation. This does not seem to have 54 MACROECONOMIC POLICY AND AGRICULTURE happened during 1982-84, however, because the increased import restrictions have dominated the additional export incentives. This means that the export-tax effect discussed earlier was the likely result of the 1982-84 policies. Use of Import Licenses From an economic point of view, the Colombian system of import licenses is similar to a quota, or quantitative restriction, system. The main difference, how- ever, is that in the case of licenses the quantity of any particular good that can be imported in any period of time is not known a priori by the public. Also, this quantity can easily be altered by the authorities, in response to varying circum- stances. In a sense then, it may be said that the Colombian system of licenses is equivalent to a flexible quota system. This import-licensing system has two basic economic effects: it restricts the volume of a good that can be imported, and it produces an increase in the domestic peso price of licensed goods above what it would be in the absence of the license. This means that the domestic price of the imported good will exceed the foreign price plus the tariff, transport costs, and other normal adjustments. From the point of view of efficiency, and to the extent that the allocation of licenses is administratively decided by the authorities without the possibility of resale of licenses, this system can lead to corruption and welfare costs exceeding those of an equivalent import tariff. To the extent that the alloca- tion of licenses is arbitrary, moreover, there will be income distribution effects asso- ciated with the scheme, because those parties that obtained the quotas would re- ceive related rents, while consumers would incur losses associated with the higher prices they would have to pay. EFFECT ON IMPORTS. The effects of the 1982-84 increases in the prior-licenses coverage have been to reduce the volume of importable goods available and raise domestic prices. The full medium-term effect of this policy on domestic prices, however, depends on the actual restrictiveness of the licenses, the elasticity of de- mand for imports of those goods, and the process by which the licenses are allo- cated. With respect to the restrictiveness of licenses, the available historical evidence indicates that at least for agricultural goods, licenses have been fairly restrictive and have had a fairly considerable effect on prices. Domestic prices of some agricultural goods have significantly exceeded world prices (corrected for tariffs, taxes, and esti- mated transport costs). In the absence of import licenses, it is probable that the domestic price of an imported good (Pd) will be equal to the international price adjusted for the tariff and other taxes and domestic transport costs." The difference between the prices observed when import licenses are in effect and Pd can then be attributed to the effect of the import licenses. The available information on the price elasticity of the demand for imports refers to fairly comprehensive categories of goods. Empirical analyses of the subject have found price elasticities of demand for imports that range from -0.17 for capital goods to - 1.52 for consumer goods, IMPORT POLICY FOR GROWTH AND STABILITY 55 which would imply that a significant price increase can be caused by additional import restrictions.'4 TARIFFS VERSUS LICENSES. If it is intended that increased protection should be temporary, it would be desirable to compare the merits of tariffs or licenses as alternative means of achieving it. The disadvantages of licenses-from the point of view of both welfare and administration-are well known. On the basis of pure efficiency, tariffs would be preferred over licenses." In the present situation of Co- lombia, however, there is an additional consideration: which of the two systems- tariffs or licenses-is more likely to be temporary, or which of these two measures will be easier to reverse once the exchange rate problem has been solved. Licenses are determined by a high-level ministerial advisory committee for the external sec- tor, and a resolution is signed by the minister of economic development. The rec- ommendation for a tariff change is made by a national council on tariff policy to the president and the minister of finance, and a decree is signed by the president. It is not clear which of the two-tariffs or license-is likely to be less permanent on institutional grounds. A point that favors licenses as a temporary protective mea- sure is that by their own nature they would easily become nonbinding if everyone who applied for a license immediately received one. If licenses were to be applied as a temporary measure, two ways of improving their efficiency could be considered: licenses could be allocated in a way that would reflect the willingness of economic agents to pay for them, possibly by auctioning them, and the resale of licenses in the free market might be allowed. In principle, auctioning could be developed on the basis of bids submitted by interested parties, who would specify the unit price to be offered for different numbers of licenses. The numbers corresponding to each unit price would then be added together to obtain the total demand at each clearing unit price. By equating the aggregate de- mand with the number of licenses the government wished to supply, a market- clearing unit price could be established.'6 Policy Conclusions The primary objective of the 1982-84 changes was to arrest the decline in inter- national reserves, rather than to provide additional protection to the domestic in- dustry. A gradual reversal of the recent restrictive measures would be desirable-as recognized by the government in its most recent actions-from the point of view of the production sectors as export policies begin to take effect. Liberalization is envis- aged to proceed in stages in order to reduce the adjustment costs to domestic pro- ducers and the short-term employment effects. In particular, a major liberalization while the exchange rate is appreciated could be problematic, and it would need to be tied to a depreciation of the peso to reach an equilibrium level. In the meantime, it would clearly be desirable to make use of controls that are temporary in nature, easy to apply, and on the whole cost-effective. Studies in 56 MACROECONOMIC POLICY AND AGRICULTLRE other countries have demonstrated that tariffs, if applied with fair uniformity across sectors, are preferable to quantitative restrictions, since tariffs allow the volume of international trade to respond to changes in domestic demand and supply with greater flexibility. At the same time the flexibility of the control instruments ulti- mately depends on the ease of making changes, given Colombian administrative and legal procedures. Even though there are no infallible answers, for the present situation in Colombia it seems advisable to reduce import restrictions as the real exchange rate reaches equilibrium and exports respond to improvements in the real exchange rate. In the short term, however, if there is no way to induce quick enough adjustments in the real exchange rate, it may be advisable to undertake a temporary policy of tariffs and export incentives together in addition to depreciation of the exchange rate. If this policy is adopted, it is essential that the newly imposed import policies are temporary and that the public is made clearly aware of the temporary nature of these policies. It would be normally desirable to rely on a system such as tariffs that maximized the probability that increases in protection would be temporary. If it is decided, however, that licenses are best suited for this purpose, they should be allocated in a way that would capture the willingness of different parties to pay for them. In view of the problem with international reserves, one criterion for licensing at present Is the availability and terms of external financing for imports. In general, however, a good number of licenses are allocated more or less arbitrarily, their recipients gain- ing the implied rents. Auctioning the licenses would increase the efficiency of the system and transfer the rents to the government. The government could periodically-say, every three months-announce through INCOMEX the auctioning of a certain number of licenses for selected com- modities. Interested parties would then submit bids that would specify the unit prices they were willing to pay for different numbers of units of a commodity. These bids would be added up, and the clearing price-consistent with the amount to be auctioned-would be determined. This procedure would allow the central government to capture the rents derived from licenses and would avoid serious problems of income distribution and efficiency. Ideally this scheme could be supple- mented with the transaction of licenses in the open market. Additional work would be needed before such a proposal could be put into effect; it might also need to proceed in stages, being focused initially on selected commodities on a pilot basis. Notes 1. On the benefits of liberalization, see, for example, Anne 0 Krueger, Foreign Trade Regimes and Economic Development Liberalization Attempts and Consequences (Washington, D.C.: National Bureau of Economic Research. 1978); Bela Balassa, The Policy Experience of Twelve Less Developed Countries, 1973- 78, World Bank Staff Working Paper no. 449 (Washington, D.C., 198 1); and Jagdish N Bhagwat and IMPORT POLICY FOR GROWTH AND STABILITY 57 T. N. Srinivasan, "Trade and Development," in International Economic Policy: Theory and Evidence, ed. Rudiger Dornbusch and Jacob A. Frenkel (Baltimore, Md.: Johns Hopkins University Press, 1979). On Colombian policy see Luis Jorge Garay S., "La politica de importaciones, 1978-82: Una evaluaci6n" (Bogota, June 1982, processed). 2. Imports fall under free registration or prior import licensing. Items under the former can be freely imported without quantitative limitations after a procedure of administrative registration has been fol- lowed. Quantitative restrictions are imposed through the latter, which requires specific approval for each import permit. INcOMEX, the foreign trade institute, is responsible for administering these controls, and in practice, uses its discretion to exercise control even over commodities in the free registration category. 3. See Miguel Urrutia, Winners and Losers in Colombia's Economic Growth of the 1970s (New York: Oxford University Press, 1985); and Francisco Thoumi, "International Trade Strategies." 4. See Sergio Clavijo Vergara, "Los dep6sitos previos de importaci6n," Ensayos sobre politica economica, no. 1 (March 1982). 5. See World Bank, Colombia 6. More accurate estimates need to be developed. 7. The concept of weighted average tariff is tricky because the higher the tariff the less of a good imported and thus the lower its weight. In the extreme case a prohibitive tariff would have a weight of zero. 8 PM = PM* (1 + t)e, where PA is the domestic price, PM* is a world price, (1 + t) is the effect of the tariff, and e is the nominal exchange rate. 9 Gonzalo Giraldo Echeverri, "Estructura de la protecci6n arancelaria y para-arancelaria en Colom- bia despues de las reformas de 1979," Revista de planeacidny desarrollo vol.11, no 2 (May-August 1979). 10. See, for example, Thomas L. Hutcheson and Daniel M Schidlowsky, "Colombia," in Develop- ment Strategies in Semi-Industrial Economies, ed Bela Balassa and associates (Baltimore: Johns Hopkins University Press, 1982); Giraldo, "Estructura"; and Ocampo, "Politica econ6mica bajo condiciones cambiantes de sector externo 11. The symmetry between import tariffs and export taxes holds in both a world with nontradable goods and one without. 12. At this stage the evidence suggesting that in growth the extent to which liberalized export- oriented economies outperform inward-looking economies is overwhelming. See Krueger, Foreign Trade Regimes; Bhagwati and Srinivasan, "Trade and Development"; and Ian M. D. Little, Tibor Sci- towsky, and M. Scott, Industry and Trade in Developing Countries (New York Oxford University Press, 1970). 13. Pd = P*[(1 + t) + 1 + r, where P* is the world price c.i.f., t is the tariff, T is other taxes, and r is domestic transport costs. 14. See, for example, Kanta Marwah, "An Econometric Model of Colombia: A Prototype Devalua- tion View," Econometrica, vol. 37, no. 2 (April 1969), pp 228-51; Alberto R. Musalem, Dinero, infla- tion y balanza de pagos: La experiencia de Colombia (Bogota Banco de la Republica, 1971); Mohsin S. Khan, "Import and Export Demand in Developing Countries," IMF Staff Papers (Washington, D.C., 1974). 15. See, for example, Krueger, Foreign Trade Regimes, andJ N. Bhagwati, Anatomy and Consequences of Excbange Control Regimes (Cambridge, Mass.: Ballinger, 1978). Tariffs are particularly superior when licenses cannot be resold. 16. See, for example, Jagdish N. Bhagwati and T. N. Srinivasan, Lectures in International Economics (Cambridge, Mass.: Mrr Press, 1983); P. Engelbrecht-Wiggans, "Auctions and Bidding Models: A Sur- vey," Management Science, vol. 26 (February 1980), pp. 119-43, and C. A. Holt, Jr , "Competitive Bidding for Contracts under Alternative Auction Procedures," Journal of Political Economy, vol 88 (June 1980), pp. 433-45. Part Two AGRICULTURAL PRICE POLICY 4 Price Interventions, Competitiveness, and Incentives THE FOCUS OF this chapter is on sectoral incentives that are the result of macroeco- nomic policies and sectoral price interventions and on their effects on profitability, and efficiency in agriculture.' As observed in chapter 1, from the mid 1970s until 198 3 the effect of macroeconomic policies on noncoffee agriculture was on balance negative, while coffee benefited from a price boom during 1976-80. Sectoral poli- cies, on the other hand, were intended to directly boost agricultural production and income. These policies include price supports, import protection and export subsi- dies, and nonprice inducements through agricultural investments of the govern- ment, input supply, research, extension, and transfer of technology. These inter- ventions favored agriculture, and, with varying degrees of efficiency, offset some of the (unintended) disincentives from macroeconomic policies. A significant intervention is the protection of importables, effected by a combina- tion of import controls and price supports. The levels of protection in Colombia have been high, given international distortions, and tended to more than offset the appreciation of the exchange rate. Only about 10 percent of agricultural output is thus protected, however, and the degree of protection would not be considered pervasive. The levels of protection declined somewhat during the first half of the 1970s and increased thereafter, so during the 1970s as a whole no significant trend can be observed. The competitiveness of import-competing products as well as ex- port items is influenced also by changes in the exchange rate and the rate of infla- tion. More generally, incentives in agriculture as a whole are influenced by the levels and changes in the prices of agricultural products-consisting not only of export- ables and importables but also of the large group of domestic products that consti- tute 30 percent of agricultural output-in relation to the rest of the economy. These relative prices are affected by government interventions as well as by market forces. The discussion of price supports and import restrictions will therefore be complemented by analyses of measures of competitiveness and relative prices in agriculture and the extent to which government policies may be significant in these respects. 61 62 AGRICULTURAL PRICE POLICY Government Interventions in Agriculture Four types of direct government intervention can be distinguished in Colombian agriculture, the first two operating in the external sector, the other two in the do- mestic agricultural sector: agricultural trade restrictions, including import tariffs, and import-export licensing; agricultural export subsidies, and in the case of coffee, export taxes; output price supports; and price fixing in output and input markets. Direct price controls are set by the Ministry of Agriculture (MoA) on cocoa, sugar, and sisal and on fertilizers and pesticides, although the ministry-through IDEMA, the Agricultural Marketing Institute-intervenes directly in the market to buy or sell only sisal and some sugar. Producer price supports for wheat, barley, corn, soybeans, sorghum, sesame, beans, and rice are set by IDEMA and the ministry. Price support policy is closely linked to import licensing. Imports of food crops are controlled by INCOMEX, the Foreign Trade Institute. Permission to import is granted after INCOMEx is satisfied that a deficit for the crop exists at the prevailing price level. Thus, the support price can in principle be maintained without a sub- sidy, even when the international price may be lower. Other than coffee, the major agricultural exports from Colombia include sugar, flowers, bananas, cotton, rice, and tobacco, each of which is eligible to receive the subsidized PROEXPO credit. In addition, several products now receive CERT, the tax-reimbursement certificate, of 20-2 5 percent-25 percent for rice and livestock, for example, and 20 percent for cotton.2 As a member of the Sugar Exporters' Association, Colombia faces a quota on its exports of sugar. The domestic price of sugar is fixed by the MoA after negotiations with the Productores de Caiia (PROCAR A, the Sugar Growers' Asso- ciation) and the Asociaci6n de Cultivadores de Caiia de Az6xcar de Colombia (ASOCA&A, the Sugar Mills Association). This price takes into account the cost of production and the level of international prices. Cotton and rice now receive a variable export incentive over and above CERT, depending on the level of interna- tional prices. The main instruments of protection for agriculture are tariffs and quantitative restrictions on imports and incentives for exports. In Colombia, exports of agricul- tural products have usually been subjected to a quota, with the exception of flowers, bananas, and tobacco. First priority has been given to supplying the domes- tic market; when this requirement has been satisfied and a surplus remains, permis- sion to export has been granted. For coffee, export quotas have been set in recent years by the International Coffee Organization (see chapter 6). Domestic producers and consumers of cotton have been obliged to reach an agreement on both quantity and price before the government decides on export permits; as would be expected, this provides more leverage to domestic consumers-that is, to the textile indus- try-than to producers. Rice export permits have been granted only if a production surplus is expected. Most import licenses are purchased by IDEMA, and INCOMEX PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 63 normally does not grant an import license without prior consultation with IDEMA and the MoA. As a result, the protection granted to the production of a particular product does not constitute the legal tariff rate or export incentive alone, but may be significantly greater. The combined effect of tariffs and quantitative restrictions is nevertheless reflected by nominal rates of protection measured by a comparison of domestic and international prices. IDEMA's Policy The main functions of IDEMA are to buy, sell, and distribute certain agricultural products; to maintain adequate operating stocks; to import and export certain other products; and to intervene in the marketing of strategic agricultural commodities. One of IDEMAS main tools has been the provision of support prices for wheat, barley, corn, sorghum, sesame, beans, and rice. The purpose of the price supports- announced twice a year at the beginning of the planting seasons-has been to guar- antee that domestic producers can recover production costs or realize a minimum level of income per unit of output.3 Price-support and import policies have been closely related, and IDEMA attempts to supply short-term deficits by imports and to build up stocks to tide over periods of short supply. In addition, an effort has been made through import policy at least to equate prices of the imported products with the support prices plus distribution costs.' Most of IDEMA'S imports and sales involve raw materials and intermediate inputs in production. Various reasons are given for this emphasis: domestic production of raw materials falls short of domestic consumption in the case of wheat, sorghum, corn, soybean cake, and milk, all of which IDEMA imports; the market structure for raw materials is imperfect; and the share of imported raw materials in the produc- tion costs of food for direct human consumption is high. More significant perhaps, IDEMA'S interventions in general help the government to appropriate economic rents from import controls, which otherwise would go to private industry, and also help to protect the domestic producers of agricultural raw materials. Level of Price Support In principle, price supports take into account estimated farm production costs and international prices. In practice, imports are kept under prior licensing, with global import quotas, by product and by semester, and buyers of the products im- ported by IDEMA continue to pay IDEMA the full difference between the domestic support price and the cost of importing. Support prices have increased at a rate approximately equal to the rate of inflation from a base determined in the early 1970s, and the real value of support prices has not changed on the average (see tables 4-1 and SA-13). The real peso value of price supports-with the exception of rice-declined after 1975, however, following steady increases in the value of the index during 1970-75. In the second half of the 1970s, the general intervention Table 4-1. Price Comparisons and IDEMA Purchases, Semesters I and 2, 1980-84 (pesos per ton) 1980 1981 1982 1983 1984 Commodity - and price 1 2 1 2 1 2 1 2 1 2 Wheat Support price 12,000 14,000 15,500 17,500 19,000 20,600 22,500 24,750 26,730 29,500 Market price n.a. 13,500 14,986 15,960 18,381 19,037 20,534 22,340 na na. IDEMApurchase price 11,925 14,744 15,887 17,825 19,129 21,401 23,114 25,425 n.a. n a. IDEMA purchases 85.6 29 4 49 5 41.7 91 7 50 1 80.7 n.a. n.a. n a Corn" Support price 9.600 11,680 14,040 15,986 18,250 20,246 22,965 23,980 26,380 27,700 Market price ia. 13,825 13,465 14,200 16,354 19,172 21,144 23,810 n.a ia IDEMA purchase price 8,950 11,480 13,525 14,958 17,642 18,863 21,997 22,709 n a. na IDEMA purchases' 03 2 8 2.1 3 2 3.5 6.6 3 5 n.a. in a n a. Soya Support price 14,930 16,000 21,200 24,000 27,600 30,300 33,000 35,475 39,400 43,340 Market price n.a. 19,150 21,300 24,000 27,634 31,194 34,000 37,213 n.a. n.a. IDIA purchase price n.a. n.a. 20,325 23,625 27,168 30,245 30,449 35,504 n.a. n.a. IDEMA purchases n.a n.a. 0 1 6.7 2 0 0.2 0.1 n.a n a. n.a. Sorgbum Support price 8,645 9,800 11,500 13,200 15,000 16,700 17,900 19,240 20,780 22,440 M arket pricc is a 11,436 12,228 11,269 14,181 16,971 17,585 19,320 i TI I IDEMA purchase price 8,289 9,568 11,182 12,624 14,730 16,189 17,463 18,611 ni a ina IDEMA purchases 3.3 2 ( 7.5 1 2 19 7 8 6 6.7 n.a. n.a. n.a. Sesame Support price 29,820 29,820 29,820 31,000 32,000 35,000 38,200 42,020 46,220 58,240 Market price n a. 29,800 24,500 28,000 30,500 40,000 40,000 44,600 n a n.a. IDLMA purchase price 29,554 29,651 29,539 30,611 31,568 34,275 37,310 41,406 na n.a. IDEMA purchascs' 93.9 18 1 89.5 10.3 33 4 4.4 3.8 n.a. na n.a. Beans Support price 30,000 37,500 40,500 45,000 46,000 52,000 58,300 64,130 69,900 85,000 Market price n a 40,000 44,000 52,500 60,800 63,500 67,700 69,540 n.a. n a IDEMA purchase price n.a n.a. 30,666 36,173 n a. n.a. 41,580 63,442 n.a. ta IDEMA purchases' n.a. n.a. 2 1 0 2 n.a. n.a 0 3 n a n.a. n a Rice' Support price 9,957 11,777 12,720 14,500 16,200 18,100 19,900 21,700 24,180 25,390 Market price n a n.a. 10,860 13,250 14,790 14,550 17,137 17,605 n a. n.a. IDFMA purchase price 9,247 10,418 10,647 12,097 13,460 14,882 16,344 17,649 n.a n.a. IDEMA purchases' 5.4 10 9 0 2 1 9 1.4 17 1 10 1 na n.a. n.a Barley Support price 9,800 10,500 13,000 15,000 16,800 18,600 20,100 22,500 24,750 29,500 Market price n.a 12,500 14,150 14,000 15,438 17,450 18,500 21,500 na n a IDLMA purchase price n.s n.a n.a. n.a n.a 35,750 n.a. 19,840 it a na IDEMA purchases' n.a n.a. n.a. n a. n a 2.1 n.a. n.a 1a n.a n.a. Not available. Note: The support price is set for a certain grade of the crop, usually a superior grade, so that the average price paid by im.DmA is typically lower The market price is the national average farmgatc piice. a. As a percentage of total production. b. Weighted average price of white and yellow corn. c. Average price of CICA 7, 8, and 9 and Metica and Oryzica rice Source: iDLMA 66 AGRICULTURAL PRICE POLICY also declined except for sesame and wheat. This reduced level of intervention was in part the result of the precarious financial situation of IDEMA as a result of the huge losses that it incurred during the period 1970-74 and continued to accumulate in the years following. The 1983 increases in price supports were nearly uniform and corresponded to inflation levels. The increases announced for semester 1 of 1983, for example, ranged from 7 percent to 9 percent for all crops. Similarly, while IDEMA claims to be concerned with stock levels, the 198 3 semester 2 support price of rice was increased 9 percent, even though IDEMA had large stocks of rice from the preceding year that it could not profitably export and, according to IDEMA'S own estimates, costs of production increased only 5 percent during the semester. Thus it appears that changes in IDEMA S Support prices have been based primarily on the rate of inflation. This implies an attempt to ensure that farmers' returns on a crop do not drop below previous levels in real terms rather than an attempt to alter the production mix. Effect of Policy IDEMA'S Support price is offered for a certain grade that is usually higher than average, and the support price is therefore typically higher that the average price actually paid by IDEMA (see table 4-1). The support price of a commodity is uniform across the country. Both the announced support price and the average price actu- ally paid by IDEMA are usually higher than the market price would have been in areas located far from consumption centers but lower than the average open market price in the country and close to estimated producer prices (see table SA-14). Con- sequently, IDEMA manages to purchase only a small fraction of the output of most crops. In general the total value of IDEMA S purchases does not exceed 1-3 percent of the total value of agricultural output in noncoffee agriculture. The principal ex- ceptions are wheat and sesame, of which IDEMA has purchased 49 percent and 43 percent respectively of the total annual production during the last three years. Therefore, while IDEMA has had a strong influence in stimulating production and increasing farm incomes for some products in some regions, its price support func- tion is not an important policy tool for stimulating production throughout the country. Protection and Efficiency in Agriculture A comparison of domestic prices and international prices provides a measure of the extent to which domestic production is protected from external competition. In table SA-15 the f.o.b. prices of major traded commodities are compared with Co- lombian prices at the official exchange rate during 1970-82. Admittedly, these comparisons are rough. In many instances international prices are the result of sub- sidies provided by exporting countries. No quality adjustments are made in com- PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 67 paring external and internal prices. The use of the official exchange rate does not account for the overvaluation of the peso that has occurred since the mid 1970s. As a result, any implicit tax on exports-that is, a domestic price lower than the exter- nal price-would be underestimated, and protection of imports-that is, an exter- nal price lower than the domestic price-would be overestimated. Finally, a more nearly correct price comparison for import-competing commodities would be be- tween the farmgate price and the Colombia import price c.i.f. plus port and trans- port charges to the consumption center, less transport cost from the farm to the consumption center. The Colombia f.o.b. price of exports would be compared to the farmgate price plus all transport and port charges. The present comparisons can be used only to suggest broad differences in the treatment of imports and exports, and general secular trends. Exports versus Import-Competing Products Export agriculture in general has been taxed implicitly, and import-competing agriculture has been protected. For most export commodities-rice, coffee, ba- nanas, tobacco, cotton, and sugar-the ratio of domestic to international prices (f.o.b.) has been significantly less than one, no matter whether the export crop is food, such as rice, or is not a food item-cotton, for example.' Even after adding international transport costs to the Colombian prices of most of these commodities, price ratios would appear to be less than one. In the case of import-competing crops-barley, corn, wheat, sorghum, soybeans, and butter-the ratio of the do- mestic to the international price (f.o.b.) has been significantly higher than one, sug- gesting that one government objective is to achieve self-sufficiency in food, al- though this aim may not be expected to be realized for each specific crop. Self-sufficiency in food, however, has been pursued at a high cost with respect to many import-competing products, as the nominal protection granted these prod- ucts-corn, wheat, and sorghum, for example-has at times reached levels between 50 percent and 100 percent. The implied cost of protection, moreover, has in- creased in recent years as levels of protection have gone up significantly. There appears to be some measure of protection of animal products, such as beef and milk, and particularly butter. The rates of nominal protection of beef and milk, however, must be interpreted with particular care, since the domestic price of beef seems to be the average of different qualities of meat, while the international price is for meat of only the best quality. The international price used in the price compari- son for milk is the U.S. producer price, as supplied by the Central Bank, for lack of a better indicator, while the quality of U.S. milk and Colombian milk is quite different. The pattern of protection during 1970-83 as a whole was rather stable, even after accounting for the growing overvaluation of the peso since the mid 1970s. Export crops remained implicitly taxed, while import-competing food crops were being protected at the official exchange rate. One export crop that exhibits wide 68 AGRICULTURAL PRICE POLICY variability in its index of nominal protection is sugar; at times it appears heavily taxed, while at others it seems to be heavily protected. The reason for such volatil- ity rests on the goal of keeping prices in the internal market relatively stable in the face of highly volatile international prices. Protection and Efficiency The levels of nominal protection have been related positively to the inefficiency in production of a small group of food crops, such as wheat, corn, sorghum, and soybeans. This conclusion is based on comparisons of average production costs with their international peso prices, although unadjusted for production subsidies provided abroad (see table SA-16). Production of corn and wheat appear to be inefficient by international standards, and the levels of nominal protection received barely serve to cover the high costs of production. The production of soybeans and sorghum is less inefficient when compared with international standards, and the corresponding levels of nominal protection of these products are smaller than for wheat and corn. Cotton and rice seem to be competitive but historically they have been implicitly taxed and subject to export controls; recently increased CERTs-tax credit certificates-and a variable subsidy are efforts that have been made to offset the disincentives from an appreciated exchange rate and declining domestic compet- itiveness. Historically, it appears that crops which can be developed as exports and which can compete successfully in international markets have not been stimulated but have been implicitly taxed. Only when export crops develop problems in external markets have support measures been devised and implemented, as they have for cotton. On the other hand, import-competing crops that do not stand up to exter- nal competition have received protection, but not much increase in their output has been achieved. Most of the protection granted to import-competing crops has gone to food crops in general, particularly cereals.7 A More Precise Estimate of Protection Strictly speaking, the farmgate price of importables should be compared to the c.i.f. import price plus port and transport charges to the consumption point less transport charges from the farm to the consumption point. Data on port and trans- port charges, however, are not readily available. If it were assumed that the sum of port charges and transport costs from the port to the consumption point is equal to the transport costs from the farm to the consumption point, a direct comparison of the c.i.f. price with the farmgate price would be possible. Since port charges in Colombia are high, this simplifying assumption would lead to an overestimation of the nominal rate of protection. In a sense, the resultant estimate gives a measure of protection both from policies of import restrictions, including tariffs, and from nat- ural protection in the form of high port charges. The nominal rates of protection thus calculated for five principal import crops are given in table 4-2. The high nomi- PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 69 Table 4-2. Nominal Rates of Protection for Selected Importable Crops, 1980-82 (percent) Crop 1980 1981 1982 Wheat 36 45 91 Corn 87 67 79 Soybeans 37 46 85 Sorghum 67 57 110 Note: These rates are measures of the percentage differences of farmgate prices over international prices c i.f. Colombian ports. Sources IDEMA, DNP, and estimates made by Mateen Thobani nal rates of protection imply that the bulk of the importable crops receive a produc- tion subsidy in the sense that they could have been imported more cheaply. The rates of protection impose an implicit consumption tax, since retail prices are based on markups from the protected farmgate prices. The distortion suggests the possi- bility of gains from lowering import restrictions. Calculating nominal rates of protection for Colombia's exportable crops is diffi- cult for two reasons: first, the comparison must now be made between the Colom- bian f.ob. price and the farmgate price plus the sum of transport costs to the port and port charges-the earlier simplifying assumption on transport and port charges can no longer be made; second, Colombia's exportables, such as sugar, cotton, rice, flowers, and tobacco, unlike the importables, cannot be exported before significant processing costs have been incurred. One way to circumvent these problems is to compare the f.o.b. price to the domestic wholesale price, on the assumption that the wholesale price approximates the farmgate price plus processing and transport costs. The problem with port charges remains, however, as do the differences in quality between domestically consumed exportables and actual exports. For these reasons a more precise estimate of protection rates for exportables has not been attempted. Nominal rates of protection for exportables are expected to be signifi- cantly lower than for importables, however, and may be negative, at least for some crops, in several years. Turning to the effective rate of protection (ERP), since agricultural inputs in Co- lombia receive lower rates of protection than do importable outputs, the ERP of the latter would exceed the nominal rates given earlier.' Therefore, importables unam- biguously receive high rates of protection. The ERP of exportables, however, is more difficult to determine for the same reasons mentioned in respect to nominal rates. Evaluating Protection: The Case of Wheat The stated purpose of wheat protection is not so much to stimulate domestic production-in which IDEMA has had limited success, since about 90 percent of Colombian consumption of wheat is still imported-as to increase farm incomes in 70 AGRICULTURAL PRICE POLICY certain regions. The concerns are net losses of efficiency and the effect of this policy on farm incomes, consumer welfare, and government revenues. In evaluating a reduction of the high rate of protection given to wheat, consideration should also be given to the substitutes or complements for wheat in either consumption or production. Removing wheat alone would lead to the movement of resources into the production of a protected substitute: a second best policy might be to reduce the ERP of wheat to that of its substitutes. The welfare gain from eliminating wheat protection alone is likely to be small, while the principal substitutes for wheat in production and consumption are also protected at fairly high rates. The more inter- esting case might be to calculate the potential welfare gain if IDEMA were to lower its support prices for all imported cereals. The welfare effect on consumers, producers, and the government that would be associated with lowering wheat protection by 10 percent, 15 percent, and 20 per- cent has nevertheless been measured.' Based on an analysis of producer and con- sumer surplus, the results show that it is the consumer who would stand to gain most from the policy, and the government and IDEMA would lose the most. The net gain in efficiency from such a wheat policy above varies from Col$163 million to Col$226 million. This net gain in efficiency is modest compared to the value of consumption, which would be less than 1 percent. Even when the elasticity of supply was increased from a short-run value of 0.6 used in the analysis to a long-run elasticity of 1.2, the gain in efficiency as a percentage of the value of wheat con- sumption increases to only slightly more than 1 percent in the 10 percent distortion case. Interventions by IDEMA The elements and the pattern of protection for import-competing cereals have varied since 1970. While the ratio of support to international prices declined in 1970-75, it increased in 1976-8 3. The reason is that during 1970-75, the increase in costs that had to be covered by higher support prices was smaller than the rise in the peso value of the international price; thus was the double objective of making agriculture more efficient and of increasing the real incentives to domestic agricul- tural production accomplished; the opposite behavior of costs and the peso value of international prices occurred during 1976-8 3, and it is for this reason that a higher level of protection was required. As for commodities in which the country seems to enjoy a comparative advantage and for which IDEMA'S support prices are lower than international prices, the reasons for IDEMA to intervene are to build up working stocks and to support producers when bumper crops are harvested at the regional level. In certain regions there may be an effect on production, but it is not significant nationally. Only in the case of paddy rice and wheat were support prices higher than average nationwide producer prices during 1970-82; for all other products- PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 71 sesame, barley, beans, corn, sorghum, and soybeans-average nationwide producer prices have in general been higher than support prices. Thus only in the case of wheat and sesame have price supports possibly been an effective incentive to boost output, both because support prices are higher than producer prices or interna- tional prices and because the participation of IDEMA in the market for these products is rather large. It is doubtful whether intervention for the rest of the products has any significant effect either in increasing production or in guaranteeing a minimum return to the producer: the intervention of the institute in these cases is to be looked on rather as a help to producers at times when problems of excess produc- tion arise at the regional level or when stocks need to be built up to complement the function of the institute as a distribution entity and to diminish risks. Support prices higher than international prices of most importables can be main- tained because imports are restricted. INCOMEX must approve the import of every crop whose price is supported by IDEMA. It grants a license only after consulting with IDEMA and the MoA and after confirming that a deficit for the crop exists. This is to ensure that consumer prices do not fall below IDEMA'S support prices. IDEMA may import the commodity without paying any tariffs, whereas other importers must pay between 15.5 percent and 24.5 percent. While IDEMA'S domestic pur- chase price may be higher than its import purchase price, the price at which IDEMA subsequently sells is based on its domestic purchase price." Thus IDEMA obtains all the rents from the system of tariffs and licenses. The farmgate price of wheat in 1982, for example, was 91 percent higher than its price c.i.f. Even after accounting for differences in the costs of transport from the port to BogotA, instead of from Pasto, the principal wheat-growing region, to Bogot6, the farmgate price remains about 70 percent higher than its import parity price. Since tariffs on wheat in 1982 were only 16.5 percent until October, when they were raised to 18.5 percent, and IDEMA does not pay tariffs anyway, the difference of 70 percent constitutes rent to IDEMA and port charges. Port charges have been estimated to be very high in Co- lombia and are likely to contain elements of rent. Production Incentives for Major Commodities Despite the protection for importables and increases in the levels of protection during the second half of the 1970s, production incentives of Colombia's traded agricultural commodities-that is, both exports and import-competing commodi- ties-have on the average declined since 1975. Evidence is provided by an index of the real peso value of international prices for selected agricultural commodities (see table SA-1 8). This index is the product of the international price in dollars and the average exchange rate for each year, divided by the implicit gross domestic product (GDP) price deflator for that year, thus combining the conditions prevailing in inter- national markets with domestic economic policies and conditions. 72 AGRICULTURAL PRICE POLICY Protection and Competitiveness To the extent that the GDP deflator reflects trends in domestic production costs, this index indicates broad patterns in international competitiveness of Colombian agriculture. For the majority of the fifteen products chosen for the exercise, the index fell 25 percent to 50 percent between 1975 and 1983, and for two of these products, butter and beef, no clear trend exists. This behavior differs remarkably from that experienced during the years 1970-75, when the index rose for most of the products, showing rather substantial margins in some instances. Among the products chosen, the country produces and imports butter, barley, corn, wheat, palm oil, and soybeans. The index for butter varied moderately during 1975-83 and no clear trend can be deduced. The 1982 value of the index for barley was 35 percent lower that in 1975, but the index had reached a bottom level in 1978 and recovered thereafter. The index for wheat, corn, soybeans, and palm oil has been falling sharply. Among export products the index for beef, rice, sugar, coffee, bananas, tobacco, and cotton has decreased significantly since 1975. In the case of cotton, the loss of international competitiveness was so great that the area planted and the output fell to a third their size after 1975. External and Internal Factors International economic conditions, particularly low international prices, have undoubtedly had an effect on the trend and the level of the index for several prod- ucts. It is worth noticing, however, that the international price of products such as coffee, bananas, beef, and tobacco was higher in real terms during 1981-83 than it was in 1975 (see tables 1-3 and 4-3). If domestic conditions had remained un- changed-that is, if prices and the exchange rate had remained constant-the index of the real peso value of international prices of agricultural exports of these products should have been higher in 1982 than in 1975. High rates of inflation and a deteri- orating-that is, appreciating-real exchange rate have been important reasons for the loss of competitiveness of agricultural exports. In addition, this decline is sub- stantial when the index of international prices-which is the index of the real peso value of international prices with stable prices and stable exchange rates-is com- pared to the actual index of the real peso value of international prices. The same conclusions are obtained when the index of competitiveness is gener- ated on the basis of the ratio of the peso value of the international price to the average production costs for some selected agricultural commodities-rice, corn, sorghum, soybean, wheat, and cotton (see tables 4-4 and SA-19). The pattern ob- served is similar to that of the previous index, although the levels vary somewhat. Using the index that incorporates production costs, it will be seen that rice, corn, sorghum, and soybeans have suffered a greater loss in international competitiveness PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 73 Table 4-3. International Prices of Selected Agricultural Exports, 1970-83 (constant prices; 1975 = 100) Year Cotton' Sugar" Bananas' Riced Beef Tobacco' 1970 110.1 36 8 137.2 91 5 172.3 157.0 1971 119 2 414 106.9 85 4 192.7 132.1 1972 116 3 60.9 112.2 88 1 226.9 131.1 1973 166.7 66.5 95.9 134.9 260.8 114.7 1974 139.5 166.4 85.5 150.7 245.5 104.5 1975 1000 100.0 100.0 100.0 100.0 100.0 1976 143 5 56.0 103.7 72.4 104.3 100.1 1977 121.7 36.2 101.1 72.1 123.3 100.5 1978 104.3 29.7 90.0 73.3 104.0 91.9 1979 100.2 32.8 91.6 62.8 154 1 89.5 1980 112 5 89 7 97.5 75.4 159.5 87.5 1981 105.7 556 106.0 90.2 144.6 103.3 1982 93.2 28.0 104.4 59.7 112 4 120 1 1983 104.8 27.3 115.9 59.7 n.a. 118.2 n.a. Not available. Note: Constant prices are nominal prices divided by the index of manufacturing unit value (muv), taken from World Bank data of January 1985. For coffee see table 1-3 a. Liverpool index. b Caribbean (New York). c. Latin America (U.S. ports). d. United States (New Orleans). e. Argentina (frozen). f. United States (all markets). Sources: Ts and estimates compiled by J Garcia-Garcia. than have cotton and wheat. The general trend observed in all of these commodi- ties, however, is the same: following a period of substantial gains, 1970-75, there has been a loss of international competitiveness in agricultural production in Co- lombia since 1975. Trends in Relative Prices of Agricultural Products The internal relative prices of agricultural production have varied significantly since 1970.11 As shown in table 4-5, the domestic prices of agricultural output rose in relation to those in the rest of the economy during the early 1970s, fell during the mid 1970s, recovered again briefly between 1976 and 1977, and declined con- tinuously thereafter. The direction of change in these price movements for agricul- ture as a whole has been connected with price changes in commodities internation- ally traded by Colombia. Comparing the terms of trade-that is, the ratio of 74 AGRICULTURAL PRICE POLICY Table 4-4. International Prices of Selected Agricultural Imports, 1970-83 (constant prices; 1975 = 100) Year Wbeat' Corn" Barley' Sorgburn Soybeans" Butter' Palm oil 1970 68.8 98.6 87 5 100.2 107.5 93.7 122.4 1971 77 1 91 0 75 0 88 6 108.9 122 0 113 6 1972 79 6 79.8 119.2 86.1 108.2 127.9 86 3 1973 133 8 116 5 126.4 126 1 188.4 87.5 125 3 1974 137.7 126 0 95 5 128.4 143.1 83 6 177 2 1975 100.0 1000 100.0 100.0 100.0 1000 100.0 1976 87 5 92.4 85.1 91.9 103.1 104.2 93 1 1977 62.7 72.4 79.0 71.6 1150 105 9 113.8 1978 65 8 65.0 56.7 66.2 93.7 116.8 107.6 1979 74.0 669 799 72 5 930 124.9 105.0 1980 73.6 67 0 74.5 84 2 85.9 133.7 86 5 1981 78.3 73.1 102.3 86 0 87.5 130.5 88 8 1982 77.6 62 7 97 4 75 0 760 123 5 70 8 1983 80.9 76 3 60 9 n.a 84.7 102 2 67 4 n a Not available. Note. Constant prices are nominal prices divided by the index of manufacturing unit value (\, taken from World Bank data of January 1985. a. United States (U S. Gulf ports) b. Yellow no 2 (U.S Gulf ports). c C.i.f Colombia, from DANE. d United States (Rotterdam). e. New Zealand (London) f Malaysia (Europe). Sources Same as for table 4-3 implicit price deflators of peso exports to peso imports, both in agriculture and in the economy as a whole (see table 4-6)-with producer prices of agricultural com- modities in relation to prices in the rest of the economy, a fairly close correlation in the directions of change can be observed. The movements in Colombia's terms of trade have been dominated by coffee prices, and changes in producer prices in agriculture in relation to prices in the rest of the economy have been stronger for coffee than for noncoffee agriculture. As a result of the sharp fluctuations in the international price of coffee, and its domi- nance in the terms of trade and the domestic prices of agricultural output in relation to those of the rest of the economy, however, a clear distinction must be made between coffee and noncoffee agriculture. The domestic price of coffee in relation to prices in the rest of the economy has followed the same pattern as that of the terms of trade for coffee: both reached a peak in 1977 and declined continuously thereafter. Variations in domestic relative prices of coffee have been less pro- nounced than variations in the terms of trade, which will be further elaborated upon in chapter 6. PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 75 Table 4-5. Producer Prices in Agriculture in Relation to Prices in the Rest of the Economy, 1970-83 (index of price ratios: 1975 100) Total Noncoffee Year agriculture Coffeeb agriculture' 1970 98.1 112 4 92.8 1971 944 95.8 93 9 1972 99 4 105.4 97 4 1973 109 1 120.3 105 1 1974 105 8 103.0 106.8 1975 1000 100(0 1000 1976 111.1 150.8 98.0 1977 126 1 201 3 103.4 1978 109.4 149.0 93.9 1979 96.1 112.3 88 9 1980 92.3 106.8 85 8 1981 82.8 82.6 83 3 1982 82 7 82.0 82 9 1983 81.7 79.8 82 5 Note. Gross value of output deflators were used, the comparison in each column is with the entire nonagriculture sector of the economy. a Broad definition; see text. b) Consisting of pergamino and processed coffee. c. Sectors 02 and 03 plus sector 12 Source. Garcia-Garcia, "Aspects of Agricultural Development," which is based on oNE data. The terms of trade for noncoffee exports as a whole have varied relatively little since the mid 1 970s, and their levels in 1982-8 3 were hardly different from those in 1974-75. Within the agricultural sector also, the terms of trade for noncoffee exports have fluctuated less than those for coffee, but since the mid 1970s-with some exceptions-a downward trend can be observed. International price varia- tions have differed significantly from product to product, as will be discussed in the following section; what is indicated in table 4-6 is only an aggregate picture. The declining terms of trade for Colombia's noncoffee agricultural exports can be dis- cerned from international price trends. With efforts to hold the country's competi- tiveness, export performance could have been better, despite this decline in interna- tional prices. Within the domestic economy the decline in the price of noncoffee agricultural products in relation to other prices has been stronger than the decline in the terms of trade. Two subperiods emerge with clear trends: 1970-74 and 1975-83. During the first there was an improvement in the internal relative price of noncoffee agricul- tural products, but no clear trend appeared for coffee. On the other hand, there was no defined pattern in the external terms of trade either for coffee or for noncoffee agricultural products during 1970-74. The improvement in domestic terms of 76 AGRICULTURAL PRICE POLICY Table 4-6. Prices of Exports in Relation to Prices of Imports, 1970-83 (1975 = 100) Noncoffee Coffee Noncoffee Agricultural agricultural Year Exports exports exports exportsa exports 1970 107.3 135.0 82.6 176.8 132.5 1971 99.0 115.9 84.6 137.1 93 5 1972 109.4 132.9 93.2 155.3 110.8 1973 122.1 150 1 103 6 128.7 117.0 1974 113.3 111.4 114 8 90.9 122.5 1975 1000 1000 100.0 100.0 100.0 1976 131.3 179.7 98.6 171.3 91 7 1977 164.8 286.7 101.7 234.2 101 4 1978 144.5 197 1 102 5 271 4 87.7 1979 123.5 141 3 105.8 154.3 87.5 1980 127 7 139 6 116 7 142.8 101.2 1981 109 7 103 5 115 1 119.6 97.3 1982 112 9 116.9 110.4 107.8 83.9 1983 113 9 114.4 113.6 108 2 90.8 Note The comparison is with prices of all imports except in the case of agricultural exports, price of which are compared with those of broad agricultural imports. a. Broad definition, including sectors 01, 02. 03, 08, and 12 in the DANE classification. Source Same as for table 4-5. trade for noncoffee agricultural products can be explained by a move to liberalize trade and promote exports that took place between 1970 and 1974 in the form of lower nominal tariffs and lesser quantitative restrictions. These permitted an in- crease in the real exchange rate which was then transmitted to the noncoffee agri- cultural sector." During 1975-83, the trend in domestic relative prices has been one of deterioration for noncoffee agriculture, with the exception of 1977, when a serious shortfall of domestic production pushed prices of agricultural products up- ward. Besides movements in the external terms of trade, domestic factors also sig- nificantly explain the decline in the internal relative price of noncoffee agricultural products since 1975. Some of these considerations were analyzed in chapter 1- improvements in the terms of trade for coffee, a lag in the exchange-rate adjust- ment, a growing fiscal deficit, and inadequate use of import policy to offset infla- tionary pressures. Policy Conclusions An important sectoral intervention-in addition to credit subsidy (see chapter 8)-is represented by import restrictions complemented by IDEMA's price supports, provided mostly to cereals and other food crops, which, with the exception of rice, PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 77 do not generally appear to enjoy a comparative advantage. The import-competing component is a relatively small part, so despite substantial levels of protection of this segment, the competitiveness and production incentives have been diminishing since the mid I 970s. The policy inference is not that the levels of import protection should be raised further: levels of protection are already high, and protection im- poses a cost on consumers, while it has, on the whole, been ineffective in stimulat- ing production. Exports of agricultural products, on the other hand, have in the past been subject to a quota depending on estimated surpluses, with the possible exception of flowers, bananas, and tobacco. In 1984, a shift in policy in favor of removal of export restrictions has been initiated. It would be advisable to maintain such a policy direction for an extended period in order to guarantee producers that they would not face impediments to exporting. If the objective of IDEMA is to protect domestic producers of importables, import restrictions might achieve that goal without price supports, unless the market is completely monopolistic. Even if some price supports are to be provided, a reduc- tion of the support prices in areas farther removed from consumption centers to take into account increases in transport costs would be desirable for the sake of efficiency. On the other hand, income distribution goals might be particularly im- portant in such removed areas, and alternative redistributive mechanisms might need to be devised. Reforms of IDEMA S seasonal price stabilization and a separation of price support operations from storage functions would be helpful. While the benefits from import controls and price supports have been small, their economic costs are more complex to estimate. In a partial and static analysis these costs do not appear large compared to the value of sectoral GDP. The losses in alloca- tive efficiency caused by policy are not very large in the case of wheat. Neverthe- less, these interventions bear a significant financial cost, particularly in comparison with the public sector operations in agriculture. At least from this point of view, improvements in efficiency in pricing policy would merit attention. Furthermore, there may be benefits from providing greater incentives to reallocation of resources and to dynamic efficiency. Also, gains from liberalization of trade, if measured in the framework of general equilibrium, are likely to be more significant. International conditions have undoubtedly contributed to the problems of Co- lombian agriculture: the terms of trade for noncoffee agricultural exports have de- clined since the mid 1 970s, as international prices for these products in real terms have declined. The country could have maintained greater competitiveness abroad, however, with more adequate macroeconomic policies. Despite low and declining protection, as measured by domestic and external price ratios for traded commodi- ties, the international competitiveness of Colombia's noncoffee agricultural prod- ucts was stronger during 1970-75 than in the post-1975 period. Declining compet- itiveness since 1975, despite rising protection, has been at least in part the result of high rates of domestic inflation and an appreciating exchange rate. In general terms, the emphasis of policy for the 1980s could be on establishing a more neutral macro- economic framework, which would not prejudice incentives to agriculture. In addi- 78 AGRICULTURAL PRICE POLICY tion, export promotion could replace import protection as a development strategy. A favorable export climate within Colombia would be essential in order to regain external markets and to restore the confidence of exporters in the government's intentions concerning export promotion. The macroeconomic adjustments under way (see chapter 9) are in the right direction from these points of view. The range of price distortions in Colombia during the 1970s, including protec- tion to agriculture and manufacturing and the pricing of capital and labor, has been judged moderate in comparison to groups of high-, moderate-, and low-distortion countries." Coefficients of protection during the 1970s have been moderate for agricultural products as a whole, high for corn, moderate for cocoa, and low for rice and beef. The fact that agricultural prices have not been unduly distorted dur- ing the last decade has also been related to the sector's good long-term record.4 Consistent with these observations, the needed shifts in agricultural pricing policies for growth and employment supported in this book are relatively modest. Notes I. Earlier discussions include Roberto Junguito Bonett, "Agricultural Incentives in Colombia" (Bo- got, 1982, processed), Eduardo Sarmiento Palacio, Inflacion, produccidny comercio internacional(Bogota: FEDESARROLLo, 1982). 2. Although many of the differences have been reduced, the degree to which exports of various agricultural products benefit from these incentives varies Flower exports to the United States, for in- stance, receive only a 1 percent CERT on account of objections by the United States, while those to other countries receive 20 percent. 3. L F Londofio Capurro, PolHticaagropecuario, 1981-1982(BogotA, 1982), p 26, IDEMA, "Filosofia de los precios de sustentacion (Bogota, November 1982, processed), p. 4; and Eduardo Sarmiento Pala- cio, "Objetivos del IDEMA (Bogoti: n.p., n d.), pp. 8 and 84. 4 A. Ramirez Ramirez, IDEMA. Dos anos de labores, 1980-1981 (Bogot, 1982, processed), p. 31. 5. IDEMA, ''La intervencion del IDEMA en el mercado de alimentos y materias primas" (Bogota, Janu- ary 1983, processed), p 8. 6 Beef seems to be protected, as shown in table SA-15 The difference between the domestic and international price, however, is probably absorbed by the marketing margin from producer to whole- saler. 7. On the matter of comparative advantage and domestic resources costs see "Ventajas comparativas de productos agropecuarios en Colombia," vol. 1 (October 1982), a report prepared for the ONP by Econometria. 8 See Edward Tower, "Understanding Shadow Prices, Second-Best Tariffs, the Effective Rate of Protection and Domestic Resource Cost from the Perspective of Simple General Equilibrium Models" (Washington, D.C.: World Bank. July 1983, processed) 9. Mateen Thobani, "Welfare Impact of Reducing Import Restrictions on Wheat" (Washington, D.C. World Bank, Country Policy Department, April 1984, processed). 10. IDEMA'S selling price is much higher than its import price plus transport and handling costs and is only feasible because of a government policy of restricting imports through licensing. Thus the support price policy and import restrictions policy go hand in hand. 11. The relative prices of two activities means the ratio of the implicit price deflator of gross output for these activities. When no mention is made to the contrary, the term "relative price" refers to domes- tic relative prices. In these measures agriculture is defined to include sectors 01, 02, 03. 08, and 12 in the PRICE INTERVENTIONS, COMPETITIVENESS, AND INCENTIVES 79 DANE classification of national accounts The rest of the economy comprises sectors 04, 05, 06, 07, 09, 10, 11, and 13-35. 12. The average nominal tariff was reduced from 70 percent in 1970 to 30 percent in 1974. See Roberto Junguito and Carlos Caballero, Problemasyperspectivas del proceso de integraci6n andina (Bogota: FEDESARROLLO, 1974) The liberalization process between 1970 and 1974 is documented byJorgc Garcia Garcia in The Effects of Exchange Rates and Commercial Policies on Agricultural Incentives in Colombia- 1953-1978, International Food Policy Research Institute Research Report no. 24 (Washington D.C : iFpRI, June 1981), chapter 6 13. World Bank, World Development Report 1983 (New York: Oxford University Press, 1983); Ramgopal Agarwala, Price Distortions and Growth in Developing Countries, World Bank Staff Working Paper no. 575 (Washington, D.C., 1983). 14. Urrutia, Winners and Losers in Colombia's Economic Growth of the 1970s. 5 Price Stabilization in Agriculture INSTABILITY OF DOMESTIC PRICES iS considered by sources in and out of the govern- ment as a serious problem for producers and consumers of agricultural commodities in Colombia. Greater price stability is often recommended on various grounds, and some policies have been put into effect and others proposed for this purpose. Poli- cies connected with stabilization concern two distinct but interrelated aspects: year- to-year price variation and seasonal price fluctuations. These two types of price instability will be discussed in turn, and related macroeconomic and sectoral poli- cies and other alternatives will be evaluated. FEDERACAFE (the Coffee Federation) reduces domestic price changes of coffee in the face of volatile variations in international markets. The Oficina de Planeaci6n del Sector Agropecuario (OPSA, the planning office of the Ministry of Agriculture) sets producer prices, usually once a year, for sugar, cocoa, sisal, and fiber, and intervenes between producers and cotton manufacturers in the negotiation of cot- ton prices. The agency may not be an effective price stabilizer, however, since it does not actually buy or sell any of these products. IDEMA sets support prices for producers and influences consumer prices through its sales and imports of staples such as rice, wheat, corn, barley, beans, sorghum, and soya. Many of these are seasonal crops; the agency's purchase, import, storage, and sale policies therefore affect seasonal variations in price. In addition, the government policy of subsidized credit for storage has important implications, which will be discussed later. IDEMA'S nationwide contribution to seasonal and annual price stabilization may be limited, however, since-in view of its financial constraints-it intervenes substantially only in a few markets, as is shown in table 5-1. Some Basic Considerations A concern of policymakers is that high seasonal prices of food items hurt poor consumers who spend a large share of their income on food. This notion confounds 80 PRICE STABILIZATION 81 Table 5-1. IDEMA's Purchases as Percentage of National Production of Each Crop, 1980-84 Crop 1980-81 1981-82 1982-83 1983-84 Beans 0.6 n.a. 2.0 n.a. Corn 2.3 5.2 5.4 7 6 Rice (paddy) 4.4 10.3 10.8 2.7 Sesame 64.7 22.9 3.0 0.7 Sorghum 4.2 14.7 6.5 3.7 Soybeans 0.9 1.1 00 n.a. Wheat 41.6 60.1 52.1 35.7 Barley n.a. 2.1 0.3 n.a. n.a. Not available. Source: IDEMA. the conceptually distinct issues of the average price level and the variations of the price during the year. Stabilizing the price would imply prices lower in some sea- sons and higher in others than any present prices. Consumers are clearly better off in the former seasons and worse off in the latter than under a regime of unstable prices. On balance, consumer welfare may actually be reduced throughout all sea- sons by price stabilization.' It is believed by some observers that price uncertainty leads producers to make wrong production decisions. Producers, uncertain about the future price, base deci- sions on the current price, thereby causing alternate booms and busts. If producers were to learn from experience, however, they would find that high prices one year tend to be followed by low prices the next year, and they would make the appropri- ate adjustment in planning decisions, leading to a leveling off of prices. Where IDEMA 1s effective, it is difficult to understand why producers would base planting decisions on current prices when IDEMA publishes future support prices far enough in advance to base plans on them. Another consideration is that reduction of agri- cultural risks could be expected to promote investments, expand production, and reduce consumer prices. Concerns regarding the macroeconomic effects of price instability are based on the assumption that consumers and producers have different marginal propensities to consume.2 Thus, it is often argued that when fluctuations in agricultural prices and production cause shifts in income between the two groups, the demand for individual products-and aggregate demand-also fluctuate, destabilizing the rest of the economy. It should be noted, however, that if incomes of producers of different crops are not highly correlated, large shifts of income from agricultural producers as a group to consumers or vice versa would be rare, since the shifts in some crops in a given year would tend to be offset by the opposite shifts in other crops. It is seldom clear in discussions of stabilization whether the objective of actual or proposed policy is to stabilize prices, incomes, or some other variable. Additionally, 82 AGRICULTURAL PRICE POLICY sometimes the task involves insulating domestic markets from variations in interna- tional prices, whereas in other instances the source of variation is instability in do- mestic production. In all instances, the effect of government intervention per se and its costs have been inadequately evaluated. The only effort being made in this study is to initiate an analysis of these issues, and the findings offered in the rest of this chapter on seasonal and annual variations in prices will need to be further analyzed. Indexes of variability and of differences between realized and predicted values of three variables for different crops have been computed: prices, profit-net income per ton and gross income per ton, if annual cost figures were unavailable-and annual return on land investment. Domestic prices at the producer and consumer levels are more stable and predictable than are international prices for the crops considered, namely, coffee, rice, sugar, wheat, cotton, corn, barley, and potatoes. It is not clear whether on the average this finding is obtained from government inter- vention, or is the consequence of farmers' and traders' own storage and other ef- forts, or both. For coffee, the stabilization program of FEDERACAFE has had a signifi- cant effect. There may also be some additional fragmentary evidence on the price stabilization effects of government policy: with respect to potatoes, with little or no government intervention, the variability of domestic prices has not been much less than the international variation, whereas in the other cases, as already noted, varia- tion in prices is less domestically than internationally. On the other hand, price is perhaps not as good an indicator of risk or uncer- tainty as income and profitability. It is well known that stabilizing prices does not necessarily stabilize income or profitability. An example is a product with random production and a demand elasticity of unity. In such an instance, although produc- tion fluctuates from year to year, low production is offset by high prices and vice versa, so income is stable in the market without any intervention. On the other hand, if the price in this instance is stabilized by an external agency, the income of producers fluctuates with production, and price stabilization destabilizes income. For potatoes, total income and income per hectare seem to be more stable in do- mestic prices than in international prices by about the same amount as other crops. Judged by these criteria, therefore, little credit can be attributed to government programs in stabilizing farm income. In any event, for all crops, including potatoes but excepting cotton, producer prices and incomes vary less if they are evaluated at domestic prices than at interna- tional prices. In selected instances, such as coffee, domestic price stabilization can be attributed to policy, but such a relation is less clear in other instances. Furthermore, income stabilization appears to be achieved, at least in respect to potatoes, with much less price stabilization than in other instances. It appears that with or without government intervention, domestic variables tend to be less volatile and, to a lesser extent, more predictable than international variables. It is not clear why this should be true, though a study in Malaysia found that producer prices were naturally buf- fered from external fluctuations because middleman margins were positively corre- lated with these prices. PRICE STABILIZATION 83 Addressing Year-to-Year Instability Any plan that solves the microeconomic problem of production risk by breaking the link between the income of producers and the expenditures of consumers may exacerbate the macroeconomic destabilization problem. With no stabilization plan, agricultural production and price shifts cause transfers of income between pro- ducers and consumers of agricultural commodities. In years when producers' in- comes and their expenditures on nonagricultural commodities are low, consumers' expenditures on agricultural products are also low, and their expenditures on non- agricultural commodities are high. Thus the overall demand for the nonagricultural commodities tends to stay at a constant level, although it may fluctuate somewhat because of differences between agricultural producers and consumers in marginal propensities to consume. But suppose the link between producer income and con- sumer expenditure were to be broken-in a case in which producer income is stabi- lized by the provision of countercyclical credit, for example, with concessionary credit made available to producers in bad years. Consumers' expenditures on non- agricultural products would be high, as before, but they would no longer be offset by a decline in producers' expenditures on these products. Shifts in total expendi- tures would be more pronounced, and the net effect of stabilizing producers' in- come would be destabilization of nonagricultural demand. It can also be shown that in general there is no presumption that a buffer-stock scheme would stabilize non- agricultural demand; it might well destabilize it. Export Price Stabilization For internationally traded crops, domestic prices can be stabilized by import or export taxes or subsidies without breaking the link between producers' income and consumers' expenditures. A law approved by the congress facilitates the use of vari- able export incentives to stabilize domestic prices of exportables. With the CAT system, changes in an export subsidy rate were made administratively, but only once a year. Under the new system, CERT, the rates-five levels, with a maximum of 25 percent, announced by Decree 637 of 1984-are expected to be changed more often, greatly increasing the flexibility of the system. When the world price of a product increases, its rate can be decreased quickly, thus preventing its domestic price from rising very much; the opposite can take place when its world price falls. As noted, this kind of plan has the advantage that consumers' expenditures are still linked to producers' incomes, avoiding the macroeconomic side effects of some other stabilization plans. It is also probably less burdensome to the economy than many other possible plans, such as direct import management by the government and quotas. 84 AGRICULTURAL PRICE POLICY One limitation, however, is that its scope of coverage is limited to exports. It is, of course, possible to expand the coverage to importables, as the European Com- munity (Ec) does with its variable levies, but in any event it cannot be used for nontraded items. A second limitation is that the size of the export subsidies is lim- ited by GATT. (In fact, the main purpose of the flexible plan is to allow incentives to be changed quickly if other signatories to GATr complain; the price stabilization function is secondary.) As a practical matter, this may not be very important for some crops, but may be a real limitation on the ability to stabilize the prices of others. One disadvantage of this sort of plan is that there is a temptation to use it to increase distortions of production incentives beyond what they would be other- wise. The Ec has used its variable levy system not only for stabilization, for exam- ple, but also for protection. Certainly, it is possible to stabilize prices without in- creasing average protection of a good beyond existing levels, but it is probably difficult politically. For an import that is not protected now, this would involve taxing imports in years of low world prices and subsidizing them when world prices are high. It may be quite difficult to subsidize imports in the face of the opposition of domestic producers. If the stabilization efforts should produce greater protection from foreign competition, the economic costs may be high. A second disadvantage is the economic inefficiency involved in failing to react to world market signals. Economic efficiency demands, for example, that an export product be produced and exported at a level such that its marginal cost is equal to its price. But if producers are insulated from world prices, the price on which they base production decisions is not the price that reflects the true social value of the prod- uct. If producer prices are stabilized, a country will not export enough of an export product when world prices are high to take full advantage of the good prices, and it will export too much when world prices are low. Rough estimates of this kind of efficiency or welfare cost-defined as the sum of loss in producer and consumer surpluses in comparison with a situation without interventions-are presented in table 5-2 for several crops. While the annual costs for any single crop are not over- whelming, the cumulative value of these costs carried into the indefinite future can be substantial. It should be noted, however, that these are welfare losses from per- fect price stabilization. To the extent that a price stabilization scheme does not make the price perfectly stable, the costs will be correspondingly less. A third disadvantage is that the plan requires government officials to make judg- ments about whether observed world price movements are transitory or part of a long-run trend. If the movements are part of a trend, but are thought to be transi- tory, the rate will be adjusted in the expectation that it will be readjusted when the price returns to its previous level. Since the price will never return to this level, the adjustment becomes a permanent change in protection of the good. This disadvan- tage could be mitigated to some extent by basing the CERT adjustment on the devia- tion from the long-run price. This long-run price would be recomputed each year as a moving average of previous years. When the price moved to a new level as part of PRICE STABILIZATION 85 Table 5-2. Welfare Cost of Domestic Price Stabilization for Some Import and Export Crops (million 1975 pesos) Net present value Crop Annual welfare cost of cost, Wheat 59.4 848.6 Corn 60.9 8700 Barley 0.9 12 9 Rice 231.7 3,310 0 Cotton 108.6 1,551.4 Potatoes 126 3 1,804.3 a. Net present value of an infinite stream of the annual cost, evaluated using a real interest rate of 7 percent. Source: See appendix E. a long-run trend, this would at first appear to be a large deviation, leading to a large compensating adjustment in the rate. But as the price remained at its new level, it would be seen as less and less of a deviation, and the adjustment would become smaller each year. Futures Market Colombia might consider diminishing the risks involved in agricultural produc- tion and processing by allowing producers and processors access to futures markets. Hedging in futures markets does not stabilize incomes, but it does eliminate uncer- tainty about price, which is one of the problems about which the government is justifiably concerned. Eliminating risk in this manner can be, by and large, without cost to the government, and it is free of any efficiency costs as well, differing in both respects from the support price policy of IDEMA and such price stabilization plans as CERT. HEDGING FOR RISK REDUCTION. Hedging involves the buying and selling of futures contracts in such a way that movements in the value of the contracts offset the movements in the price of the commodity that the hedger will buy or sell in the future. An idealized hedging transaction might proceed as follows:' An importer of wheat who will need to buy wheat on December 1 buys a futures contract on a U.S. commodity exchange calling for delivery of wheat on that date, paying $ 150 a ton. The contract is an enforceable agreement, and if he wanted, the importer could accept delivery of the wheat. But, for any of several reasons, the importer may not wish to buy the wheat through the futures market and need not accept delivery to relieve himself of the risk of paying a price different from $ 150. Sup- pose, for example, that on November 30, the spot price of wheat-that is, the price for immediate delivery-is $160 a ton. Then his futures contract, which on this 86 AGRICULTURAL PRICE POLICY date is essentially a contract for immediate delivery, is worth $160. The importer could sell his contract for $160, making a profit of $10 on his transaction in the futures market. He could then apply this profit to buying wheat from another source for $160, paying $10 more than he expected to pay. He would finally have paid, in effect, a price of S 150. (Note that if the price of wheat from this source is not $ 160, then the final price he has in effect paid is not exactly $ 150. But as long as the prices of wheat asked by his source are well correlated with prices in the U.S. market, the profits or losses in futures market transactions will tend to offset move- ments in his prices and decrease his risk accordingly.) Similarly, an exporter of, say, cotton could reduce his risk by hedging. If the exporter wished to sell cotton on December I at an assured price, he could sell a futures contract for delivery on that date at the going price for such a contract-say, eighty cents a pound. On December 1, if the actual price he received for his crop was seventy cents a pound, he would nevertheless receive in effect a price of eighty cents, because he could liquidate his position in the futures market-that is, buy back the contracts he sold for eighty cents-at a price of seventy cents, thereby making a profit of ten cents a pound. OTHER ADVANTAGES. In addition to the advantage of giving importers and ex- porters the opportunity to reduce risk from price fluctuations, participation in fu- tures markets would have an important concomitant advantage.' The spread be- tween the current spot price and the futures contract price provides valuable information on the way the market expects the price to move, since the futures contract price is a good predictor of the spot price that will prevail on the date the contract matures. This kind of information is valuable in making decisions about storage and inventory control. One analyst argues that this function of futures mar- kets is of greater value than the reduction of price uncertainty.' POLICY PROBLEMS. While the value of futures markets, especially as they relate to reduction of risk, has been widely recognized by academic and government sources, their use by Colombians has been rare. In fact, it appears that only three economic agents have used futures markets-for several reasons.' First, and perhaps foremost, government exchange controls have made participation difficult. The government has, to some extent, been reluctant to authorize the use of foreign exchange for this purpose for fear that hedgers will begin to speculate and require large quantities of foreign exchange to pay their losses. Even in those instances in which it is willing to do so, the licensing procedure is time-consuming and does not allow hedgers much flexibility in the timing of purchases and sales of contracts. Second, strict controls on imports decrease the usefulness of hedging. As implied by the foregoing examples, for maximum reduction of risk the physical commodity must be purchased in the same market as the futures contract. Otherwise, even on the contract maturation date, the value of the contract may differ from the value of the commodity. Import controls make the purchase of commodities in world mar- PRICE STABILIZATION 87 kets more difficult. Of course, some reduction of risk is still possible, to the extent that world and domestic prices are well correlated, but the effectiveness of hedging is reduced. Similarly, for goods imported by IDEMA, or imports of which are li- censed by IDEMA, hedging is made less effective in reducing risk by the fact that decisions are made a very short time before the importation actually occurs. Third, the unpredictability of government trade policy makes hedging riskier. If an importer hedges in the futures market, planning to import a commodity, for example, and the government then places that import on the prior-license list, the importer is in effect changed from a hedger to a pure speculator, and his risk has increased manifold. Fourth, for some agents hedging is not very useful because prices are controlled by the government, a policy that makes prices on the whole predictable. This is true of producers of crops whose future support prices are fixed and are announced before planting time. It is also true of export crops to the extent that their domestic prices are fixed by the CERT plan. This factor appears to have had much to do with the decision of one cloth manufacturer to stop hedging after the government began fixing domestic cotton prices. Finally, inadequate informa- tion about the value of futures markets may also help explain why they are rarely used. Some of these problems would he mitigated by the development of a domestic futures market. In the long run, it is possible that the Bolsa Agropecuaria will de- velop into a domestic futures market. In fact, contracts are now available through the Bolsa which call for delivery-in 150 days-of crops that are nonexistent when the contract is signed. But since these contracts are not tradable, the transfer of risk to speculators is not possible. The Bolsa also has quite limited market participa- tion-only about 2 percent of Colombia's agricultural output is sold through the Bolsa-and is completely dominated by IDEMA, which participates in 70-80 percent of all the transactions-7 3 percent in 1982. The Bolsa is still young and is struggling to work out a number of problems that prevent it from attracting greater participa- tion. In time, it may succeed in attracting a sufficient number of participants to make IDEMAa relatively small part of the Bolsa and to develop the economies of scale that are necessary if a futures market is to function efficiently, but until this comes about, there is little chance that it can function effectively as a futures mar- ket.7 In the meantime the government might remove obstacles to participation in international futures markets. Seasonal Price Fluctuations One objective of the government is to ensure that food prices do not rise unduly in the periods between harvests. Such seasonal stabilization can in principle be facil- itated by varying the timing and quantities of imports and exports of agricultural products. In practice the country's internal storage system is equally important in 88 AGRICULTURAL PRICE POLICY assuring adequate supplies and moderating prices in the nonharvest months. This section will be concerned with the effects of storage policies, with particular atten- tion to two issues: first, whether the existing system of subsidized loans for storage, the bonos de prenda (BP), is desirable and efficient, and second, whether IDEMA'S sales policies tend to discourage private storage activities. Considerable storage capacity is owned by IDEMA, industrial processors, pro- ducers' associations, and twelve general facilities, called Almacenes Generales de Dep6sito (AGD), which are subsidiaries of private banks.' Identifiable on-farm stor- age capacity (excluding coffee plantations) in 1977 was only 0.91 percent of the total.9 This figure is probably an understatement of the available capacity, but field observations suggest that on-farm capacity may be limited. Most individual storers use rented space in AGDS, which are generally owned by the banks with which those storers do business. Since some banks specialize, for example, by catering to the needs of growers of particular crops, these banks also specialize in the storage of particular commodities. When someone stores a commodity in an AGD, he receives a deposit certificate, or Certificado de Dep6sito (co), specifying the quality and quantity of the merchandise. This certificate then becomes a title to the property and can be bought and sold in the market. It may be returned to the bank as collat- eral for a loan under the BP system, which will be described below. Subsidized Credit for Storage Under the bonos de prenda system, a bank may make a loan at a subsidized interest rate, using the CD as collateral, and rediscount part of the loan through the Central Bank. Only certain crops qualify for loans under this system, but the list is fairly extensive: African palm oil, sesame, cotton fiber, cottonseed, anise, paddy rice, cocoa, coffee, barley, sisal, beans, powdered milk, corn, potatoes, sorghum, soybeans, tobacco, wheat, and products for export. A serious limitation is that only growers, processors, and IDEMA are eligible for loans under this system, thus exclud- ing wholesalers and other middlemen. In general, the loans can be for as much as 80 percent of the value of the stored crop, evaluated according to a basic price estab- lished by the Monetary Board.10 This share is known as the discount margin. The subsidy element in the 1P loans can be substantial and has changed quite a bit since the system was established, as shown in table 5-3. Of course, the "subsidy" is not measured strictly by the difference between the market rate and the BP rate, because the subsidized loan is made on only 80 percent of the value, based on the basic price, which may deviate substantially from the market value. That is, to use the BP system is as if a storer incurred all the normal storage costs, but received a rebate in the amount of 0.8 PbQ(r, - r,) a year, where Pb is the basic price, Q is the quantity stored, r, is the real market interest rate, and r, is the real subsidized rate." This subsidy element depends as much on Pb as it does on the difference between rm and r,. The value Pb may vary with the passage of time and with different crops; therefore, the subsidy element is different in different years and PRICE STABILIZATION 89 Table 5-3. Market Interest Rates and Bonos de Prenda Rates, 1970-81 (percent) Interest rate Real market bonos de prenda Market Rate of interest Year interesta inflationb (ex post) Nominalc Real (ex post) 1970 13.3 7.2 6.1 100 2.8 1971 16.4 12.6 3.8 10.0 -2.6 1972 15.6 14.1 1.5 10.0 -4.1 1973 20.3 22.1 - 1.8 10.0 - 12 1 1974 30.4 25.2 5.2 15.0 - 10.2 1975 23.8 17.5 6.3 17.0 -0.5 1976 22.4 25.4 -3 0 17.0 -8.4 1977 22.9 27.5 -4.6 17.0 - 10.5 1978 25.9 19.7 6.2 17.0 -2 7 1979 36.5 26 5 10.0 17 0 -9.5 1980 41.5 24.5 17.0 24.0 -0.5 1981 52.5 25.6 26 9 28.8 3.2 a. CAT, 120-day maturity, average annual rate. b. Consumer price index. c. Average discount rate. Sources. Calculations by the DNP, based on Banco de la Republica, Resoluciones de la Junta Monetaria, various issues, and the Asociaci6n Bancaria (Asobancaria, the banking association) with different crops. Table SA-20 shows the evolution of the ratio Pb/Pm for dif- ferent crops, where Pm is the market-producer price of the product. The subsidy can be written as a fraction of the market value of the crop.12 Table 5-4 shows the value of the subsidy element as a percentage of market value for several crops in 1980 and 1981, the last years for which full information was readily available. (Where appropriate, the 0.80 has been adjusted to 0.70 or 0.75.) These figures are expressed as annual percentages and are thus directly comparable to the financial cost of storage. Storing barley with a subsidy of 12.05 percent and a market interest rate of 41.5 percent, for example, is exactly as costly as storing barley at a market interest rate of 29.45 percent (= 41.5 percent - 12.05 percent), with no subsidy given. So from the point of view of storers the BP system in 1980 and 1981 can be considered equivalent to a reduction in the market interest rate by the amounts shown in table 5-2. The subsidy rates vary from a low of 5.12 percent (palm oil in 1980) to a high of 17.26 percent (barley in 198 1), with unweighted averages of 9.65 percent and 13.55 percent for the two years. Both the discount margin and the interest rate charged on the rediscount are set by the Monetary Board. As of July 29, 1981, the discount margin was 40 percent and the interest rate on the rediscount was 25 percent. At the same time, the dis- count rate on the BP was 30 percent and the margin of discount was 80 percent for most crops; that is, for every 5 100 worth of stored crops, evaluated at the official 90 AGRICULTURAL PRICE POLICY Table 5-4. Value of Subsidy to Storers as Percentage of Market Value of Stored Crops, 1980-81 (percent) Subsidy' Crop 1980 1981 Barley 12.05 17 26 Beans 6.41 11.35 Cocoa 10.91 13.43 Corn 7 17 12.86 Cotton fiber 14.32 14.28 Cotton seed 10 30 14.09 Palm oil 5.12 14.53 Rice (paddy) 10.29 12.45 Sesame 13 23 13.27 Sorghum 8.44 12.14 Soybeans 8.47 15.67 Wheat 8.98 11.32 Unweighted average 9.64 13.55 a. S 0.8(PhIP,)((r, - r,), where S = annual subsidy, Pb = basic price, P, = market price, , = market interest rate, r, = subsidized bonos deprenda discount rate. For wheat and palm oil, 0.8 in this formula was replaced with 0.7 Sources: Computed from tables 5-3, SA-20, and SA-21. basic prices, a storer could borrow S 80 from the commercial bank at an interest rate of 30 percent; the bank could then rediscount $32 (40 percent of $80) at the Cen- tral Bank at an interest rate of 25 percent. The bank's effective rate of return on its loan can be computed as [0.30 - (0.25)(0.40)] / (1 - 0.40)= 0.33. It is clear that in recent years the incentive for banks to make loans through BP rather than through normal market channels has declined, contributing to the decline in the real volume of RP loans. Costs of Subsidized Credit First, it is costly to the government and to lending institutions to subsidize loans to storers. Table SA-22 shows the total cost of the program for each major product in 1981 and how the cost is divided between banks and the government.3 Al- though the exact ratio varies from crop to crop, depending on the ratio of discounts to rediscounts, in general private banks bear at least as much of the cost of subsidiz- PRICE STABILIZATION 91 ing crop storage as does the government and much more of the cost of subsidizing manufactured and export goods. The main recipients of the subsidy are a few large firms in concentrated indus- tries. In 1978, the last year for which these data were readily available, one firm used 18.6 percent of the total loans rediscounted, the largest four firms used 46.6 percent, and the largest eight firms 59.1 percent." This tendency is even more striking in certain industries: in cotton fiber, barley, tobacco, and cocoa, four firms used between 85 percent and 100 percent of the loans rediscounted. The use of the subsidy by growers, including growers' associations, was only 1 percent of the total. If the banking system is more or less competitive, the portion of the cost borne by it-half or more-would ultimately be passed on to savers (in the form of lower interest on deposits) and to borrowers (in the form of higher interest charged on loans). It was not possible to estimate the relative magnitudes of these burdens without a detailed analysis of elasticities of supply and demand for lendable funds. The portion of the cost paid in the first instance by the government is ultimately financed by an "inflation tax." The rediscounts of the BP by the Central Bank are financed by the creation of money. This can cause a rise in the general price level, in effect taxing the holders of cash balances and transferring purchasing power from holders of cash balances to the recipients of the subsidy. In general, the inflation tax would be expected to be regressive; that is, the rate of taxation-as a percentage of wealth-declines with the absolute level of wealth, because the poor hold a larger share of their wealth in the form of cash. On balance, therefore, the aP system operates to transfer resources from savers, investors, and holders of cash to a rela- tively small number of industrial firms. The second type of cost involved concerns efficiency. The system subsidizes firms' storage activities, and by doing so, it encourages the diversion of investment from other activities to storage, to such an extent that the true social marginal value of the storage is less than the marginal value of alternative investments. It is difficult to estimate the magnitude of the efficiency loss without a detailed analysis of elastic- ities. The size of the distortion is certainly large enough to create a large loss, how- ever, especially for certain crops (see table 5-2). It is, of course, possible that the subsidy to storage could be optimal in the presence of other distortions. This does not seem likely, however, because the type of distortion that would make this sub- sidy optimal-subsidies of a similar size for other investment activities of the firm- are not generally present. Reforms of the System First, if the government desires to encourage agricultural storage, it could do so in a way that would benefit all storage activity. There does not seem to be any legiti- mate reason for excluding middlemen from the subsidy, especially since the princi- pal alternative beneficiaries at present are large processing firms. Second, the subsidy should be divorced from the money supply function of the 92 AGRICULTURAL PRICE POLICY Central Bank. This linkage between rediscount of BP and the expansion of the mon- etary base has two peculiar effects. One is that seasonal increases and decreases in agricultural production and the consequent demand for storage to some extent tend to reduce control of the supply by the monetary authorities on a year-to-year basis and introduce a seasonality into the monetary base. This problem could be serious if the rediscount formed an important part of the monetary base, although that part was only around 1.8 percent in 1984. The effect on monetary stability is actually greater than this figure would imply, however, because changes in BP rediscounts are reflected, peso for peso, in changes in the monetary base. Changes in the base, in turn, change the money supply through a multiplier effect. Thus, for each peso by which Br rediscounts vary, the money supply changes by several pesos. If BP rediscounts should ever again form a significant share of the monetary base-8.4 percent, for example, as in 1974-this would be a source of concern. The more serious linkage runs the other way: changes in monetary policy tend to have a direct effect on storage activity. In periods of tight monetary policy, the rediscounts available for BP tend to be contracted, discouraging storage. In one sense, this is as it should be; tight money always raises interest rates, raising the cost of storage. When monetary policy affects storage through its intermediate effect on interest rates, however, it also affects other competing investment activities, both private and public, in the same way. Given the way in which the BP system is set up, it is not possible to assure this result; BP rediscounts may be affected to a greater or lesser extent with respect to alternative investments by monetary changes, depend- ing on political decisions. Perhaps a better system would be to include interest subsidies for storage as a line item in each year's budget. A person taking out the type of loan currently eligible for BP would be charged a nonsubsidized rate of interest by the bank but would then be reimbursed directly by the government for some part of the interest cost, perhaps using the bank as an administrative agent. Presumably the fraction would remain constant for several years. In this way, storage would always be given the same advantage as other investments, but its cost would rise or fall as it should along with the cost of other investments as monetary or real factors in the economy caused the interest rate to rise or fall. An additional advantage is that it would finance the government's share of the subsidy from general revenues, rather than by the creation of money, thus avoiding the possibly regressive inflation tax. The share of the subsidy now financed by the banks could be taken over by the govern- ment-perhaps financed by a tax on the banks-or could be left with the banks by mandating that they give storers a rate higher than the current discount rate (since banks would no longer be partially reimbursed by subsidized rediscounts) but be- low the market rate. Third, the mechanism for allocating credit among various crops might be changed. At present the amount of credit available to a storer depends on the basic price set for each crop by the Monetary Board. This price is used to establish a value for the physical quantity stored, and this value is the basis for the size of the PRICE STABILIZATION 93 subsidized loan available. These basic prices have little connection to market prices but rather, to a large extent, are set on the basis of political considerations. As might be expected, the basic prices are the subject of lobbying efforts by producers' associ- ations." Consequently, the ratio of basic prices to market prices-and the subsidy element-vary widely among products, and the products with weak producers' associations generally lose out. The contrast between the subsidies available for cotton, with a strong producers' association, and beans, an important food crop with a relatively weak producers' association, can be seen in table 5-2. This distor- tion tends to cause inefficiency by encouraging storage of some crops more than of others, with no apparent rationale. As table 5-2 demonstrates, the size of the bias tends to change in time. When the ratio of basic prices to market prices of cotton and beans, for example, is compared, cotton always has an advantage, but the magnitude of the advantage varies from 1 percent (in 1976) to 22 3 percent (in 1980). This bias could be eliminated by setting the ratio of basic price to market price at the same level for all crops. The overall ratio could be adjusted each year as a means of controlling the demand for subsi- dized credit. Alternatively, the ratio could be set at unity every year, and the credit could be rationed by changing the rediscount interest rate or the margin of redis- count. Of course, this rationing would all be unnecessary if the whole scheme were revised in such a way that basic prices would be set at the same level as market prices each year, and the demand for storage loans each year would be determined by a uniformly subsidized interest rate. The average demand for several years, of course, would be determined by the size of the subsidy offered. IDEMA's Function in Storage IDEMA is the single most important agricultural storage agent in Colombia. It is full owner of 10.7 percent of total storage capacity, is partial owner of another 5.6 percent (Almagrario and Almapopular), and sometimes rents space from privately owned AGDS.16 It is much more important in the storage of some crops-especially rice, wheat, corn, and sorghum-than others. As indicated by table 5-5, in some products, IDEMA'S storage and sales policies can have a significant effect on market conditions, including incentives for private storage. The profitability of storage is determined by the difference between the purchase price of a commodity and the sale price after storage compared to storage costs." In a competitive storage indus- try, storage each year would be sufficient to ensure that, for the marginal firm or the marginal units, the expected price spread was just sufficient to cover the costs. The policies of IDEMA affect the price spread. GOALS AND POLICIES OF IDEMA. The pricing policies of JDEMA are rooted in two of its perceived goals, in addition to supporting producer prices: to keep consumer prices low and to aid the government in its fight against inflation. In the past, these goals have caused IDEMA to pursue unrealistic pricing policies. After buying and 94 AGRICULTURAL PRICE POLICY Table 5-5. IDEMA's Share of Total Stocks of Selected Crops, 1979-81 (percent) White Imported corn Year Month rice Wheat and sorghum 1979 June 32 89 21 December 22 16 19 1980 June 29 42 11 December 39 32 59 1981 June 38 56 51 December 14 52 47 Source: Computed from IDEMA: Dos anos de labores, 1980-81 (Bogota: IDEMA, 1982). cuadro no. 12, p. 42. storing commodities, IDEMA would sell them for little if anything more than it paid, thereby failing to cover its own storage costs. Its recent policy has been an attempt to operate on a financially sounder basis, while still meeting the same goals. It com- prises the following three operational rules: (1) the selling price is not to exceed an amount equal to the purchase price plus carrying cost of 3.5 percent a month; (2) the selling price is not to rise more than 20 percent a year, in any event; (3) if market conditions at any time are such that the market price is lower than the price that would be indicated by rule 1, IDEMA Is to sell at the prevailing market price. Rule 1 reflects the goal of financial responsibility, rule 2 reflects the goal of fighting inflation, since 20 percent is the government's inflation target, and rule 3 reflects the goal of assuring low consumer prices. EFFECTS OF IDEMA S POLICIES. To the extent that the policy outlined above is effec- tive, it is likely to have a negative effect on private storage, although the analysis would have to be revised if the actual policy were to function differently in practice. First, even if prices always rose 3.5 percent a month, this would be insufficient to cover private storage costs. IDEMA calculated the cost of storage as 3.5 percent by using the subsidized BP rate of interest as the financial cost of storage, then adding its other costs, such as physical spoilage. BP loans are only available, however, to cover 80 percent or less of the value of the stored crops evaluated using the basic prices. The remaining 20 percent must be financed at the market rate. In addition, since the ratio of the basic price to the market price is less than 1 for almost all crops and differs substantially among crops, the true financial cost of storage is always higher than the BP rate and differs from crop to crop. It also changes with variations in the market interest rate. Table 5-3 showed the direct interest subsidy equivalents of the BP system in 1980 and 1981. That is, with the schedule of basic prices in those years, and market and e rates as shown in tables 5-3, SA-2 1, and SA-22, the effective financial cost of PRICE STABILIZATION 95 storage was reduced by the amount shown in table 5-3 for each crop. The effective simple annual financial storage cost of rice, for example, with a subsidy equivalent of 10.29 percent, was 31.21 percent (41.5 percent- 10.29 percent) in 1980, or, since the interest is compounded bimonthly, a true cost of about 36 percent a year or 3 percent per month. IDEMA sources suggest that when the BP rate is 24 percent, as it was in 1980, the financial cost of storage might be 2.31 percent a month. This may be a substantial underestimate of the true cost, and the underestimate is even more serious for crops whose subsidy equivalent is smaller. Obviously, the magni- tude of the underestimate becomes smaller as the market interest rate approaches the BP rate, but it would always be an underestimate unless the two were equal. In addition, the 3.5 percent figure seems to be an estimate of IDEMAS marginal cost; that is, it does not include fixed costs, such as the cost of construction of storage facilities, nor does it adequately cover other nonfinancial costs, which are estimated to be significant." Consequently, storage would be uneconomical for private stor- ers-and, indeed, for IDEMA also-if IDEMA succeeded in limiting the price increase to 3.5 percent a month. Second, the 20 percent a year ceiling on price increases would make storage for a fairly long term uneconomical in a period of 20 percent general inflation. (And, presumably, it is because inflation is expected to be 20 percent or greater that the 20 percent ceiling was chosen as an anti-inflation tool.) The fundamental reason for this is that the price of the stored crop must rise in real terms-that is, at a rate greater than the rate of inflation-in order to cover the cost of storage. Consider the case most favorable to a policy of limited price increases, a hypothetical crop that is harvested almost the year round and has to be stored for consumption for only a short time; even so, a 20 percent ceiling is insufficient. Suppose, for example, that the harvest is spread nearly evenly over nine months of the year, so that con- sumption must come from storage for only three months. In a period of 20 percent annual inflation, the price of the crop would rise during the nine months of harvest by about 15 percent because of the general inflation. After the end of the harvest, when consumption must come from stocks, suppose that IDEMA regulates its stocks so that the price rises 3.5 percent a month. At the end of one month and one week of storage, the price will be 20 percent higher than it was at the beginning of the harvest about ten months earlier. If IDEMA then holds the price constant, storage will be a losing proposition for the rest of the year. In general, if the underlying rate of inflation were about 20 percent, the price would rise 20 percent from the beginning of one harvest to the beginning of a harvest one year later. But this does not mean that the price during the year should not rise to a level higher than 20 percent above the original price. It must rise to a higher level in order to make it profitable to carry stocks, even though it will fall when the new harvest begins. That is to say, on average the price would behave as shown by the solid line in figure 5-1. If marginal consumption must come out of storage beginning at to + 9, storers must be reimbursed for the real costs of carrying stocks-real interest costs, physical losses, and so on-over and above the nominal 96 AGRICULTURAL PRICE POLICY Figure 5-1. Inflation, Seasonal Price Rise, and Storers' Profits Price In(I 28P,)- In(1.2P,) - In (1.15P,) ---- -- I in P, I I I ITime to t) + 9 to + 10 to) + 12 to + 21 harvest harvest harvest harvest begins ends begins ends Source: Estimate based on price analysis for 1982. gains in the value of their stocks caused by inflation. To place a 20 percent ceiling on price increases is to make storage for the last two months of the year unprofit- able. Finally, the interaction of rules I and 3 tend to make storage unprofitable on the average. The fact that profits from storage are stochastic is ignored. Storage deci- sions are normally based on expectations about the prices at which the stored crop will be sold in the future. When the crop is actually taken out of storage and sold, the price may be higher than expected, in which case profits are made, or the price may be lower, in which case losses are incurred. In a competitive market, the profits in some years balance the losses in other years, so on the average, costs of the marginal firm or marginal unit are just covered. But IDEMA'S rule 1 guarantees that the price will not rise rapidly enough to do more than cover costs, even in the best of times. And rule 3 guarantees that in times when prices are lower than a firm expected them to be, it would lose money. There is fairly convincing prima facie evidence that IDEMA has suppressed the rate of price increase in at least one crop- rice-below what it must be to make private storage profitable. Castillo Nii6o ex- amined the evidence from 1978 to 1982 and concluded that prices after harvest had frequently risen at a rate such that private storage would not generally have been profitable.' PRICE STABILIZATION 97 Reform of [DEMA'S policies in a variety of ways is now being studied and begin- ning to be executed, particularly in an effort to increase private-sector storage. In the future, to avoid displacement of private storage, the financial cost of storage could be calculated more realistically along the lines suggested here. So long as the BP loans are based on basic prices, the true financial cost will be different for each crop, and this should be reflected in a different pricing policy for each. Second, IDEMA should not try to act as an anti-inflationary tool, since it cannot suppress inflation by its pricing policies, and its attempts to do so will only distort its pricing decisions. Finally, if it is desirable to subsidize consumer prices, this might be done by lowering prices equally in all seasons rather than by decreasing the rate of price increase between harvests, in order to avoid long-term distortion of storage deci- sions and consumption patterns. Policy Conclusions The government has been considering ways and means of introducing greater price stability, particularly for internationally traded commodities. A variable ex- port incentive scheme, CERT, is now in effect, and an export price stabilization fund to be financed by the private sector is now being studied. An alternative and more efficient price-stabilizing scheme is that of allowing hedging in international futures markets, which would reduce price uncertainty for several imported products and some exported products. If importers and exporters could be encouraged to use international futures markets, it would provide the benefits of risk reduction directly to them and indirectly to domestic producers and consumers, to the extent that domestic prices move with international prices. In the absence of a viable domestic futures market, the government might consider taking steps to remove the obstacles to participation in international futures markets. First, legitimate hedging might be exempted in certain ways from rules governing other foreign exchange transactions. So long as the futures market activity is truly hedging-that is, is coupled with a transaction for the physical commodity-the possibility of significant losses in foreign exchange are remote. General rules could be developed for this purpose to ensure that futures markets were not used for speculation but to leave potential users sufficient flexibility to hedge effectively. Such rules might limit the exemption from exchange controls to agents who actu- ally deal in the commodity, limit futures market transactions to one set (buying and selling) per transaction in the physical commodity, and limit the size of the futures market transaction to the size of the transaction in the physical commodity. Second, a commitment might be needed to exempt hedging from changes in trade policy during the period of the transaction. This step would eliminate the risk that a policy change would prevent a hedger from consummating the part of the transaction involving the physical commodity, changing the position of the hedger to that of a pure speculator. Third, futures market activity might be substituted for 98 AGRICULTURAL PRICE POLICY government control of domestic prices for reducing the uncertainty of producers. While prices guaranteed by the government remove the incentives for hedging, guaranteed prices are less efficient and more costly to the government, and they are perhaps no more effective in reducing uncertainty for producers. Fourth, import restrictions on agricultural commodities might be gradually eliminated, and hedging in more commodities might thereby be encouraged. Storage permits the transfer of consumption from periods of low marginal value-that is, low price-to periods of high marginal value. Storage is costly, how- ever, and the optimal amount of storage is therefore less than that required to make prices constant throughout the year. The optimal storage should, in fact, be much less than this in an economic environment such as that of Colombia, where the economic cost of storage is high because of the high returns on alternative invest- ments. In general, market signals would be sufficient to give incentives for the opti- mal level of storage, and storage subsidies, such as the bonos de prenda, the subsi- dized credit system, are likely to be unnecessary. The bonos de prenda system has subsidized storage of different crops at unequal rates, with no apparent economic rationale for the differences other than special circumstances and problems of individual products and the disparate ability of vari- ous beneficiaries to exert political pressure. The system causes transfers of income from savers, investors, and holders of cash to a relatively small number of large producers and creates an undesirable linkage between storage and the monetary base. If it were necessary to give a subsidy to storage, it would be better to include it as a line item in the federal budget and administer it as a direct rebate to anyone- producer, distributor, or manufacturer-who stores agricultural commodities. The amount of the rebate could be determined as some fixed fraction of the financial cost of storage-the opportunity cost of funds tied up in the stored commodity, for example-based on the true market value of the crop at the time of storage and the market rate of interest. In the past, IDEMA has been expected to carry out such widely divergent and conflicting objectives as assuring high prices for producers and low prices for con- sumers, holding price increases over the course of the year to the target rate of inflation, making private storage profitable, and supporting itself financially. Its function is now being redefined to emphasize those goals that are consistent. If IDEMA S primary goal is to support producer prices, this goal might be more effi- ciently met through import adjustments alone rather than through the elaborate additional purchasing program now being used; if the aim is to improve the distri- bution of income through price support-which may not be an efficient instrument for this purpose-such an effect might be evaluated, explicitly recognized, and fi- nanced separately by the government. IDEMA is in the process of reforming its policies so that for the medium term, the private sector will increase its participation in storage and marketing and the institu- tion's financial burden will be reduced. In this connection, policies regarding price increases between harvest periods need to be changed if private storage is to become PRICE STABILIZATION 99 financially feasible on a larger scale and if IDEMA'S financial losses are to be con- trolled. First, IDEMA should not be expected to help control inflation through its pricing policies, not only because of the inefficiencies generated but also because inflation is for the most part a macroeconomic phenomenon. Second, IDEMA should retain its goal of allowing prices to rise at a rate sufficient to cover storage costs, but its estimates of these costs need to be reconsidered. It would appear that the costs, both financial and nonfinancial, of storage are significantly underesti- mated at present. Probably the best guide to true storage costs would be the average historical rates of price increase during periods of minimal outside intervention. These figures for different crops could be considered along with direct cost esti- mates to decide upon more realistic target rates of price increase. Notes 1. B. F. Massell, "Price Stabilization and Welfare," Quarterly Journal of Economics, vol 8 3 (May 1969), pp. 285-98. 2 See, for instance, Yesid Castro Forero, "Precios agricolas y su incidencia econ6mica," Revista nacionaldeagricultura, no. 861 (February 1983) 3. A more detailed discussion is contained in Alfredo Fuentes Hernandez, "Participaci6n de Colom- bia en los mercados de futuros," Revista naconalde agrcultura, no 863 (June 1983), pp. 97-106 4 Other agents, such as producers and consumers in domestic markets, could also reduce risks by hedging in international futures markets, to the extent that domestic prices are correlated with interna- tional prices, or by hedging in a domestic futures market 5. Holbrook Working, "Futures Trading and Hedging," American Economic Review, vol.43, no 3 (June 1953), pp 314-43 6. See Fuentes Hernindez, "Participaci6n de Colombia en los mercados de futuros," part 2, Revista nacionaldeagricultura, no. 864 (August 1983), pp. 109-20. 7. One of the necessary conditions for the smooth functioning of a futures market is that it be cheap to purchase and sell contracts, thus encouraging sufficient buying and selling to establish a legitimate market price at all times. This, in turn, means that brokers' commissions must be low. Unless market participation is quite heavy, brokers will not receive a competitive return on their labor, commissions will be increased, participation will decline further, and so on, until the market collapses. In addition, participation must be heavy enough to support a large number of brokers, so that collusion will be impossible and the operation of the market will be competitive 8. Exceptions are Almagrario and Almapopular, which are also such general facilities but are partially owned by IDEMA and consequently are considered semiofficial rather than private. Almacafe is also an exception since it is owned by FEDERACAFE and not by a private bank 9 According to a report of the Division de Regulaci6n Tecnica of the Ministry of Agriculture, "Ca- pacidad instalada de almacenamiento y secamiento en bodegas y silos 10. This share is 70 percent, rather than 80 percent, for palm oil, sisal, and wheat and 75 percent for export products 11. Obviously, for the crops eligible for only a 70 percent or a 75 percent loan, the 0.8 should be replaced accordingly. 12. 10.8 Pb Q(r. - r,)]/PmQ = 0.8 (Pb/PX)rm - r). Throughout this analysis, computations are based on the assumption that the interest rate is the rate paid on the full amount of the loan; that is, it is assumed that a rate of interest of 25 percent means the borrower pays $25 interest on a $100 loan. On some Colombian loans, the interest is taken out of the principal, so the interest rate is an understatement 100 AGRICULTURAL PRICE POLICY of the true rate. On a loan at 25 percent, for example, the borrower would receive only 575 (S 100- S25), leading to a true rate of 33 percent (25/75) Computation of interest in this way would not alter the qualitative results of the analysis, although it would change the numbers. If interest is computed in this way, for example, the subsidy as a fraction of market value should be the following: 0.8 (Pb/Pm)(rm - Ts)I(l - rjm)( - r,). 13. The cost in table SA-22 may be a slight underestimate for the following reason: the interest rate used as a market rate is the rate in the CAT market and is thus essentially a risk-free rate of interest. To the extent that BP loans are risky, the true market rate for them would be higher and the computed subsidy greater. Since these loans are well secured, however, it is doubtful that the risk premium on them would be very large in a free market 14. Alvaro Slva, R. Monsalvo, and Gabriel Montes, "El almacenamiento de productos agrope- cuarios en Colombia," Revista deplaneaci6ny desarrollo, vol 11, no. 3 (September-December 1979), p 100. The four largest firms, not including IDEMA, used 43.0 percent of the total. 15. Silva and others, "El almacenamiento," pp. 92-93. 16. Figures are for 1979, the most recent year for which information is readily available, source "Informe estadistico sobre la capacidad instalada en Colombia para almacenamiento de productos agri- colas" (Bogota OPSA, July 1980). 17. If goods are stored by the producer, not by a third party who purchases them for resale later, the purchase price is to be interpreted as the market price at the time the goods are placed in storage-that is, the opportunity cost. 18 John Nash, "Non-Financial Costs of Storage" (Washington, D.C.. World Bank, Latin America and the Caribbean Country Programs, Colombia Division, April 1984, processed). 19. Alvaro Castillo Niio, "Inestablidad de los precios agricolas" (Bogota, 198 3, processed) 6 Coffee Policy under Changing Prices and Technology WHILE COFFEE has been to a large extent responsible for Colombia's economic ad- vances during the 19 70s, the difficulties of the coffee sector in the first half of the 1980s have included the buildup of stocks and related financial and economic pol- icy dilemmas. Although overproduction may not necessarily continue indefinitely, developments in the early 1980s suggest the need for adjustments in domestic in- centives and in supporting infrastructure in order to encourage an increase in the movement of certain resources into noncoffee activities. This chapter contains a review of price and nonprice policy options being considered by the authorities to address these current issues while protecting the long-term performance of the cof- fee economy. Colombia's Coffee Economy Coffee has represented a significant share of Colombian GDP and an overwhelm- ing proportion of exports (see table 6-1). Coffee is grown on one million out of a total 4.5 million hectares of cultivated land. Nearly 50 percent of the coffee farms are less than 4 hectares. Coffee cultivation employs some 35 percent of the agricul- tural labor force on a full-time or part-time basis, while approximately 10 percent of the total population are considered to depend directly on production, processing, and primary marketing of coffee for their livelihood. Coffee has also been a signifi- cant source of government revenue. Between 1975 and 1979 the ad valorem ex- port tax on coffee contributed about 9 percent of the annual current revenue of the national government before declining to about 6.6 percent in 1980 and 3.4 percent in 1981, owing to a decline both in coffee prices and in the tax rate itself. During the 1970s coffee production expanded faster than the average rate of growth of agriculture, but no clear difference emerges once the 1980s are included in the analysis.' Coffee has accounted for an estimated fifth of the incremental 101 102 AGRICULTURAL PRICE POLICY Table 6-1. Trends in the Share of Coffee in GDP and Exports, 1970-82 (ratio of values in current prices) Agricultural Agricultural Year GDPa ExportSb GDPc exportSd 1970 6.5 63.5 23.9 82.2 1971 5.0 58.0 19.6 79.2 1972 5.3 49.7 20.2 73.5 1973 5 7 50.8 21.3 78.2 1974 4.2 43.9 160 73.0 1975 50 45.9 18 7 67.2 1976 7.5 55.4 27.2 78.3 1977 9 6 61.3 32 5 80.9 1978 7.7 65.9 29.1 85.4 1979 6 7 60.8 25 3 84.0 1980 5.8 59.8 25.2 78.6 1981 3.8 48.1 18 7 70.7 1982 3.6 50.5 18.1 76.0 Note. Only the direct contribution of coffee within the agricultural sector is included. a. Based on table SA-5, using DANE data on GDP at current market prices b. Based on table SA-8, using export figures in current U.S. dollars. c. From table SA-5; using DANEs categories, agriculture in this column is defined to include pergamino coffee (01), other agriculture (02). livestock (03), coffee processing (08), and sugar (12) d. From table SA-8; agricultural exports are defined to include coffee, livestock, bananas. sugar, flowers, cotton, tobacco, and rice. SourceS: DANE, Cuentas nacionales de Colombia. revision 3. 1970-82; Banco de la Republica; and FEDERACAFE. growth of the agricultural sector since 1970. The greater part of this contribution has stemmed from improvements in yield, since (according to official statistics) there was no net increase in total land under cultivation during this period. The value of coffee exports also expanded in real terms at a rate slightly higher than the average for all agricultural exports. Coffee has represented an average of some 77 percent of agricultural exports since 1970; in the economy as a whole this share of coffee in the value of exports averaged about 55 percent. Coffee Technology Three main techniques are practiced in Colombia: traditional technology (pri- marily for tipica), shade technology, and sun technology. The latter two are used in cultivation of the caturra variety, which became commercially available in Colom- bia around the mid 19 70s and was increasingly adopted as coffee prices rose during the second half of the 19 70s. Yields vary significantly among the three techniques. A FEDERACAFE survey suggests that the shade and sun methods increase yields on the average by 2 and 2.8 3 times, respectively, over the traditional technology. In den- COFFEE POLICY 103 sity of planting, whereas traditional coffee plantings average 1,800 trees per hec- tare, the new caturra varieties average 4,000 trees per hectare under shade and 4,500 trees per hectare under sun technology.2 Under shade, the available light is reduced so that coffee trees are planted at a wider spacing, with fewer trees per hectare. Both the new shade plantings and the traditional plantings are in most instances grown under the shade of trees and banana plants, frequently a source of additional revenue. Another important difference between the technologies is that whereas traditional plantings are seldom fertilized, caturra coffee in the sun uses about 1,500 kilograms of fertilizer per hectare, and in the shade about 910 kilo- grams per hectare. The caturra plantings use more man-days per hectare than tradi- tional plantings. On a man-days per kilo of coffee basis, however, the sun technol- ogy uses slightly more and the shade technology slightly less labor than does traditional. Estimates of labor requirements given in table 6-2 mask considerable variations among farms that use the same techniques but differ in planting densities and are at different stages of the production cycle. One hectare of caturra grown in sun at a density of 5,000 trees per hectare, for example, may require as little as 96 man-days in the second year after planting, or as many as 402 man-days in its fourth year, when it reaches peak production. While the shift from traditional technology to the new would increase total labor requirements per hectare, the effect on total labor use may not be positive-and might even be marginally negative-when the higher yields of the new techniques in the face of an output constraint have been taken into account. The new technology is also likely to make labor use more seasonal, compounding the problems of a large migrant labor force. Each hectare converted from traditional technology to the new shade method increases labor use for culti- vation by only 3 1 man-days but increases labor use during the harvest periods by 60 man-days. These problems could be mitigated by diversification in the coffee- Table 6-2. Labor Use, by Type of Planting (man-days per hectare a year) Technology Caturra Caturra Labor Traditional shade sun Cultivation 34 65 83 Harvest 65 125 180 Total 99 190 263 Note. FEDERACAFE quotes different estimates of labor use, depending on the farm models considered The measures given here indicatc only a broad order of magnitude. In the face of a constraint on output, man-days per bag of coffee, which may decline somewhat with the new technologies (see text), might be a more appropriate unit of measurement Source: FEDFRACAFE. 104 AGRICULTURAL PRICE POLICY growing areas to crops that require labor when labor demand for coffee harvesting is low. More than a third of the coffee-growing areas are planted in the new varieties, about evenly divided between sun and shade technologies (see table 6-3). This area is estimated to produce less than two-thirds of total output. There is little informa- tion available regarding the type of farmer who has adopted the new technology, although field observations suggest that the larger farmers with better education and better access to capital more readily adopt the new techniques than do others. This is especially true of the new sun technology, which requires much more fertil- izer and other inputs than traditional technology. Since FEDERACAFE-subsidized loans have been available for conversion, however, many small producers also ap- pear to have switched. In assessing the long-term effect of this revolution in produc- tion technology, it would be useful to have the benefit of a detailed study of the farmers who have switched to the new varieties. Changing Fortunes of Coffee The effect of coffee on the national economy was particularly strong during 1976-80 as a result of price increases and technological advances. World coffee prices began to increase in 1976, peaked in 1977, remained high in 1978-79, and declined after 1980 (see table 6-4). The higher prices were mainly responsible for increased export values in 1967-77. By 1978 export volumes began to respond to the price boom and, combined with higher prices, produced record sales receipts in 1978-80. During 1981-83 the export volume declined from its 1978-80 levels Table 6-3. Coffee Production under Traditional and New Technologies, 1980/81-1981/82 Area Production, 1980/81a 1980/81 1981/82 Thousand Thousand Thousand 125-kilo- Technology bectares Percent hectares Percent gram bags Percent Traditional 665 8 66.0 651.2 64.0 3,381.2 43 2 Newb 343.7 34.0 362.5 36.0 4,442.1 56.8 Total 1,009.5 100.0 1,013.7 100.0 7,823.3 100.0 Note: The coffee year runs from October 1 through September 30. a. Detail for 1981/82 by technology is not readily available. b. Includes both sun and shade techniques. Source: FEDERACAFE, Informe del Gerente General of ILI Congreso Nacional de Cafi, Annex 2 (Bogota, 1982). COFFEE POLICY 105 Table 6-4. Trends in Coffee Exports, 1970-83 (1975 = 100) Year Value' Priceb Volume' 1970 63.9 80.3 79.6 1971 56 5 70.3 80.3 1972 68.3 85.1 78 9 1973 84.3 101.9 82 7 1974 85.6 101.3 84 5 1975 100.0 100.0 100.0 1976 144.6 188 0 76.9 1977 228.8 350.3 65 1 1978 268.2 242.8 110.5 1979 278.8 204 7 136.1 1980 312.1 230.0 135.8 1981 246.2 222 2 110.8 1982 237.6 219.2 108.4 1983 250.8 222.5 112.7 a Based on foreign exchange surrendered at the Central Bank. b. Derived from value and volume. c. Volume exported, based on declarations of exports. Source: FEDERACAFE. and stabilized at levels that were still higher than the 1970-75 averages. The 198 1- 8 3 price was significantly lower than the 1977-80 level but in real terms was com- parable to the 1970-75 level. The share of coffee in total exports rose in 1976-80 but subsequently fell to the average levels for 1970-75 (see table 6-1). Domestic production, at some 13 million sixty-kilogram bags of green coffee a year, was higher during 1980-83; the average was about 11.5 million during 1978-80.1 Exports, on the other hand, fell to about 9 million bags in 1980-83 from a peak of 11-12 million bags in 1979-80. Historically Colombia had main- tained a share of world stocks well below its share in exports. This pattern contin- ued until the crop year 1980/81, although between 1977/78 and 1980/81 the stock share reached levels broadly similar to the export share. In 1981/82 this pattern began to change, and, as shown in table 6-5, Colombia was estimated to have held more than 19 percent of world stocks at the beginning of 1983/84, a figure based on stock estimates derived from estimated production, domestic con- sumption, and exports; its export share in total world exports was 14 percent. On the basis of verified stock estimates of the International Coffee Organization (ico), however, Colombia's share in stocks changed to more than 21 percent on October 1, 1982, and 20 percent on October 1, 198 3. Policymakers realize that it would be advantageous for Colombia to supply its export commitment with a lower level of stocks, as long as the level is sufficient to ensure steady sales even in the face of unexpected upswings in demand. 106 AGRICULTURAL PRICE POLICY Table 6-5. Colombia's Changing Shares in World Exports and Stocks of Coffee, Selected Years (percent) Coffee year' Productionb Exports, Stocksa 1961/62' 10 8 14.2 n.a. 1964/65 15.0 14.0 3.8 1970/71 13.4 11 9 9.4 1976/77 15.6 10.1 6.9 1977/78 15.2 14.9 13 0 1978/79 14.8 18.0 17.5 1979/80 15.4 19.1 15.6 1980/81 16.0 15 2 11.6 1981/82 13.2 14.2 18.3 1982/83 15.8 14.1 18.9 1983/84' 149 14.2 194 n.a Not available a. The coffee year runs from October 1 through September 30. b USDA, Foreign Agriculture Circular. Coffee; and FEDERACAFE. c. Based on ico data concerning exports to all destinations. d. Initial stocks; the ico estimate, based on verified stocks, is more than 21 percent. e Refers to calendar year 1962. f. Preliminary estimate. Sources: FEDERACAFE, the International Coffee Organization (ico), and the U.S. Department of Agri- culture (USDA), The Federation of Coffee Growers For more than fifty years the National Federation of Coffee Growers (FEDERA- CAFE), a private associatton of coffee producers, has been charged by the govern- ment with administering coffee policy and has received remuneration from the latter for its services. The federation's most important task has been to support incomes of coffee growers by facilitating the marketing of coffee at a guaranteed minimum price. With this policy it has acted as a regulator of domestic coffee prices in the face of wide variations in international coffee prices. The federation's presence is strong in the coffee zone, reaching down to the village level through its many projects, some of which, such as health, education, and rural electrification, are not directly related to the production of coffee. It pro- vides rural infrastructure and technical assistance for the coffee sector, controls do- mestic and export marketing, and advises the government, working with it to deter- mine domestic prices and taxes on coffee. The federation manages the National Coffee Fund (NcF) and has considerable freedom in managing the coffee economy and advising the government on policies in this subsector. The NCF was originally created, in 1940, in anticipation of the need to finance surplus stocks that would COFFEE POLICY 107 arise from the international export quota arrangement under the International Cof- fee Agreement (ICA) introduced in that year. The functions of the fund have subse- juently increased in scope as coffee revenues accumulated and various investment activities such as shipping (through Flota Gran Colombiana), banking (Banco Cafe- tero), and so on, were undertaken. Recently the finances of the fund have come under considerable pressure as world coffee prices declined and Colombia's coffee stocks accumulated. This situation is reversing itself in 1985 as the rate of peso depreciation has been in excess of the internal price increases accorded to coffee producers (see below). Adjustments in the Coffee Economy A large part of the export price has been siphoned off through price interventions such that the domestic farmgate price offered by the federation ranged from 3 8 to 6 3 percent of the external price during 1970-8 3 (see table 6-6). Comparing the end years of the period, the farmer's share of the export price has hardly varied, al- Table 6-6. Price Stabilization in Coffee, 1970-83 Internal Internal price in World real price' real priceb relation to peso Year (1975 = 100) (1975 = 100) export price' 1970 124.8 1148 0.58 1971 102.6 98 3 0.59 1972 107.2 103 9 0 57 197 3 120.2 110 0 0 53 1974 105.8 1000 051 1975 100.0 100.0 050 1976 190.3 169.2 047 1977 263.8 162.9 0 38 1978 176.7 142.0 047 1979 154.1 113.2 0.43 1980 137.4 106.1 048 1981 97.4 90.5 0 63 1982 109.9 85.6 0 58 1983 104.8 83.3 0.59 a. From table 1-3 These indexes are based on quoted prices, not actual sales prices of Colombia. b. Purchase price of coffee produced to the federation's quality specifications, deflated by the con- sumer price index for workers, from DANE. c. This is the ratio of the nominal domestic price paid by the federation for pergamino to coffee farmers, divided by the nominal world price of green coffee, equivalent to 0 719 pergamino in weight, converted to pesos at the official exchange rate Source: FEDERACAFE. 108 AGRICULTURAL PRICE POLICY though since 1981 the trend has been downward. The real domestic price of coffee has almost always moved in the same direction as the real world price of coffee, but the absolute changes in the former have generally been about half those of the latter. The domestic price variation during the coffee boom has, nonetheless, had significant effects on substitutions of caturra, which by now have reached prime bearing age. Since that time, the number of new plantings has declined, but the substitution of the new varieties for the old has continued. The result has been a dramatic increase in production, causing Colombia's export capacity to exceed its sales quota under the ico by 2-3 million bags and obliging FEDERACAFE to accumu- late an all-time high stock level of some 12.2 million bags as of the end of Septem- ber 1984, exceeding annual export needs. While little new land is now being planted in coffee-only 2,110 hectares during the first semester of 1981/82-about 10,000 hectares of coffee were replanted during the first semester of 198 1/82. The federation is not at present making loans for new planting or replantings. Most, if not all, of the replanting replaces tradi- tional varieties with caturra. Each hectare thus replanted increases the long-run coffee yield: if this increase is about 18.7 sixty-kilogram bags, it would imply that replanting of 20,000 hectares a year could eventually increase production by 374,000 bags a year, adding to stocks.' If exportable production declined and stabi- lized at a level of about 11.5 million bags and the country's exports remained at the increased 198 3/84 level, on the other hand, the stock buildup could be stopped. Further buildup of stocks from present levels would incur costs with perhaps little benefit. FEDERACAFE has estimated storage costs at $44 a ton; at this rate it would cost about $40 million a year to hold the stocks of 12.2 million bags estimated to have accumulated by the end of 1983/84 (see table 6-7). In this situation, the Table 6-7. Production, Exports, and Stocks of Coffee, 1958/59-1983/84 (thousand sixty-kilograrn bags, unless otherwise indicated) Year-end stocks Coffee Total Year-end as percentage of year Production exports stocks exports 1958/59 7,442 6,431 114 1.8 1960/61 7,500 6,043 1,081 1.8 1964/65 8,547 5,743 3,589 62.5 1969/70 8,266 6,874 5,583 81.2 1974/75 7,981 7,542 2,400 31.8 1978/79 12,300 11,431 4,870 42.6 1979/80 11,848 11,540 3,450 29.9 1980/81 13,037 9,031 5,978 66.2 1981/82 12,893 8,990 8,289 92.2 1982/83 12,810 9,174 10,230 111.5 1983/84 13,464 9,966 12,175 122 2 Source. Table SA-23. COFFEE POLICY 109 financial requirements of purchasing coffee that cannot be exported have turned out to be a burden on the NCF. Circumstances under which such large stocks could become fully useful-such as severe crop damage in Brazil-seem somewhat unlikely, although there is consider- able speculation concerning the true level of estimated world stocks. Even at the peak of the boom that followed the mid 1970s frost, Colombia exported only 11.5 million sixty-kilogram bags. Alternatives to a reduction of stocks do not seem promising. Brazil has destroyed coffee in the past, but to do so may not be politi- cally feasible in Colombia. Increasing domestic consumption to any significant ex- tent does not seem possible. Illegal exports to ico members cannot be an official policy., On the other hand, crop disease has already begun to affect Colombian coffee production: a strain of coffee leaf rust, roya, which has attacked most coffee-pro- ducing countries in Central and South America, attacked a few farms in the Depart- ment of Caldas in 1983 and had spread to about seventy-five municipalities by the end of 1984, affecting 4.5 percent of the coffee-producing area. In the past the federation has spent considerable resources to prevent appearance of roya in Co- lombia and to study ways of preventing its spread, should it gain entry. In particu- lar, a new variety of coffee tree has been developed-variety Colombia-which is actually a mixture of several strains and supposedly resistant to roya. Seed for the new variety is now in commercial production, but replanting with it is still rather limited. Existing stocks of coffee are sufficient to maintain present levels of export, and even to make up for any reduction in national production because of roya- which could reduce production by as much as 15-20 percent, depending on cli- matic conditions and the effectiveness of efforts to control it-while replanting with disease-resistant varieties is taking place. Since there is now a campaign to keep the disease under control, the final effect on production is yet to be determined. One of the main reasons for the recent stock buildup was the sharp increase in production following record prices and the adoption of new technology during the second half of the 1970s. The stock buildup is also a result of the country's partici- pation in the ico strategy of holding back sales in order to maintain a certain level of prices. It should also be noted that an ico exporter's quota is determined in part by that country's share in total stocks held by all exporting countries. A basic ico quota is determined for ico countries that export more than 400,000 sixty-kilogram bags. In 1983/84 this quotatotaled53.633 million bags, which was apportioned among exporting countries in this category. (A separate quota, which in 198 3/84 totaled 2.567 million bags, is established for small exporters.) About 70 percent of the total basic quota is assigned among the members on the basis of their past exports to ico members, while the remaining 30 percent is in theory apportioned according to their shares in total stocks held by all exporting nations. Thus, in theory, an in- crease of 1 million bags in Colombia's stock level, holding other countries' stocks constant, could imply an increase of 200,000-300,000 bags in Colombia's export quota.' Such a benefit would be unlikely in practice, because, among other reasons, it is not realistic to assume that there would be no changes in the stocks of other 110 AGRICULTURAL PRICE POLICY countries. In any event, even if such a benefit were to be realized, it should still be compared with the cost of a stock buildup and the financial and economic costs of the additional production that is not sold. Coffee Price Policy FEDERACAFE and the government jointly set a buying price for a certain quality of coffee-a weighted average of Col$ 15,499 per 125-kilogram bag of pergamino in 1984-at which FEDERACAFE will buy all the coffee of the prescribed quality offered, which it will then export, sell domestically, or store. Establishment of this guaran- teed price is perhaps the most important price policy in coffee. Private exporters can and do buy coffee from farmers primarily for export, but in so doing they must pay an ad valorem tax on the value of exports and a retention coffee quota and must sell to the government an amount of pasilla-low-grade coffee-and ripio (waste) parchment, all of which are determined by the government in consultation with the federation.' The retention quota and the pasilla and ripio "taxes" are generally paid in kind, the exporters physically turning over part of the coffee to the federation as tax. Changes in the federation's guaranteed price often require chang- ing one of the taxes if private exporters' incentives are not to be affected, and the retention quota has been the variable most frequently changed for this purpose. This quota was changed thirteen times during the period 1979-82, while there were only three changes in the ad valorem tax. The quota has ranged from a high of 85 percent in June 1976 to a low of 15 percent in 1980-81 and stood at a weighted average of 66.1 percent during 1984. Price Policy and Exports Given the export receipts that an exporter must deposit at the Central Bank (on the basis of their export volumes and average prices assumed for the period by the Central Bank) and given the domestic taxes on coffee, there would be a maximum price an exporter would be willing to pay the coffee farmer. The amount that private exporters would actually buy would then depend on the way their offer price compared with the minimum purchase price guaranteed to farmers by FEDERACAFE. Thus both the price interventions of the federation in the domestic market and the taxes it levies determine the volumes that private exporters find it profitable to purchase and export. Actual exports are also carefully monitored through licenses, provided by INCOMEX, which specify the buyer and destination in conformity with the arrangements negotiated under the ICA. The coffee must also be certified by the federation, weighed on the docks, and cleared of all the taxes and payments before shipping. This close supervision ensures that all actual sales by the exporters are registered, the federation supplements these quantities by its own sales in order to meet Colombia's ico export quota and non-ico demand. COFFEE POLICY III To illustrate the effect of prices and taxes on exporters' margins, we begin with the exporters' obligation to the Central Bank. In mid 1983 an estimated $ 191 had to be turned in at the Bank-reintegro minimo, "the minimum surrender," the level of which was a weighted average of $205.95 during 1984-by the exporter for the sale of a seventy-kilogram bag of excelso (export quality) coffee. This would be based on a New York price of S204 per seventy-kilogram bag, less an estimated $13 a bag for transfer cost from the Colombian port to New York-that is, S1.32 a pound at New York. Once the reintegro has been fixed, the exporter bears the risk of external price variations. In this example it is assumed that the reintegro corresponds to the actual world price received by the exporters. A 9 percent ad valorem export tax- reduced to 6.5 percent in 1984-meant that the exporter would receive in pesos the equivalent of $174 per seventy-kilogram bag-that is, $191 less 9 percent, or $17-which would be about ColS 13,050 per seventy kilograms at that time. Sub- tracting an estimated Col$700 per seventy kilograms for the internal cost of con- verting pergamino into excelso, a net receipt of Col$ 12,350 per seventy kilograms for excelso is obtained for the exporter. For every seventy kilograms of excelso, ninety kilograms of pergamino is required, and 40 percent-66.1 percent in 1984-of pergamino purchases, or thirty kilograms, had to be turned over to the federation as the retention tax. Thus, with Col$ 12,350, the exporter must buy 126 kilograms of pergamino, implying an offer price of Col$98 a kilogram. In compari- son the federation's guaranteed price at mid 1983 was Col5 102 a kilogram. This example is not necessarily indicative of actual price comparisons at present. The ColS700 "internal cost" must include the exporter's profit margin. Since most of his other costs are fixed by government tax rates, this profit margin can be squeezed by a decline in the world price of coffee below $204, assuming no change in the reintegro, or by an increase in the federation's guaranteed purchase price, which would increase the competitive farmgate price, or by an increase in tax rates. Any of these events can decrease, sometimes drastically, the percentage of produc- tion purchased by private exporters. On the other hand, if the world price rises while the reintegro remains constant and the peso depreciates, exporters' profits increase. In practice, the reintegro is adjusted to follow the world price movements with some lag, so in periods of rising world prices exporters' profits increase, and in periods of declining world prices their profits decline. Price Policy and Revenues Variations in these prices and taxes during 1975-84 are set forth in table 6-8. The minimum surrender has followed the international price, with some lag. The federation's guaranteed price in real terms-which has been well below interna- tional prices-has followed the direction of the latter with a lag, although the spread between the two prices has varied in the course of time. The ad valorem tax has been reduced steadily, and the retention quota has been the main instrument of tax policy; pasilla and ripio taxes are minor. Total taxes have been as high as 60 percent 112 AGRICULTURAL PRICE POLICY Table 6-8. Trends in Prices and Taxes on Coffee, 1975-84 (percent, unless indicated otherwise) Federation's Reintegro guaranteed minimo price Coffee Private (US. dollars (pesos Ad taxes + exports + per 70 per 125 valorem Retention value of total Year kilograms)a kilograms)b tax' quota' production exports 1975 107.87 2,730 19.0 32.8 34.3 60o0 1976 208.16 5,532 18.0 33.8 41.3 79(0 1977 366.08 7,179 17.0 30 0 48 6 70,0 1978 275.15 7,300 160 30.0 59.6 36 0 1979 234.14 7,270 16.0 50.4 60.0 23 0 1980 255.06 8.663 15 9 51 8 50.4 2.0 1981 187 60 9,453 12.7 20.8 194 38 0 1982 208.75 11,171 11.3 37.0 28.0 43 4 1983 193.74 13,010 8.7 40.6 26 3 42 7 1984' 204.50 14,439 6.5 60.0 n.a. n a. n.a. Not available. Note. These prices, taxes, and the retention quota are changed several times during the ycar: the figures given are yearly averages. a. Yearly averages, weighted by export sales in different months. b. Yearly averages, weighted by the federation's purchases, c January-March estimates. Source FEDERACAFE. of the value of production during times of high coffee prices. At such times of high taxes, the share of private exporters in total exports tends to fall, the federation undertaking the bulk of the exports. The principal recipient of the coffee taxes is the NCF, since it receives the entire retention tax and a part-36 percent at present-of the ad valorem tax.8 In 1974, the Ncv received 63 percent of total coffee taxes, estimated at Col$6,064 million, while in 1982 it received 84 percent out of Col$34,143 million. The national gov- ernment receives a good part of the ad valorem tax-56 percent today-while de- partmental committees of the federation receive about 8 percent. NCF Finances Following is an illustrative and partial calculation of inflows and outflows of NCF funds for 1982/83, focused on only those annual receipts and expenditures associ- ated with annual coffee purchases and sales by the federation. The federation's receipts come from two sources-its own sales of coffee and the share of the Na- tional Coffee Fund from ad valorem taxes on all coffee exported, both privately and by the federation. In 1982/8 3, the federation is estimated to have exported 5.1 COFFEE POLICY 113 million sixty-kilogram bags of excelso coffee and surrendered, at $ 168.57 per sixty kilograms, a total of $859.7 million to the Central Bank. Deducting an estimated 8.8 percent for 1982/83 ad valorem tax and using an average exchange rate of Col$74.4, a value of ColS58,334 million is obtained as revenue from coffee sales. The receipts of the NCF from ad valorem taxes on all coffee exports can be estimated as follows: total exports were about 9.2 million sixty-kilogram bags, with total rein- tegro of $1,541 million, or Col$ 114,650.4 million. The Coffee Fund receives 3.2 percent of this, or Col$ 3,669 million. Total receipts of the federation from both sources would then be about Col$62,003 million. In 1982/83, the federation pur- chased about 4.5 million bags of 125-kilogram pergamino at an average price of Col$ 12,464 per 125 kilograms, implying an outlay of about Col$55,702 million. These measurements are, to be sure, rough and partial. On the receipt side the value of domestic sales and on the cost side the transport, packaging, and storage expenses have been ignored.' It has also been assumed that all FEDERACAFE exports are to ico members. Receipts and costs of all other operations of the NCF and those from earlier debts and investments have likewise not been considered. On an annual operating basis and excluding these factors, a broadly break-even situation is indi- cated for 1982/83. Technology and Yields To resolve the stock-accumulation problem, one approach might be to discour- age replacement of traditional plantings with caturra, a step that could eliminate production increases-3 74,000 bags a year at current rates of substitution in the example given earlier. The traditional planters are generally the poorer coffee growers, however, and should perhaps not be denied the benefit of using techno- logical advances to improve their incomes from coffee. Furthermore, the traditional method is also a relatively high-cost way-per unit of harvested coffee-of produc- ing coffee under many circumstances, as indicated by the rapid adoption of the new technologies. Recognizing these conflicting interests, the federation is employing other measures to discourage further increases in production. Stumping Another long-term strategy could be to encourage the stumping, or pruning, of all caturra trees every six years, the time at which their production begins to de- cline.'o This puts them out of production for about fourteen to eighteen months, after which time they begin to bear again. Their growth then follows the pattern of a newly planted tree, beginning to slow down after another cycle of five to seven years, at which time they could be stumped again. If incentives, possibly in the form of low-cost loans to cover the direct costs of pruning, were provided to farmers to stump systematically, this would reduce annual production. One plan 114 AGRICULTURAL PRICE POLICY under discussion by the authorities was to seek the stumping of 80,000 hectares of caturra trees in rotation, so that at any given time, 80,000 hectares of coffee land would be out of production, implying production forgone of 2.46 million sixty- kilogram bags a year, at least in the initial years of the program. Eventually, how- ever, a tree produces more in its third and fourth years after stumping than it would without. Thus, the increased productivity of previously stumped trees might com- pensate for the production forgone from the currently stumped trees, unless they were stumped again. FEDERACAFE indicates that stumping is good husbandry and essential to protect long-term yields, and that farmers might prune even without special incentives. Many producers do indeed already follow this practice, although not on a scale that reduces output significantly in the short term; the federation's incentives would at best expedite the process. Loans from the Fondo Financiero Agropecuario (FFAP, the agricultural fund) have been available for pruning under the Ley 5a program, and a growing number of FFAP loans for coffee renovation have been given for renovating caturra by pruning rather than for replanting traditional holdings with more of the same. While the resources of the FFAP appeared sufficient to provide loans for the existing demand for stumping, additional funds would be necessary if a larger program were to be envisaged and implemented. Planting of Shade Trees An additional scheme could encourage the planting of shade trees among the pruned coffee trees, thereby converting plots from the sun technology to the shade technology. The shade method uses less fertilizer and produces a smaller yield- 12.59 cargas per hectare, rather than 16.84, or 531 kilograms per hectare less-but a better quality of coffee. Since it is proposed to prune 80,000 hectares of caturra, on half of which sun technology is probably practiced, the conversion of some 40,000 hectares from sun technology to shade could decrease output by about 354,000 sixty-kilogram bags. One possible way of achieving this aim is to require a farmer to convert his plot from sun technology to shade in order to be eligible for subsidized loans for pruning. Incentives could also be offered for conversion to shade without linking it to the incentives for pruning. Some farmers are already adopting shade technology because it is less fertilizer-intensive and therefore re- quires less working capital." Stronger incentives could be expected to accelerate this trend. Coffee Diversification The continuing coffee diversification program, implemented through the NCF credit program and free technical advice of the Programa de Diversificaci6n y Desa- rrollo de Zonas Cafeteras (PRODESARROLLO, a part Of FEDERACAFE) and through FFAP- COFFEE POLICY 115 subsidized loans under the Ley 5a program, is seen as both an alternative and a complement to the objective of stabilizing coffee production. The interest rate charge on Ley 5a loans was about 23 percent during 1983/84, whereas NCF loans were available at 2 1 percent and the current inflation rate is about 18-20 percent. A farmer typically obtains some credit under the Ley 5' program and finances the remainder of his diversification plans with an NCF loan. From the second half of 1980 through the first half of 1982, 20,684 hectares were financed for diversifica- tion-partly for eradication of coffee, but mainly for renovating or expanding the acreage planted in other crops in coffee-growing areas. Of the 20,684 hectares af- fected, about 3,225 hectares involved eradication of coffee. Social Benefits versus Private Benefits In evaluating the possible social benefits of coffee substitution, it may be assumed in a first illustrative example that additional production of coffee above current demand levels has zero marginal social benefit at present, since every extra bag produced will presumably be stored indefinitely. This assumption should be quali- fied in a more detailed analysis.12 At the margin, the annual net social benefit of substituting another crop for coffee can be considered to equal the net present value of the variable cost of coffee production and maintenance not incurred, plus the costs of storage avoided, plus the net value of the alternative crop, minus the cost of the eradication of coffee, and minus other real costs of the efforts that make diversi- fication possible." BENEFITS. The estimates are as follows: The variable cost of producing coffee was estimated to average around $1,487 a hectare." The coffee production forgone at 0.77 ton per hectare of excelso coffee, at an annual storage cost of $44 a ton would imply a stream of storage costs avoided of about $34 a hectare initially and increas- ing by $ 34 a hectare each year. This gives an annualized present value of storage costs avoided of $523.15 Estimates for the net value-that is, gross value less pro- duction costs-of alternative crops vary a great deal by crop, but one relatively popular alternative is cocoa, with an annualized net value of about $5 14 a hectare annually. Cocoa is chosen purely for illustrative purposes. Other alternatives should be evaluated, particularly since cocoa might face some of the same problems as coffee in international markets. Total annual benefits calculated in the case of cocoa are about S2,524 a hectare. cosTs. The estimated costs are as follows: Reduction of coffee was estimated to cost about $284 a hectare; this would be equivalent to a yearly cost of about $20 a hectare, using a real interest rate of 7 percent. The real opportunity cost, assumed to be 7 percent, of loans totaling about $14,576,708, or $1,020,370 a year for 3,225 hectares eradicated under the program-that is, $316 a hectare-evaluated using a real interest rate of 7 percent, can be considered a cost of carrying out this program. Furthermore, the cost of the free technical assistance might be consid- 116 AGRICULTURAL PRICE POLICY ered, on the grounds that in the initial years it is necessary to promote diversifica- tion; if it is assumed to be equal to its value in the private sector, this would cost about $291,534 a year, or $90 a hectare. Total costs were thus around $426 a hectare a year. Annual net benefits would be about $2,098 a hectare (2,524 minus 426), the benefit-cost ratio would be about 5.9, and the program would seem socially advan- tageous in the foregoing example. The most significant modification to this exam- ple would occur if the assumption were made that additional stocks have nonzero marginal value. As explained earlier, the strategic value of stocks in influencing exports cannot be ignored. In recent Ico negotiations the level of stocks may have affected export quotas in an ad hoc way, although a formal instrument-30 percent of export quota on the basis of a country's share in stocks-has existed since 1976. It is not easy to predict the effect of incremental stocks on export quotas in the future because it will depend on the behavior of other countries, and presumably many other factors. If it depended only on Colombia's behavior and if the formula relating stocks and exports were strictly enforced, the value of additional stocks can be readily seen (see footnote 6) lowering the benefit-cost ratio of diversification. Private Benefits and Costs The private producer sees the benefits and costs quite differently than does soci- ety. First, a producer sells the crop to the federation or to a private exporter. From the producer's viewpoint, these are revenues that would be lost if coffee were not produced and so should be deducted from the benefits of coffee reduction. Assum- ing a yield of 0.99 tons of pergamino per hectare (equivalent to 0.77 tons of green coffee), a guaranteed price of Col$ 12,100 per 125 kilograms of pergamino (the price in early 1983), and an average exchange rate of Col$74 to the dollar, these revenues forgone come to $1,295 a hectare. The producer does not incur storage costs, so from his point of view storage costs forgone of US$523 cannot be consid- ered a benefit. From the example given in note 13, the private cost of coffee pro- duction is slightly lower than the social cost, but the difference is only $34. Making these adjustments reduces the benefits to $672. If it is assumed that from the private point of view only half the real interest costs are relevant and since technical assis- tance is free, costs are reduced to $158 a hectare. The annual private net benefit from diversification would be only $514. Comparing the risks connected with mar- keting crops other than coffee to those connected with coffee may make the net benefit even lower." It should also be noted that many coffee growers have yields in excess of 0.99 tons per hectare; for them, the benefits from diversification would be correspondingly lower. From the federation's point of view, however, the bene- fits of diversification are great. The reason is that diversification relieves the federa- tion of buying coffee at the guaranteed price and again incurring further storage costs. COFFEE POLICY 117 Policy Conclusions Colombia's efficiency in capitalizing on favorable world prices of coffee has con- tributed enormously to the generation of Colombian exports, incomes, and em- ployment. More than a fifth of agricultural growth since 1970 can be attributed to this commodity. With the recent problem of stock buildup, however, the focus has shifted to finding ways and means of minimizing the social cost of supplying the country's export quota under the ICA and any additional sales. More work is needed to determine whether the stock buildup would continue under present poli- cies and under alternative assumptions about the life cycle of the existing trees, incentives for increased production, and projections of world demand. At the same time, policies to support the long-term performance of the caturra variety also need further study. While most of the needed adjustments in cropping patterns in the future may be expected to be initiated by the coffee producers, compliance with the ICA neverthe- less implies that domestic prices must be kept substantially below international lev- els to stabilize production. Taking into account transitory changes versus more en- during changes in international prices, price policy would also need to continue to avoid translating temporary increases in international prices into higher domestic prices, particularly since the latter are not easily reversible.17 If the demand for coffee should recover, reasonable additions to Colombia's export quota might still be met without increasing production. The federation might also provide-as it does now to varying degrees-nonprice encouragement to reduce the growth of coffee production and to develop com- modities other than coffee. In this chapter several options have been reviewed- discouragement of new technology in coffee production, stumping of coffee trees, planting of shade trees, and diversification of coffee. Stumping somewhat post- pones output and is considered essential for protecting long-term caturra yields. The obstacles to diversification are partly psychological-producers are reluctant to for- sake their tried-and-true coffee crops for other activities which to them are yet unproven. They are also partly based on the farmers' inadequate knowledge of cultivation techniques and markets for noncoffee products, such as fresh fruits and vegetables. Most important, the federation's assured market for coffee at a guaran- teed price makes other alternatives less attractive, particularly for the small farmer. While the social benefit-cost ratio of diversification is large, the ratio from the private farmer's point of view is significantly less attractive, given the alternative of an assured coffee price. FEDERACAFE is acutely aware of the emerging difficulties and is financing a diversi- fication program. The program has been less successful in inducing coffee reduction than in encouraging other crops. At the same time, since the decline in the real 118 AGRICULTURAL PRICE POLICY internal price of coffee has made coffee production only marginally profitable in relation to other activities in some areas, this might be a propitious time to encour- age some coffee substitution. The recent outbreak of the disease roya has made coffee production riskier, which might make diversification more attractive. More of the funds might be used to provide incentives for the replacement of old coffee trees, particularly in the medium-size and larger farms, by other crops, and to pro- mote marketing and processing, thereby creating greater price assurance for the noncoffee commodities. New loans or grants under the program might be linked to stumping and coffee substitution; alternatively, farmers might be reimbursed for the value of perhaps one year of coffee output postponed. The structure of incentives governing the distribution of funds to the regional committees of FEDERACAFE might also be reconsidered. At present, the share of the budget to each committee for improving its region is determined by the share of that region in coffee production. It may not be in the interest of an individual committee to give its full cooperation to a program such as coffee substitution in its region, because to do so would decrease the ensuing budget allocations made from the Coffee Fund. Finally, the federation's practice of providing subsidized fertilizer for coffee growers could be abandoned; such a process was begun in 1984. Even if production were to be stabilized at current levels, given the iCA, stocks might accumulate, unless a sizable increase in world coffee prices and world short- ages were to materialize. If not, reduction of production would seem to be needed, and the federation's objective, stated at the 1984 Coffee Congress, is to bring the output down gradually in the coming years. The options, such as pruning and diversification, for production reduction while sustaining long-term yields retain adequate flexibility for farmers to increase coffee yields and output when it is re- quired. Finally, greater attention might continue to be given to increasing total exports of coffee. Colombia has little control over world demand, however, and only limited control over its quota share under the ICA. The issue of increasing its exports to nonmember importing countries might be handled within an agreement among member exporters. Notes 1. See tables SA-5 through SA-7 for details of the participation of coffee in the economy. 2. FEDERACAFE, Ceso cafetero, 1980 (Bogoti, 1981). These are averages; a wide range of planting densities is practiced on various farms for all three types of production technology. 3. FEDERACAFE, Economia cafetera (Bogoti, various issues); see also table SA-2 3. 4. The figure of an incremental yield of 18.7 bags per hectare from replanting is based on the differ- ence in average yields for traditional and caturra varieties. It is slightly below the estimated incremental yield of 19.2 bags used in ico, Coffee in Colombia 1979/80 (London, September 1980), p. 44. Using the ico's estimate would reinforce the conclusions of the present analysis. 5. It was reported by the U.S. Foreign Agricultural Service, however, that 200,000 bags of contra- band coffee were smuggled out of Colombia in 1981-82. One practice is to secure a license to export is COFFEE POLICY 119 to a country that is not a member of the ico, then transship it to the United States through another producing ico member country. (This last step is necessary because ico consumer members are obliged to purchase only from ico producers.) Much of the contraband was discovered when it was noticed that sales to members of the ico from the country of transshipment were exceeding the quota assigned to that country. 6. The following is an illustrative example of the effect on an export quota if the stocks were in practice considered a factor. In 1983/84, 70 percent of the initial basic quota of 53.633 million bags- that is, 37,543 million bags-would have been distributed among the ico exporters on the basis of their share of past exports to ico members. If a 15 percent export share is assumed for Colombia, the resultant export quota, on this basis alone, would be 5.631 million bags. (The issue of the starting point for calculating a country's historical share of exports is a source of contention among ico exporters It would be in the best interest of the relatively new producers, for example, to compute the export shares on the basis of recent experience.) The remaining 30 percent of the basic quota, or 16.09 million bags, would be distributed according to the stock shares. The formula used to determine this variable part, QV, is QV = 0.3 S, - D., where S, is Colombia's share in the preceding year's stocks and D., is the projected ico import demand. A 19 percent stock share for Colombia, assuming world stocks of some 54 million bags and Colombian stocks of 10.2 million bags, would imply an additional export quota of 3.057 million bags. The total quota for Colombia in this example is 8.688 million bags, which turns out to be about 16 percent of the basic quota. An additional million bags of Colombian stocks can, other things being equal, increase the country's stock share to about 20.5 percent, which in principle increases the stock-based part of Colombia's export quota by some 241,000 bags, in addition to the original ico allocation of 5.631 million bags. 7. See appendix F for a fuller description of the various mechanisms by which coffee is taxed and subsidized. 8. The NCF receives a portion of the ad valorem tax equal to 3.2 percent of the total value of exports (reintegro), another 0.8 percent going to departmental committees of the federation, to be spent on projects in the coffee zone. Since the ad valorem tax rate changes from year to year, this 4 percent of total exports represents a variable fraction of the ad valorem tax receipts 9. The value of the retention tax that the NCF receives may also appear to have been ignored, but this factor is actually included. The retention tax is paid in kind-that is, in bags of coffee These bags are only of value in 1982/8 3 to the extent they are sold, not stored, and their sales value was included in the federation's total sales receipts Storage costs are not easy to estimate without knowing the pattern of stock accumulation. If stocks had accumulated at a steady rate throughout the year, beginning with 8.289 million bags and ending with 10.230 million bags, storage costs would have been about Col$1,811 million, Packaging and internal transport cost are estimated at ColS2,700 per ton, so for 5.1 million bags of FEDERACAFE'S purchases, these costs would be about ColS826 million 10. Stumping is done only on caturra, not on traditional plantings. Pruning increases the yield, be- cause caturra trees are so densely planted that they eventually compete for light, which reduces the yield. Traditional plantings, being much less dense, do not benefit from pruning. 11 Informe del Gerente General al XLI Congreso Nacional de Cafeteros, Anexo 2, November 1982, p. 29. 12. First, there is the possibility that some stocks will be needed in the future. Second, additional stocks can have some strategic or economic value because of the way in which ico quotas are established. 13. The following figures are based on estimates provided by FEDERACAFE. The assumption underly- ing the production figure is that fields using traditional, sun, and shade technologies are taken out of production in the same proportions as the proportions of total coffee area they represent and that land taken out of production would otherwise produce its current level of production. The variable cost of production, using social costs for inputs, was estimated by the ico in 1976 to be about Col$ 13,516 per hectare using traditional methods and CoIS59,568 using modern methods. Converted to dollars at the 1976 exchange rate and then converted to 1983 dollars, these figures equal $688 and 53,034, respec- tively. The area-weighted average variable cost of production is $1,487. The comparative figure at market prices is almost the same, at S1,453 Using 1982 estimated production cost figures from Caja 120 AGRICULTURAL PRICE POLICY Agraria-Col$66,707 with traditional techniques and Col$137,691 with modern techniques-the weighted average cost at market prices is $1,211 a hectare. 14. The incidence of the disease roya may increase costs significantly if it becomes sufficiently wide- spread to necessitate yearly fumigation in all growing regions This would add to estimated production costs about ColS 1 per coffee tree, or approximately S 50-55 a hectare for the caturra plantings or 525 a hectare for traditional plantings, at average planting densities. 15. This assumes that each year's production will be stored indefinitely. Thus, without diversifica- tion, storage costs would be $34 the first year, $68 the second-$34 for the first year's crop still in storage, plus $34 to store the new crop-S 102 the third, and so on. The net present value of such a stream is 57,473. Its annualized net present value is(7,473)(0.07) = S523, usinga real interest rate of 7 percent. 16. FEDERACAFE is trying to set up marketing programs for crops other than coffee to minimize the risk of diversification, It is clear, however, that the risk in these other crops is, and probably always will be, much greater than in coffee, which is tried and known 17. A continuing analysis has suggested that real producer prices during the period 1985-95 may have to be more than 10 percent lower than 1982 prices if a significant reduction in production is to be achieved; see Takamasa Akiyama, "Analysis of Coffee Policy Instruments and Supply Response in Co- lombia" (Washington, D.C.; World Bank, Economic Analysis and Projections Department, February 1985). Part Three PRODUCTION POLICY IN AGRICULTURE 7 Investment in Agriculture THIS CHAPTER deals with sector-level issues of the size, pattern, and efficiency of investment in agriculture and their possible effect on performance.' Most agricul- ture-related expenditures are made by the private sector, with strong producer asso- ciations proposing policy measures to the government and providing support ser- vices to their members. The main contribution of the government to agriculture has been in defining the policy framework. In agricultural investment, moreover, the government's contribution has also been significant in selected areas such as infrastructure development, including rural roads; irrigation and land development; research and extension; and provision of credit. In this chapter are developed some preliminary estimates of efforts of the private sector in agricultural investment. Severe data constraints, however, limit this analy- sis. Important issues connected with activities of the private sector concern the government's policy interventions and the production environment: the deteriora- tion in security in the countryside has contributed to the cautiousness of private initiatives in recent years. An analysis of the effect of the sociopolitical factors be- hind the investment climate, however, is beyond the scope of this chapter. The bulk of the discussion concerns the function of the public sector in agricultural investment. Differences in the definition and classification of expenditures by vari- ous agencies, and considerable overlap in their ascribed functions, have severely constrained this analysis as well. Public Sector Expenditures During the past decade, even as agriculture benefited from favorable interna- tional prices, public expenditures in the sector have, according to official data, de- creased steadily in real terms (see table 7-1).2 As a proportion of total allocations under the national budget, they are estimated to have fallen, which is in contrast to growth in total budgetary expenditures of 8.3 percent during the decade. These estimates of agricultural expenditures, however, exclude the substantial and per- 123 124 PRODUCTION POLICY Table 7-1. A Tentative Measure of the Direct Share of Agriculture in Public Expenditure, 19 70-82 Share of agriculture Agricultural expenditures in the total Year (million 1970 pesos) (percenty 1970 5.187 25.1 1971 4,762 21 2 1972 4,940 20.4 1973 5,211 21.1 1974 4,497 17.1 1975 2,978 11.7 1976 4,481 14.9 1977 4,862 14.8 1978 3,963 9.7 1979 3,748 8.8 1980 4,041 8.5 1981 3,788 7.6 1982" 3,975 7 1 a. Excluding indirect expenditures that benefit the sector. b Preliminary estimate Source. Table SA-24. odic financing by the government of debts incurred by Caja Agraria and IDEMA, which appear under budget allocations for public debt repayments.' They also ex- clude expenditures on such categories as rural roads and electrification which indi- rectly support agriculture. There may also have been some reclassification of func- tions at one time or another that make comparisons difficult. Nevertheless, a disaggregation of the principal agencies also reveals the declining trend, although its exact magnitude may be hard to establish. The bulk of public expenditures have been on commercial-that is, marketing- services and physical infrastructure (see table 7-2). This line is constituted in part by IDEMA S own resources generated from its marketing operations and from the credit it raises domestically, such as the bonos de prenda obtained from the Central Bank against goods in bonded warehouses. The Instituto Colombiano Agropecuarlo (ICA) also generates its own resources from fees charged on the quarantine facilities it operates, from the sale of agricultural products produced at research stations, and from the sale of other services. Transfers and debt services, on both internal and external credit, compose an- other large item in public sector expenditures. The magnitude of this item suggests an increase in ad hoc additional allocations made during the year to particular agen- cies for special projects-for example, to the Corporaci6n Financiera de Fomento Agropecuario y de Exportaciones (COFIAGRO) in one year and to IDEMA In another for constructing wholesale markets. The final disbursements made by the agencies INVESTMENT IN AGRICULTURE 125 Table 7-2. Principal Categories of Public Sector Expenditure in Agriculture, 1976-81 (percent) Category of expenditure 1976 1977 1978 1979 1980 1981 Commercial services and phys- ical facilities 50.8 46.1 33 2 23 3 33.8 21.6 Debt service and transfers 22 1 25.1 34.5 39.9 30.4 32.2 Recurrent capital 10.3 8 6 10 3 11 5 11.3 13.0 Research and extension 7.0 5.4 6.3 7 7 6.3 8 7 Design and operation of irriga- tion and drainage 1.1 3.0 2.3 3.6 3 1 4.5 Infrastructure 3.1 2 5 5.1 3 9 48 6 9 Agro-livestock credit 3.2 5 1 3.0 3.3 3.8 4.0 Other 2 4 4.2 5 3 6.8 6 5 9.1 Total 100.0 100.0 100.0 100.0 100.0 1000 Source: Estimates based on Ministry of Agriculture data. could thus be rather different from their initial allocations and portfolio of projects. While this could be a useful trend toward greater flexibility of investment in the public sector, one serious criticism has been the difficulty of tracing cross-transfers to ensure that appropriated funds are actually spent on those items for which they are designated and that they do not get lost in the recurrent expenditures of the implementing host agencies. It becomes difficult to analyze investment policy when categories of expenditure are not always comparable. The remaining public expen- ditures are evenly distributed among investments such as research, extension, drain- age and irrigation, and social services. Fiscal Problems Taking the expenditure pattern of the four main public agencies ascribed to the Ministry of Agriculture as an indicator of public investment in agriculture, the ra- tios of budgetary allocations to own resources of these agencies increased during the period 1976-82, although a significant reversal of this trend was estimated for 198 3 (see table 7-3).' These proportions could in fact be larger, given the fact that own resources do not necessarily mean revenues that are generated from the service or commercial operations of the agencies. In fact, "own resources" appears to be simply a convenient classification for funds other than direct budgetary allocations. Thus, for example, INcoRA's own resources are largely from loan recoveries. In another example, until 1975 ICA received tax credits of 2 percent directly from CAT, which were considered own resources. In 1976 these were eliminated, so ICA'S own resources dropped in that year. These examples indicate that the public agencies do not themselves generate suffi- cient revenue to meet their own financial needs and cannot compensate for de- 126 PRODUCTION POLICY Table 7-3. Sources of Financing in Agricultural Public Agencies, 19 76-83 (million pesos) Total Budget Own financing allocation resources Year (1) (2) (3) (2) + (3) 1976 3,055.5 1,466.0 1,589.5 0 92 1977 4,321.4 1,987.3 2,334.1 0.85 1978 3,932.7 2,628.7 1,304.0 2.01 1979 5.123.7 3,528.5 1,595.2 2.21 1980 6,924.1 5,280.6 1,643.5 3.21 1981 8,834.5 6,253 3 2,581.2 2.42 1982 9,417.6 7,089.6 2,328.0 3.05 19831 15,264.5 9,752.4 5,962.1 1.64 Note. ICA, INCORA, INDERENA, and HIMAT only; investment plus recurrent expenditures. a. Preliminary estimate. Sources: Ministry of Agriculture and World Bank estimates. creased budget allocations. While they may be limited by the nature of their opera- tions, the possibilities for generating revenue may not have been exhausted or considered seriously enough by the agencies. In the course of time their growth has imposed a strain on the national budget. Greater dependence on government trans- fers appears to be general throughout the public sector. These fiscal problems have been compounded by the creation of new public agencies, which perform some of the functions of the established institutions but which have not occasioned any cutbacks of the latter. The allocation of public expenditures in the agriculture sector by the main partic- ipating agencies is shown in table SA-2 5. The four main public agencies ascribed to the Ministry of Agriculture-ICA, INCORA, INDERENA, and HIMAT-received some 27 percent of the financing in 198 1, including both budget allocations and own re- sources of these agencies. The autonomous regional corporations, which are con- cerned only in part with agriculture, received about 31 percent of the total alloca- tions, while the rest of the entities-the Ministry of Agriculture itself and other public enterprises that are involved in agriculture to varying degrees-received about 42 percent. From 1970 through 1981 the ascribed entities maintained a somewhat decreased share of the budget, while the rest of the entities decreased their participation from 66 percent to 42 percent, the shortfall being explained by the expansion of the new regional corporations. So far, eleven autonomous regional corporations have been formed in response to regional and local interests. Among their functions are the development of local resources, land and water control, integrated agricultural services, technical assis- tance, power development, and conservation. The three corporations that at present have agriculture-related projects of any significance are Cauca Valley Cor- INVESTMENT IN AGRICULTURE 127 poration (cvc), Corpouraba, and Codechoco. cvc has irrigation projects and is in- volved in soil studies, property titling (under Ley 5a), and an integrated agricultural aid program. The other two corporations also implement integrated agricultural development programs financed in part by contributions from the national budget and directly under Dutch technical assistance. The work of these regional corpora- tions has duplicated to some extent the functions of INDERENA, established in 1968 to manage the development of natural resources in the country. In practice, it is difficult to reduce the budget or personnel or redefine the prescribed functions of bureaucracies once they have been established, and this problem may have caused some measure of parallel financing and duplication among the agencies. Programs such as the Plan Nacional de Alimentaci6n y Nutrici6n (PAN, the Na- tional Food and Nutrition Plan) and the Programa de Desarrollo Rural Integrado (DRI, the Integrated Rural Development Program) have attempted to coordinate the efforts of specialized government agencies by financing a concentration of their activities in selected areas, and these additional resources have helped to ease budget constraints of these agencies, ICA, INCORA, and INDERENA. This effect is not minimal, considering that in 198 3, the budget of DRI and PAN was some Col$ 7,500 million, of which about 30 percent was for direct investments in agriculture, while that of the four principal public sector agencies was Col$ 15,265 million. This example highlights the difficulty of properly evaluating the public sector's role in agriculture in the face of the various types of cross-transfers among agencies in and out of the sector. The Ministry of Agriculture directly executed only 0.5-0.7 percent of expendi- tures from investment and recurrent budget allocations during the period 1976-8 1. Specialized agencies and programs operate independent of the ministry. Thus, for instance, the regional corporations receive their investment funds by reallocation through the DNP; in 1982 the DRI and PAN programs received funds through the Ministry of Finance;' and the FFAP functions as an ascribed department of the Cen- tral Bank. Apart from these entities, there are the Servicio Nacional de Aprendizaje (SENA, the National Apprenticeship Service), which undertakes training, including that for agriculture; the Secretarias de Agricultura (agricultural secretariats at the departmental level); crop-specific private research and agricultural technology insti- tutes; various producers' federations; and others. The Budgetary Process and Problems An important factor that limits the flexibility of budget allocation is the practice of earmarking tax receipts for functional areas such as transport, health, and educa- tion. The proportion of national government revenues that are earmarked in- creased gradually from 23.6 percent in 1974 to 29.0 percent in 198 1.6 In 198 3, the level remained at 29 percent, which leaves the government discretionary control over only the remaining 71 percent. Of the discretionary 71 percent, the first claims are for recurrent expenditures and foreign debt service and only the balance 128 PRODUCTION POLICY could be considered for financing investment proposals submitted by individual ministries. In the budget only 2 3 percent of total resources was allocated to invest- ment in 1983. Attempts to free a larger portion of current income for discretionary allocation would require reduction in unnecessary earmarking and a continual review of nec- essary earmarking. According to 198 3 estimates, about 16 percent of current reve- nues of the government are earmarked directly for various purposes. About 15 percent of the remaining 84 percent that is not thus earmarked-that is, 13 per- cent-forms the basis for the so-called Situado Fiscal (sF), which goes to depart- ments for current expenditures in health and education. Together, the 16 percent and 13 percent make up about 29 percent, which is the total earmarked portion of current revenues. Recipient departments should be able to take on more responsibility for raising their own funds, thereby releasing at least a part of the revenues now earmarked through the sF. The primary source of the SF has, until recently, been the Special Exchange Account (SEA). A measure was taken by the administration in 1983 (which became effective in 1984) to reduce earmarking by diverting most of SEA- an estimated ColS50,000 million in 1983, or 16 percent of total national current income-for direct public investment; 75 percent of this amount went into the Public Investment Fund. Such a diversion correspondingly diminishes the base for calculating the sF. When applied to the 198 3 budget estimates, it would have meant that 15 percent of ColS50,000 million, or Col$7,500 million, was the reduction in earmarking, implying a net effect of reducing total earmarked revenues as a propor- tion of current income from 29 percent to 26 percent. In 1984, however, contrary to expectations, the anticipated reserves of the SEA for investment were not realized because of declining international reserves. The system of budgeting revenues and expenditures was described in the 198 3 Economic Report of the World Bank.0 By the National Budget Statute ("Normas orgAnicas del presupuesto general de la nacion," 197 3), the maximum allowable in- crease in forecast revenues above those of the preceding year is 10 percent and the maximum allowable decrease is 30 percent. This in principle limits the initial autho- rizations to similar levels, and projects must rely on additional authorizations dur- ing the year for continued implementation, although a variety of provisions make such additional authorizations a matter of routine. The Ministry of Finance presents monthly estimates of recurrent and capital ex- penditures and quarterly estimates of maximum obligations of capital expenditures to the Treasury. These statements-"Acuerdo de ordenaci6n de gastos y acuerdo de obligaciones"-are compared with the Treasury's estimate of expected revenues available for the next period. Here the accrual system of budgeting presents prob- lems in that the carryover of unspent resources and the continual reestimation of accrued expenditure obligations causes much confusion about the size of the actual current surplus or deficit. The Treasury must be careful not to overcommit itself beyond expected liquidity levels in the next period. In the complex administrative INVESTMENT IN AGRICULTURE 129 process of rescheduling and counterbalancing the receipts and expenditures, trying to ensure that the allocations do reflect the original programming objectives and targets becomes a difficult task. The timing of additional appropriations also causes problems. In the past, late appropriations or those that were not effected within the target period have con- tributed to underuse of capacity in the sectoral agencies, and financing has been out of phase with planned execution of targets. In some instances, funds that were meant for particular uses may have been used to stem other shortfalls in immediate requirements; this again has tended to blur the original targets. Shifting the whole budgeting process to a cash basis rather than an accrual basis might reduce confusion about the actual size of current and projected government deficits and clarify the extent of the actual financing gap arising from a particular budget proposal. This issue has been of concern to government practitioners and warrants more detailed study of budget allocation procedures. It might be possible, in the interim, to implement a system whereby the Treasury would establish lines of credit that would guarantee the supply of funds to agencies for the next period while it replenished its resources from inflows, in effect creating a grace period during which the resources available for the next period but one would become more evident. In any case, a more sophisticated system of matching revenues and expenditures is called for, to relieve the uncertainty of funding projects and programs. During the early 1960s, the national government adopted the practice of budget programming. The system worked well administratively and in the establishment of clear objectives. In time, however, the allocative process-for the agriculture sector as well as for the rest of the administration-has tended to develop "budget inertia," characterized by automatic increases of allocations to sustain established programs and agencies, often without sufficient reassessment of need, cost, or effi- ciency. Thus, for example, relatively new programs such as Cordoba 11 and Ca- queta II have had to compete for residual funds and have therefore taken nearly ten years for their completion, even though they have been designated as priority pro- grams. There is a need to review the budget allocation process to enable the execu- tion of priority programs and the realization of declared objectives and to reestab- lish a coherence in budget programming. This need is apparent, not just at the sectoral level, but also at the national level, given the inflexibilities of allocations in the national budget as a whole. The lack of proper program evaluation has made it impossible to establish rele- vant technical, social, and economic indicators by which to justify the desirability of increased resources for priority projects. Another problem has been the lack of standardization of what is included under "Investment expenditures" and "Recur- rent expenditures." Salaries and wages, for example, may be classified under either category, depending on the agency. This makes it difficult to budget adequately for the payment of this large item. The tendency has been to divert funds rather hap- hazardly from investment or recurrent allocations to pay for this more immediate 130 PRODUCTION POLICY claim, producing, for instance, situations in which researchers lack equipment and inspectors lack funds for gasoline to fuel their vehicles. Private Expenditures A comparison of public and private investment trends is given in table 7-4. These data, based on national accounts figures, imply that public expenditures in the sec- tor fell in real terms by some 2.2 percent a year during the period 1970-82. Private expenditures are estimated to have grown in real terms by about 4.0 percent a year and have compensated for shortfalls in public expenditures, causing the total to grow by about 3.2 percent a year. These are tentative measures, and considerable additional work would be useful to determine these trends with greater accuracy and establish the reasons for them. In 1982 public expenditures-that is, current expenditures plus investment-in the sector made up less than 10 percent of total agricultural expenditures. Private expenditures were thus much larger in the sector, and the bulk of the differences in total expenditures between these two sources-that in, between public and private sources-arose on account of the far larger volume of current private expenditures. Roughly 90 percent of total public and private expenditures in agriculture is esti- mated to belong to the category of consumption expenditures and 10 percent to that of investment (fixed capital formation) during the period 1980-82.7 According Table 7-4. Tentative Measures of Public and Private Expenditures in Agriculture, 1970-82 (million 1975 pesos) Percent public Year Total Public Private in total 1970 68,280 11,527 56,753 16.9 1971 70,761 10.915 59,689 15.6 1972 76,259 11,572 64,790 15.0 1973 86,257 12,741 73,714 14.5 1974 90,834 10,888 80,221 11.7 1975 87,929 7,124 80,805 8 1 1976 93,240 11,051 82,900 11.1 1977 109,192 11,922 98,165 10.1 1978 103,159 9,361 94,068 8.8 1979 100,214 9,017 91,434 8.8 1980 110,396 9,651 100,999 8.5 1981 93,873 9,054 84,996 9.5 1982 96,873 8,833 87,634 9.3 Note: Includes both current and investment expenditures. Sources: Ministry of Agriculture, DANE, and World Bank estimates. INVESTMENT IN AGRICULTURE 131 tO DANE data, the private sector is estimated to have accounted for some 93 percent of consumption expenditures. These proportions were representative of the 1970s, and growth in agriculture has been sustained by a steady rate of replenishment of capital investment by the private and public sectors in approximately equal propor- tions, without any marked increase in infrastructural investments. The private sector in Colombia appears to be well organized and, given appropri- ate encouragement, could be relied on to enhance government efforts to increase agricultural productivity. Agriculture has become increasingly commercialized, in part because of rapid outmigration of rural labor to the cities in search of jobs and amenities. Organizations of private producers undertake a wide range of activities in support of the production and marketing interests of their members. These activ- ities range from investments in research, extension and training, and provision of credit, marketing, and processing facilities, to sales-promotion activities abroad and lobbying the government. Major Organizations of Producers Eleven of the bigger producer organizations were studied. The largest of these is FEDERACAFE, the federation of coffee producers, which represents almost all the 300,000 coffee producers in Colombia (see chapter 5). The other organizations are smaller in scale but nonetheless command impressive memberships. The Federa- ci6n Nacional de Arroceros (FEDEARROz, the federation of rice producers) represents almost all the rice producers in Colombia; the Federaci6n Nacional de Cultivadores de Palma Africana (FEDEPALMA, the federation of oil palm growers) and the Asociaci6n Colombiana de Productores de Flores (ASOCOLFLOREs, the Colombian flower producers) represent about 85 percent of their growers; the Asociaci6n Co- 1ombiana de Productores de Semillas (ACOSEMILLA, seeds) about 55 percent, and so on. Membership in such organizations as the Federaci6n Colombiana de Produc- tores de Papa (FEDEPAPA, the potato growers, 7 percent of the total) tend to be lower, because this is a crop cultivated largely by smallholders in conjunction with wheat, barley, and other highland crops and probably some livestock. Eighty per- cent of the members of FEDEPAPA are small farmers having perhaps total farm sizes of five to ten hectares, of which one hectare is in potatoes. Seven of the eleven organizations have their own research programs, while the others rely on work being done by the icA; seven undertake extension and training for their members on subjects ranging from accounting and business management to improved production techniques, through short courses, news broadcasts, field days, and organized tours to farms on which innovations are being tried. Credit is arranged by only the coffee and cotton producers' organizations, although the rice, potato, seed, and palm oil associations are active in the sale of inputs to members, usually at slightly more favorable prices than if they were procured elsewhere. Joint processing is organized by the cotton federation, and agroindustries to some extent by FEDERACAFE, which has an active and well-integrated program to promote the 132 PRODUCTION POLICY diversification of coffee farmers into other cash crops and to implement such down- stream activities as agroprocessing, marketing, and retailing. FEDERACAFE also under- takes the provision of infrastructure far in excess of the normal purview of a pro- ducers' organization and in effect functions like a minigovernment among its coffee constituents, collecting coffee payments in kind; providing schools, nursery schools, hospitals, and roads; extending electrical power lines; and operating collect- ing centers, supermarkets, and banks. With such different scales of activity, the numbers of staff employed and annual operating budgets of the various federations of producers vary tremendously. Sources of finance typically include membership dues; levies on production-on kilograms of paddy milled, for example, or grains, or seed sold; and taxes on im- ports such as wheat and corn; as well as grants from the Ministry of Agriculture, such as that given to the Federaci6n Nacional de Cooperativas (FENALCO, the federa- tion of cooperatives), to support training activities among its members. Apart from membership dues, taxes and levies in support of these private organizations are enabled by state legislation, yet another indication of successful lobbying by the federations of producers and the close collaboration that is possible between the government and an organized private sector in fostering agricultural development. That the government recognizes this function of the private sector is evident in the fact that the individual federations and their apex organization, the Sociedad de Agricultores de Colombia (SAC), are often entrusted with reviewing and even draft- ing agricultural legislation in the process known as concertation. With such institu- tions and processes in place, it would appear that with the exception of the small farmers, the highly commercialized private sector is poised to complement govern- ment policies in the agricultural sector, given sufficient incentives for investment. Agricultural Borrowing A significant gap in present knowledge concerns the extent and nature of agricul- tural borrowing, information on which would shed light on the trends and prob- lems in agricultural expenditures. Most of the available information concerns insti- tutional borrowing. Institutional lending for private agricultural investment is channeled almost entirely through Caja Agraria, commercial banks, and financial corporations.' In 1981 they accounted for a total of Col75 billion in loans out- standing, provided by the agencies in the following proportions: Caja Agraria, 44 percent; commercial banks, 44 percent; and financial corporations, 12 percent. The relative importance of these institutions in agricultural lending remained essentially the same throughout the period 1977-8 1. As a percentage of their total portfolio of outstanding loans, agricultural loans assume varying degrees of importance; fig- ures for 1981 show that 76 percent of the outstanding loans made by Caja Agraria were to this sector, while corresponding shares of commercial banks and financial corporations were 16 percent and 12 percent, respectively. INVESTMENT IN AGRICULTURE 133 The most important single source of funds onlent for private investment in agri- culture is a fund instituted especially for this purpose-the Fondo Financiero Agro- pecuario (FFAP). During the period 1977-82, funds provided by the FFAP typically made up about 37 percent of the total agricultural loan portfolio. The FFAP, enabled by legislation (Ley 5') in 1973, was an attempt to coordinate agricultural credit activities under one umbrella. Funds are raised through obligatory investments im- posed by the Central Bank, which requires that all banks invest part of their re- serves in the bonds and issues of the FFAP. FFAP funds are rediscounted by the Cen- tral Bank to banks for agricultural loans; these banks should apply a certain percentage of their own resources-say, 20-30 percent-to match the remaining rediscounted funds-say, 70-80 percent-from the FFAP. These rediscount mar- gins-80 percent or so-as well as the interest rates to be charged on various catego- ries of loan are determined from time to time by the Central Bank. In June 198 3 interest rates on short-term loans-loans of up to one and a half years' duration- were 2 1 percent, and on medium- to long-term loans, interest rates varied from 15 percent (forestry), to 18 percent (smallholder cattle), 20 percent (larger cattle opera- tions), 21 percent (cash crops such as cocoa, fruit, and African palm, construction of wells and farmhouses, and purchases of farmland).9 Recent Government Efforts Recognizing the decline in public investment in agriculture, the present adminis- tration has been making a conspicuous attempt to increase allocations for 1984-86. Projections were available only for the four major public agencies of the sector, namely, the ICA, INcORA, INDERENA, and HIMAT. These agencies received propor- tional reductions in funds during the period 1970-83, a decrease of 2.5 percent annually, similar to that for the entire agricultural sector. For 1984-86, their pro- jected allocations envisage a dramatic rate of growth, but it remains to be seen how much of this increase will be realized. It has been suggested that priority to agriculture in the national budget could be augmented with better preparation of projects for domestic funding and for exter- nal funding, which would in turn require the government to make concomitant counterpart financing. A strong interest has been expressed in building up a pipeline of fundable projects and seeking such foreign participation that priority for budget allocations can also be assured. These moves are in line with the present administra- tion's policy of emphasizing agricultural development as a vehicle of economic growth. They are consistent, too, with the function traditionally assumed by the public sector of investing in programs with high economic and social returns, which would generate the infrastructure and the climate for private investment and reduce the risks in connection with such investments. 134 PRODUCTION POLICY Policy Conclusions It is widely recognized that the private sector in Colombian agriculture is dy- namic and responsive to incentives. Private investments have taken the lead in pro- moting agricultural growth, partly offsetting a decline in public sector investments during the last decade. The able stewardship of private organizations of producers is attested to in this book. The efforts of FEDERACAFE, FEDEARROz, and ASOCAlA in research and infrastructural development are examples. Recent declines in esti- mated private expenditure in real terms may have stemmed from, among other things, insecurity in the rural areas and declining incentives in the sector. Macro- economic policies to reverse agricultural disincentives can be expected to stimulate agricultural investment. The public sector has a critical function to perform in the provision of key infra- structural developments, such as rural roads and irrigation, and in supporting facli- ties that have more general application for a wider constituency than that of special- ized producer or interest groups, which would indirectly support private initiatives. Since 1970, however, total public expenditures in agriculture, particularly long- term investments, are estimated to have decreased, both in real terms and as a pro- portion of total national budget allocations. The bulk of public expenditures has been in support of IDEMA'S operations and for transfers and debt services, with investments in drainage and irrigation typically receiving 3-5 percent and research and extension 5-9 percent. A general difficulty that affects other sectors as well concerns the growing depen- dence of public agencies on budgetary allocations and on transfers; there is need for public agencies to generate more revenues to meet their own financial require- ments. The increase in the number of public agencies of various levels of govern- ment seems to have compounded the problem of transfers. A related issue is that of the inflexibility of budget allocations for investments, derived in part from the ear- marking of revenues for a variety of current expenditures. In this respect, recent efforts to free more resources for investment should be noted, although their effect has yet to be fully assessed. In agriculture, the financial problems of Caja Agraria, the agricultural credit institution, and IDEMA have further constrained the availabil- ity of resources for investment. Budgetary procedures appear to create considerable uncertainty regarding actual resources that will be available to the sector. The process spanning initial authoriza- tion, additional appropriations, and actual allocations is characterized by govern- ment practitioners as complex and cumbersome. Streamlining the procedures would seem to promise good returns, and a better system of matching revenues and expenditures may be necessary to relieve the uncertainty of project funding and reduce delays in execution. INVESTMENT IN AGRICULTURE 135 Notes I. Historic analyses of public expenditures in the Colombian context can be found in VictorJ. Elias, Government Expenditures on Agriculture in LItin America, International Food Policy Research Institute Research Report no. 23 (Washington. D C.. IFsu, May 1981); and Marcelo Selowsky, Who Benefits from Government Expenditure? A Case Study of Colombia (New York: Oxford University Press, 1979). 2 Expenditures in agriculture are defined here as those, including investment and current expendi- tures from budget allocation and own resources, by the Ministry of Agriculture, its subsidiary agencies- the Instituto Colombiano Agropecuario (ICA), the Instituto Nacional de los Recursos Naturales (INDE- RENA), the Instituto Colombiano de la Reforma Agraria (INCORA), and the Instituto de Hidrologia, Meteorologia, y Adecuaci6nes de Tierras (HIMAT)-and agricultural expenditures of autonomous re- gional corporations and other public enterprises. 3 In 1982, for example, government repayment to the Central Bank on behalf of Caja de Credito Agrario, Industrial, y Minero (Caja) and IDEMA for their past debts to the Central Bank were roughly CoIl 814 million and Col$ 577 million, respectively. 4 This pattern of increased relative dependence on government transfers appears to be general throughout the public sector. The principal increases have been in current transfers, both to decentral- ized agencies and to local government authorities. 5. As of October 1983, the DRI/PAN Program is managed by the Ministry of Agriculture but it still receives its funds through the Ministry of Finance. 6. World Bank data. 7. The predominance of current expenditures over investment expenditures is also demonstrated in Caja Agraria's allocations for agricultural sector loans, where only 5-9 percent of total lending for 1978 and subsequent years is envisaged for investment expenditures-machinery and equipment, agricultural land and infrastructure, houses and wells-while the rest, 9 1-95 percent, is for current expenditure for crop and livestock production, although the latter, in short-term loans, might really include capital ex- penditure as well. 8. INcORA and the Programa de Desarrollo Rural Integrado (DRi, the Integrated Rural Development Program) channel their loans through Caja Agraria Most of the allocations from the National Coffee Fund under the diversification program of PRODESARROLLo are handled through Banco Cafetero, and some through Caja Agraria. Commercial finance companies are an additional source of agricultural loans, but in 1980 they had only Col 328 million in outstanding loans, or 0.6 percent of the total. 9. On some FFAP loans the larger farmers pay an additional 1 percent interest, which goes to the Fondo de Asistencia Tecnica a Pequeios Agricultores y Ganaderos, a fund that subsidizes extension services to the smaller farmers. The charges for technical assistance pertain only to extension services linked to loans for major field crops, flowers, sugarcane, livestock breeding, and the installation of ma- chinery and wells, but not for coffee, reforestation, smallholder food crops, or milk production. 10. Published as World Bank, Colombia, pp. 77-82 8 Agricultural Technology, Input Policy, and Marketing THIS CHAPTER is devoted to the physical aspects of agricultural production, comple- menting the macroeconomic and pricing analysis in the first two parts, and extend- ing the discussion of investment trends in chapter 7. In particular, input policies and problems will be examined, particularly constraints to a more rapid adoption of modern inputs. Government interventions in input pricing have been few com- pared to those in many other countries, although real possibilities of lowering pro- duction costs through government policy have been significant. Modest reductions in cost can be achieved through improvement of import policy and mechanisms, while more substantial long-term benefits can be obtained through investment in high-priority areas. One set of factors reviewed at the outset that affect performance is technology and infrastructure, including research and extension, water development, and man- agement of natural resources. Government investment in this area has lagged, and efforts to increase the size and efficiency of investment are in order. Important inputs that influence production trends are labor, fertilizer, machinery, and seeds. Achievement of a reduction in the cost of these inputs is a critical issue, while in the case of credit the current levels of subsidy in relation to market rates and their efficiency are issues worth examining. Agricultural marketing affects almost all crops, although the nature of marketing constraints is not well understood. Trends and Problems in Input Adoption In a recent study of Colombian agriculture during the last decade the positive association between rates of growth of agricultural production and the use of im- proved production inputs such as fertilizer, certified seeds, and machinery has been indicated.' As shown in table 8-1, the annual growth rate in the use of these inputs diminished significantly after the mid 1970s, particularly from 1978, during which time the expansion in agricultural production also showed a distinct downturn-an 136 TECHNOLOGY, INPUT POLICY, AND MARKETING 137 association worthy of further investigation. The 1975-81 slowdown in fertilizer affected noncoffee agriculture, since during this period the use of fertilizer in coffee growing increased above the sectoral average. These trends have given rise to recent policy measures to improve the supply of inputs and to reduce their real prices in order to encourage greater use of inputs and thereby boost agricultural yields and production. There has been considerable variation in production performance and use of in- puts among various agricultural activities (see tables SA-26-28). Value added in agricultural production as presented in table 8-1 is apportioned almost equally among coffee production, other agricultural production, and animal production, in shares that have remained broadly stable since 1970.2 During the period 1970-82, the smallest expansion in output took place in animal production and the largest in coffee, other agricultural production falling in between. Within "other" produc- tion, rice, potatoes, and bananas grew at impressive rates. Important technological improvements have been made in these crops, as they have in coffee, and they are relatively input-intensive. On the other hand, domestic production of livestock, other food products such as yuca, and imported products such as wheat, corn, soya, and barley have not grown much, if at all. On the average these crops, most of which are produced by the small-farmer subsector, employ more traditional tech- nologies than the export-oriented products, and they use relatively small amounts of modern inputs. It should be noted, however, that estimates of domestic products suffer from severe data limitations. In itself, a slowdown in the rate of increase in the use of modern inputs may not necessarily be a cause for concern, since rates of use could taper off after the initial stages of rapid adoption. This pattern occurred in the case of rice in Colombia during the 1960s and coffee in the 1970s. For the sector as a whole, however, it cannot be said that the recent slowdown in use of inputs has followed high rates of adoption in earlier periods. According to World Bank estimates of fertilizer con- sumption per hectare of arable land, Colombia ranks about halfway among a group of countries and about 17 percent above the average for the middle-income coun- tries. About 74 percent of the nitrogen consumed in Colombia in 1981 was con- centrated on the production of coffee, rice, potatoes, and sugar.' Furthermore, while Colombia has achieved competitive levels of yield in crops such as coffee, rice, cotton, and sugarcane, considerable potential appears to remain for improve- ment of the yields of crops such as barley and corn (see table 8-2). The slowdown in the adoption of modern inputs is at least partly attributed to the decline in incentives for noncoffee agricultural production. Furthermore, input costs, already believed to be high in Colombia, have outgrown output prices since 1970 (see table 8-3). Increases in the prices of labor, machinery, and diesel fuel have been especially pronounced. A general conclusion supported by various informal estimates of farm budgets, and one that deserves further study, is that agricultural producers have in recent years been facing a growing profit squeeze. For the long term, increases in productivity may have justified some of the increases in the prices Table 8-1. Agricultural Production and Use of Inputs, 1970-82 Input or element of production 1970 1971 1975 1978 1981 1982 Annual growth rate 1970-82 1975-82 1978-82 Value added in agriculture (million 1975 pesos)' 86,488 88,059 108,490 123,624 136,285 134,483 3.8 3.1 2.2 Certified seed (thousand tons) 32,109 33,429 59,123 59,703 61,713 71,670 6.9 2.8 4.7 1971-81 1975-81 1978-81 Fertilizers (thousand tons) n.a. 479,014 565,754 759,449 661,147 n.a. 3.3 2.6 -4.5 Simple n.a. 55,706 87,760 115,497 86,260 n.a. 4.5 -0.3 -9.3 Urea n.a. 174,711 177,273 223,363 184,641 n.a. 0.6 0.7 -6.1 Compound n.a. 248,597 300,721 420,587 390,570 n.a. 4.6 4.5 -2.4 Machinery Number of units 22,713 23,469 24,187 27,871 29,693 n.a. 2.5 3.5 2.1 Average horsepower 58.4 62.8 62.5 64.5 62.0 n.a. - - Total horsepower (thousand) 1,327 1,474 1,514 1,798 1,841 n.a. 3.0 3.3 0.8 n.a. Not available. - Not applicable. a. DANE estimates of the combined value added in sectors 01, 02, 03, 08, and 12 (see tables SA-29 and SA-30), which, as noted in chapter 1, differ from data provided by the Central Bank. Sources: Oficina de Plancaniento del Sector Agropccuario (oPsA), DNP, and World Bank estimates. TECHNOLOGY, INPUT POLICY, AND MARKETING 139 Table 8-2. Comparison of Yields of Principal Crops, 1982 (kilograms per hectare) North and Interna- South Central tional Crop Colombia America America average Temporary Cereals 2,587 1,906 3,797 2,307 Rice 4,304 2,022 4,461 2,871 Wheat 1,605 1,610 2,382 2,009 Barley 1,667 1,108 2,836 2,068 Corn 1,401 1.942 5,994 3,465 Sorghum 2,611 2.966 3,534 1,447 Rootsandother 11,330 11,194 18,828 11,619 Potatoes 12,500 11,198 27,070 14,421 Cassava 10,392 11,594 6,030 8,885 Beans, dry 652 535 891 558 Oilseed Soybeans 2,000 1,643 2,160 1,772 Sesame 579 570 551 286 Seed cotton 1,739 1,681 1,803 1,286 Permanent Coffee 773 573 573 499 Cocoa 575 483 398 332 Sugarcane 87,705 60,594 58,265 58,682 Tobacco 1,419 1,327 2,106 1,446 Note: The area compared is heterogeneous; this is only a rough comparison of yields. Source: FAo, Production Yearbook, vol. 36 (1982). Table 8-3. Increases in the Prices of Agricultural Output and Inputs, 19 70-82 (annual percentage) Output or input 1970-82 1970-75 1975-82 1978-82 Output price 20.8 18.6 22.3 21.5 Input price' Labor 26.3 22.6 29.0 26.3 Machinery 22.7 25.6 20.7 25.0 Diesel fuel 26.4 15.0 35 2 26.0 Fertilizer 24.3 37.0 16.0 19.5 Seed 18.5 19.1 18.1 16 5 Insecticide 19.1 22.0 17 0 19.8 a. In 1982, for rice, beans, potatoes, corn, sorghum, barley, sesame, and wheat, labor was estimated to constitute 26 percent of the cost of production, purchased inputs 49 percent (roughly consistent with the data in table 8-5), "others" 8 percent, and indirect costs 17 percent. Sources. DANE. DNP, and OPSA data. 140 PRODUCTION POLICY of inputs in relation to output prices, although more recent trends seem to give cause for concern. Research, Technology, and Infrastructure By and large, basic research in the past appears to have been adequate. Except for a few crops such as sorghum and cotton, the combined efforts of plant breeders and pathologists have generally both enhanced genetic yield potential and improved resistance to disease. The efforts of agronomic research and extension have varied significantly for various crops, but on the average, they have declined during the last decade. An important area of neglect has been in the adaptation of cultural practices to location-specific ecological and socioeconomic circumstances. This has been especially true with respect to crops such as soybeans, wheat, corn, and cot- ton, world market prices of which have also been declining. Cultural practices hav- ing to do with time of seeding, land preparation, the use of water and fertilizer, and pest control urgently need to be improved and propagated in order to restore prof itability and competitiveness of crops in domestic and export markets. A reason for the relative inactivity in the generation of new technologies has been the decrease in funding of the Instituto Colombiano Agropecuario (ICA, the Colom- bian Agricultural Institute). In addition, the institute has been burdened to a grow- ing extent with regulatory functions that are not related to, and are sometimes at odds with, its research and extension objectives. These factors, as well as the loss of a number of first-rate scientists, have adversely affected the quality of ICA'S research but are now being redressed through increased budgetary allocations for research. The efforts of privately supported crop research institutes for export crops such as sugarcane, coffee, and bananas have helped to make up some of this shortfall, but their programs may not be comprehensive enough, and they do not include other crops of national importance, such as food crops grown largely by traditional farmers. Extension has also suffered from the fragmentation of the country's extension services and lack of policy orientation. There are some 2,000 private extensionists as well as numerous public or semipublic institutions responsible for extension, among other things, and these appear at times to overlap geographically and func- tionally. The record of the various extension services in Colombia is mixed and remains to be assessed. Use of certified seed is concentrated in commercial crops such as rice, cotton, and soya, with low participation in the case of others. Seed use per hectare increased only for rice and soya between 1971 and 1981. During the same period, total use of certified seeds in commercial crops increased at an annual rate of 7.9 percent, although since 1978 the rate has actually been declining. There were about fifty producers of certified seed, with a total production capacity of nearly 400,000 tons a year, in the early 1980s. During the 1970s the growth of this agroindustry permitted the country to stop its imports and become an exporter of seeds. TECHNOLOGY, INPUT POLICY, AND MARKETING 141 To turn now to infrastructure development, about 600,000 hectares, the equiva- lent of about 15 percent of the total cropped area, are estimated to have some sort of irrigation, drainage, or flood control available at present; 80 percent is under private schemes and the rest under government schemes. Another 120,000 hect- ares under government control is cultivable, but is not actually irrigated. The issue of irrigation and drainage development in Colombia concerns both the improve- ment of existing facilities and the provision of new ones by private agencies, with or without government support, and in some instances directly by public agencies. While private irrigation schemes-80 percent of which are located in the fertile Cauca Valley-are generally run efficiently and involve lower costs, most public irrigation districts are underused as a result of poor maintenance of infrastructure arid poor land leveling. In response the government has embarked on a phased rehabilitation scheme beginning with eight districts, or 76,000 hectares, in the first phase of the project; seven other districts are to be included in a second phase. On the provision of additional irrigation, care needs to be exercised in limiting investment to areas where water is indeed the constraint to higher yields and to obtaining two crops a year. For rice, yield levels are as high as can be reasonably expected. Yields for other large-scale field crops, such as sorghum, corn, and soy- beans, on the other hand, are low, primarily as a result of the fact that farmers continue to follow traditional farming practices-that is, using little or no improved seed and agrochemicals. A lack of water has not always been a binding constraint in the adoption of new technologies. Under these circumstances, it would be essential to restrict investment of scarce public resources to places where it can be clearly shown that water is limiting increased production, or where a transition to high- value crops or more intensive cultivation has taken place and high returns are plau- sible. On the other hand, public investment in agrohydraulic work for flood control and drainage could change areas that are now suitable only for extensive grazing into land that can be used for permanent crop farming. Such investment also seems cost-effective to the extent that it usually involves an outlay of only about 25 per- cent a hectare of that needed for irrigation. Various problems related to recovery of operation and maintenance costs in flood control and drainage districts are encoun- tered, however, because a national tradition in the operation of this type of district has yet to evolve. In addition technical problems are encountered in the drainage of flat, low-lying tropical areas. Investment in irrigation can be more easily justified where the marginal cost of completing projects already begun is low, as it is, for example, in the case of the World Bank's first irrigation rehabilitation project. A problem requiring particular attention is that of minimum district size. In the past, selection of public invest- ments has been guided more by political criteria than by agricultural or engineering criteria, bringing about the establishment of such small districts that diseconomies of scale for operation and maintenance were the result. Although the potential for expanding agricultural production on new land facili- tated by some form of water control is substantial in Colombia, much of it remains 142 PRODUCTION POLICY generally a more costly possibility than that of crop intensification on existing culti- vated areas and in many instances probably also the opening of new rainfed frontier land. Where new irrigation and water control schemes are warranted, the benefi- ciaries, the kind of agricultural development that is likely to take place, and the kind of cost recovery charges that are feasible need to be identified and examined. It is important to ensure that sound policies regarding water charges are adhered to. Meanwhile, private investments in irrigation and drainage work should continue to be supported when they are technically and economically justified, with credits and appropriate incentives, and research should be developed with a view to bringing about greater efficiency in the use of water through the development of improved land-leveling techniques, and, more generally, through the use of cost-reducing al- ternatives, such as puddling for rice cultivation. Renewable Natural Resources Colombia is endowed with vast expanses of heterogeneous forests that have been exploited at a growing rate in both tropical lowlands and cold highlands. During the past twenty years, forest resources have been reduced by an estimated 10.5 million hectares, corresponding to almost 20 percent of the approximately 55 million hec- tares of forest that existed in 1960. Deforestation, as a result of the establishment of new settlers and the traditional replacement of the forest cover and other natural vegetation with crops or pastures, has created serious problems of erosion and soil conservation. The effects of this erosion include more frequent and more serious flooding and reduced water flows during dry periods for hydropower generation and for water supply systems. There is evidence that the siltation of reservoirs is increasing. Protection and management of river basins and a rational use and man- agement of tropical forests are critical to the preservation of Colombia's abundant land and water resources and for ensuring a continuous supply of wood for domes- tic and export markets. The government has been aware of these issues for many years. In 1969 it set up the National Institute for Renewable Natural Resources and the Environment (INDERENA), which subsequently established a natural resources code listing the rec- ommendations and rules for managing the country's resources properly. But the task is enormous, and INDERENA has been hampered by shortages of funds and trained personnel. Nevertheless the initial steps, such as the identification of critical areas for conservation and reforestation programs in the areas of influence of hy- dropower projects and the development of the Upper Magdalena Pilot Watershed Management Project, are being taken. A greater capacity to execute programs for the conservation and rational use of renewable natural resources needs to be developed, and greater public awareness of the problems will clearly be required. INDERENA S institutional capabilities for policy implementation and project preparation need to be strengthened. In particular, TECHNOLOGY, INPUT POLICY, AND MARKETING 143 conservation education programs need to be stepped up. A program of research and baseline studies of critical areas is called for to enable INDERENA to evaluate policies and projects, funding for which might need to be sought externally. The ad hoc approach to forestry development needs to be replaced by a more systematic policy. In particular, the execution of pilot projects, such as the Upper Magdalena Pilot Watershed Management Project, and the follow-up of larger-scale projects might be placed within the framework of a national forest and natural resources development program. The forestry components of various rural devel- opment projects, despite their small size, could be geared to addressing problems of properly harvesting, using, and managing the tropical rain forests of Colombia, designing and administering "forest colonization" projects, and developing and managing large-scale watershed management projects. Inexpensive small-scale refor- estation activities could be stepped up, particularly under rural development and settlement projects such as the Integrated Rural Development Program (IRDP) and Caqueta. The potential for further large-scale industrial reforestation could also be investigated. Spontaneous settlement would be guided by the development of soil classification and of forest colonization projects, as opposed to traditional coloniza- tion, which is based entirely on agriculture, which is often inadequate for moist tropical zones. A research program analogous to the National Agricultural Re- search Plan could be developed on the basis of the National Forestry Research Plan (PLANIF) and implemented with external assistance, if it is needed. Rural Labor and Income Movements in the real wage show that between 1935 and 1964 the purchasing power of agricultural workers probably did not increase significantly, and its level was in fact 10 percent lower than that of unskilled construction workers.' In 1964 a far greater proportion of the poor were in the rural areas than at present. Some improvements in rural wages came about during the late 1960s, but it was only with the employment expansion of the 1970s that real agricultural wages began to rise significantly. Changes in survey methods between 1971 and 1976 may have overstated the true gains to some degree; nevertheless, between 1970 and 1976 real agricultural wages are estimated to have risen about 23 percent, and another 22 percent between 1976 and 1980. Consistent with the recent slowdown in growth of the agricultural sector, a small decline in real wages is reported to have occurred in 1981. Rapidly growing urban employment in the 1970s offered jobs to rural immigrants. At the same time, labor-intensive agricultural development was taking place in the coffee regions with the new plantings of the caturra variety, and cotton areas were competing for labor with clandestine drug plantations. Rapid expansion of exports had a strong salutary effect on labor demand and rural wages. These factors even led to seasonal and regional labor shortages, which in some instances 144 PRODUCTION POLICY also brought about increasing mechanization-the use of cotton harvesters in cot- ton areas, for example. Enhanced demand for agricultural labor and slower growth in its availability pro- vided much of the basis for the estimated growth in real wages-as an average for tropical and temperate regions in the country-for the second half of the 1970s. The rate of growth in the productivity of agricultural labor-value added per worker-during the period 1974-79 has been estimated at roughly 3.7 percent a year. Furthermore, beginning in 1973 real minimum wages also rose. Increases in the real agricultural wage during the 1970s directly or indirectly benefited most of the rural community, which includes the poorest segments of the Colombian popu- lation. Yet because agricultural and overall growth have slowed since 1978, real wages have probably stagnated according to some calculations on the basis of infor- mation from DANE. Nevertheless, as indicated in table 8-3, nominal wage increases in agriculture have outgrown increases in the prices of agricultural output, which has been noted by many observers as a main cause for a profit squeeze in the sector. In view of the foregoing, wage policy presents difficult decisions. The increases in rural real wages during the 1970s reflected the favorable consequences for the dis- tribution of income of rapid growth in Colombia. A problem facing economic management has been how to translate the slowdown in economic performance in the recent past into wage adjustments. As the economy attempts to catch up with the accumulated overvaluation in the nominal exchange rate, nominal wages ought not to adjust in line with movements in the exchange rate, since past changes in money wages have fully or more than fully accounted for inflation, whereas changes in the exchange rate have not. Wage negotiations for 1984 have caused a 28 percent increase in the minimum rural wage, well ahead of the expected infla- tion rate adjusted for increases in productivity. Fertilizer Pricing Aside from coffee, rice, potatoes, and sugar, which use fertilizer intensively, fer- tilizer use in Colombia remains only comparable to or below the average levels in Latin America, implying the scope for more intensive use of this input. More fertil- izer application is particularly cost-effective for small-scale farmers growing subsis- tence crops such as corn, beans, and cassava. In the case of crops such as potatoes, net gains could be realized from a more balanced use of fertilizer than is now prac- ticed. The use of fertilizer for these crops, for example, is estimated to be far from technically optimal levels, because commercially available generalized formulas are not always suitable to the large variety of ecoclimatic conditions found in Colombia or because of inadequate extension services. The increasing use of foliar analysis and more refined assessments of the need for fertilizer and the dissemination of this knowledge by the extension services could mean greater cost-effectiveness in the use of this input. TECHNOLOGY, INPUT POLICY, AND MARKETING 145 Fertilizer use has declined since 1978, and the DNP'S analysis identifies several contributing factors: rising relative prices of agricultural inputs, declining profit margins, and overall stagnation in production and acreage. From an economic viewpoint a suboptimal use of the input has been caused by high farmgate prices for fertilizers-reaching more than twice the world market levels, except perhaps for coffee, for which the input price is less than that for others. These prices are to a small extent the result of taxes and tariffs and more significant, of port charges and high domestic transport costs (see table 8-4). Fertilizer import and price policies have been under active surveillance and state control, and intervention has been frequent because of pressures from one group or another. There has not been a concerted and deliberate policy, however, with clearly spelled-out objectives and definition of the desirable levels of subsidies and taxes to be borne by the various parties involved-the industry, the agricultural producers, COLPUERTOs (the Colom- bian Port Authority), the truckers, and the distributors; further analysis of these issues is clearly in order. Credit Policy Virtually all farmers receive institutional credit that is subsidized-in relation to free-market rates-directly or indirectly by the government. The volume of institu- tional credit for which the farmer is eligible usually represents more than 50 percent of the farmer's costs of production. Several intermediaries, including commercial banks, are involved in the distribution of credit. Two sources of funds, however, the FFAP (Agricultural Finance Fund) and the Caja Agraria (Agricultural Develop- Table 8-4. Illustrative Calculation of the Price of Urea, 1983 (million pesos) Stage Price F.o.b (in bulk) 10,360.00 C i.f. 12,225.00 Taxes and duties 1,036.00 Standard port charges 1,393.00 Storage, cost of delays, and other 455.00 Financial and administrative charges 2,726.00 Ex-port cost (in bulk) 17,835.00 Bagging cost 1,400,00 Ex-port cost (bagged) 19,235.00 Estimated cost and transport 3,000.00 Cost at retail 22,235.00 Source: World Bank calculation on the basis of data from the private fertilizer company ABOCOL 146 PRODUCTION POLICY ment Bank), account for about 70 percent of the institutional funds lent to agricul- ture. Caja lends its own funds and some of FFAP s. The remaining FFAP funds are lent by commercial banks and other banks, such as Banco Ganadero and Banco Cafe- tero. The term of the loan and the interest rate charged to the farmer depend on the lending institution, the crop being financed, and the assets of the farmer. The Caja Agraria, for example, charges lower interest rates to farmers whose levels of assets are low. There does not seem to be a government policy on the level of the interest subsidy. When market interest rates were falling, for example, the nominal interest rate for the largest line of credit, the FFAP, did not change. The Caja Agraria low- ered its interest rates 5 to 7 percentage points in 1983 on loans for food crops. This move appears to have been prompted by a desire to stimulate food production, however, rather than by a desire to maintain a constant level of subsidy in the face of falling interest rates. Most farmers eligible to receive the subsidized credit have been able to do so in the past. The subsidy does not cause some eligible farmers to be deprived of the credit, as often happens when a good is subsidized. The eligibility criteria include a title to the land, and the volume of credit is based on the crop and the area planted. In 198 3, 5 percent of the FFAP funds that were committed to agricultural loans were not lent because of a lack of eligible farmers to demand the loans. The cause may have been bureaucratic delays in processing loans and in transferring lines of credit from one crop to another. In fact, there seems to be some diversion of money earmarked for agricultural credit into other activities. Therefore, the agricultural credit subsidy comes at the expense of the availability of credit in other sectors. The subsidy is not limited to specific crops and is available to small farmers. In a study conducted jointly by the Inter-American Institute for Agricultural Sci- ences (InCA) and the Ministry of Agriculture, average rates of interest on agricultural loans were found to be between 45 percent and 60 percent of market interest rates during the period 1976-80.' The credit subsidy during this period was equal to 2.2 percent of agricultural GDP. In 1982 the weighted average effective interest rate on Caja Agraria loans to farmers was 31 percent, while FFAP funds were lent at 24 percent.6 Market interest rates for unsubsidized loans in the formal sector were about 50 percent.' The benefit to farmers from government-subsidized loans may not simply be the difference between the government-subsidized rate and the mar- ket rate in the formal sector because of two other important considerations that work in opposite directions. On the one hand, government bureaucracies make the cost of transaction of government loans rather high for the farmer; on the other hand, most farmers have limited access to the formal market and would have to obtain credit in the informal sector, where interest rates are estimated to be 55 percent or higher. While data on loans given to farmers through the informal sector are not available, the proportion of loans made through the informal sector is thought to be low. At first glance, the effect of the agricultural credit subsidy on the composition of agricultural output or on the choice of farming techniques does not seem to be TECHNOLOGY, INPUT POLICY, AND MARKETING 147 distortionary, since credit is readily available for most purposes at roughly compara- ble interest rates. Some agricultural projects that may not be economically benefi- cial, however, will be undertaken because the private cost of borrowing is made less than the social cost by the subsidy. Insofar as these low-return projects are concen- trated in a given subsector of agriculture, the composition of agricultural output will be affected. The subsidized credit in agriculture is also a source of distortion for other sectors, since part of the cheap and subsidized credit in agriculture comes at the expense of reduced availability of credit in other sectors. This increases the market interest rate in other sectors to a level above the cost to the economy in the absence of subsidies. Projects in other sectors that may be socially beneficial at the social cost of borrowing are therefore not undertaken because the private cost of borrowing is higher than the social cost. The lowering of the credit subsidy to agriculture is a delicate issue, because of assumptions concerning the effect of such a measure on production and distribution. The effect of reducing the credit subsidy on agricultural production and farm incomes is likely to be modest. On the basis of an analysis of twelve major crops, including coffee, the interest costs of credit account for about 11 percent of the annual costs of production. Since the subsidized interest rate is about 55 percent of the market rate and the credit subsidy 45 percent, the subsidy accounts for about 9 percent of total production costs. Removing the subsidy would raise the production costs by less than 9 percent if, as expected, it led to lower market interest rates and farmers obtained credit from both sources. In comparison, fertilizer accounts for about 12.5 percent of annual production costs, while machinery rental accounts for about 14 percent (see table 8-5). It should be noted, however, that the measures in table 8-5 may understate the true credit costs and the difficulties faced by farmers in obtaining this input. Although the Caja figures are meant to be full interest costs, it is possible that the costs of noninstitutional credit are excluded. The calculation of the credit subsidy would still be adequate, however, since it affects only the institu- tional component. Much more needs to be known about agricultural credit and financing-both institutional and noninstitutional credit, the efficiency of existing arrangements, and their beneficiaries. There has been an increasing number of field observations about the solvency and liquidity problems of farmers, and the effective cost of credit, including admin- istrative and institutional costs-which are yet to be quantified-may be greater than the estimates mentioned here. The present comparisons nevertheless suggest that while the subsidy component of credit costs may not be insignificant, it may be an incentive for which other measures-depreciation of the exchange rate and re- duction in costs of other inputs (by a reduction in tariffs or increased efficiency in transportation or ports)-can reasonably be expected to be substituted. The financial situation of Caja Agraria, the largest bank in Colombia in geo- graphical coverage (863 branch offices and 460 input and supply stores) and num- ber of staff (14,500 employees), continues to be problematic. For equity and politi- cal reasons, Caja is required to provide a level of services far in excess of its resources 148 PRODUCTION POLICY Table 8-5. Input Costs as a Proportion of Total Costs of Production, 1982 (percent) Machinery Crop Credita Fertilizer rental Pesticidestb Seeds Wheat 6.89 11.68 24.01 4.28 11.23 Barley 6.62 10.63 26 57 6 24 758 Corn 8.03 6.10 11.37 0.59 4.70 Soybeans 7.58 10.99 14.34 8.58 9.56 Sorghum 7.41 20.24 14.46 649 2.57 Sesame 8.27 8.00 10.76 2.29 1 57 Beans 7 98 9.05 13.53 0.04 13.51 Coffee 13.85 13.76 18.76 2.19 1.30 Rice 7.84 10.99 10.01 9.07 13.74 Cotton 8.43 9.99 7.30 11.58 2 35 Potatoes 7.90 13.89 6.94 4.12 16 18 Yuca 13.54 5.58 10.19 1.97 4.84 Weighted average (with coffee) 10.93 11.86 13.73 4.33 6.02 Weighted average (without coffee) 8.82 10.49 10.08 5.88 9.45 Note. An arithmetic average of production with high-yielding and traditional varieties, except for coffee, which is a weighted average based on area planted a. Interest costs from farm budgets for semester B of 1982; weighted average for traditional and modern farms. b. Includes pesticides, herbicides, fungicides, and insecticides Source, Caja Agraria and capabilities. This has caused net losses to the institution, since no compensatory financing is provided for these activities by the central government. On the income side, Caja cannot increase its lending interest rates-as a matter of fact, at the gov- ernment's request it lowered its average lending rate in 198 3-nor can it increase the proportion of its portfolio (fixed by law) invested in more productive lines of credit. On the cost side, Caja cannot easily reduce the number of branch offices, of which about 70 percent are operating at a loss, nor its input and supply stores, and it has not been able to keep its personnel costs under control by countervailing the demands of its very strong labor union. In the past, Caja's financial problems have been managed through the use of special rediscount lines from the Central Bank, the paying of relatively low interest rates on savings deposits, and sporadic government contributions to replenish Caja's capital. At present, however, there is a consensus that these problems have to be dealt with in a more systematic manner, which requires structural and policy changes; on the institutional side, some of the required changes are Caja's responsi- bility, whereas others require government action to modify external factors that affect Caja's operations. Basically, the costs of the social function of Caja and the TECHNOLOGY, INPUT POLICY, AND MARKETING 149 costs that arise from any inefficiencies need to be identified and quantified. The government, through specific periodic allocations suitably monitored, would as- sume responsibility for the social costs and would undertake to modify the external factors which are affecting Caja's operations negatively. On the other hand, Caja would assume responsibility for improving its operational efficiency. Agricultural Marketing Some improvements that have been made in wholesale marketing are evidenced by an estimated decline in the wholesale margin measured as a percentage of the producer price or the consumer price (see table 8-6). In some recent reviews these margins are considered large, implying the scope for improvements in marketing as those that were made under the programs Of PRODESARROLLo and DR]/PAN. Retail margins seem to have increased during the 19 70s.' These increases, however, might represent, at least in part, improvement in quality and greater shares of processing and packaging in the final products, in response to gradual increases in real income and perhaps lags in transferring consumer price increases to wholesale and producer Table 8-6. Average Marketing Margins for Selected Agricultural Commodities, 19 70-80 (percent of consumer price) Commodity 1970-72 1973-75 1976-78 1979-80 Rice Wholesale 240 25 7 21.6 17.7 Retail 12.6 10.8 17.5' 183 Corn Wholesale 12.6 110 8.6 8.0 Retail 14.2 164 13.7 20.9 Beans Wholesale 36.0 31.4 29.3 23.7 Retail 16.1 14.7 18.0 23 6 Potatoes Wholesale 29.5 22.9 23.8 21.4 Retail 23.6 20.0 31.2 31.1 Beef Wholesale 21.7 20.0 21.5 15.1 Retail 23.8 24.1 23.5 37.2 Milk Wholesale 30.0 35 1 45.4 28.3 Retail 3.9 4.3 4.2 7.1 Note: Average marketing margins are wholesale and retail margins as percentages of the retaili price. Sources: DNP and World Bank estimates made by German A. Rioseco. 150 PRODUCTION POLICY levels. Inadequate improvements in marketing in the face of increasing demand for greater marketing might be another part of the explanation. While the wholesale markets are working fairly efficiently, high margins seem to be based on the high cost of transport between markets, ranging from 3.3 cents to 6.0 cents a ton per kilometer. For fertilizers shipped from Barranquilla ports to Bogota, the cost of transport is equivalent to an increase of approximately 25 per- cent over the ex-port price, and equivalent to $47 a metric ton, while the cost of freight from European ports to Barranquilla is $ 30 a metric ton. A comparable difference of 26 percent over the producer's price is observed in the costs of trans- port that affect rice exports. High costs of trucks and transport equipment on ac- count of tariffs and import restrictions constitute a part of high transport costs. Estimated operating costs and taxes in January 198 3, according to the Ministry of Public Works, were Col$ 30.5 a kilometer for a six-metric-ton truck and Cols45.0 a kilometer for an eighteen-metric-ton truck, taxes representing more than 15 per- cent of these costs. Also, freedom of entry into the trucking industry is limited in Colombia on two counts: first, the import of trucks (assembled or to be assembled in the country) requires prior approval of the Instituto del Transporte, and second, in order to operate, a trucker must be affiliated with a trucking company. Seasonal and year-to-year price variability is also believed to be high, particularly for perishables, and these do not appear to have been reduced significantly through the years. Constraints involve the lack of adequate countrywide market and crop information and data, the lack of storage facilities and other needed infrastructure, and a shortage of marketing credit for investment in infrastructure and for recur- rent expenditures related to purchases and costs of stock carryover. In spite of the improvements, the overall adjustment of marketing may not have been fast enough to keep up with the requirements of rapid urbanization. Thus, wide fluctuations in price, relatively high marketing margins, and heavy losses, par- ticularly in perishables, are attributed to deficiencies in marketing. Inadequate infra- structure in collection, storage, refrigeration, and processing and the lack of ade- quate crop-price and market information are serious problems. Given Colombia's fairly sophisticated and sizable entrepreneurial skills, any inadequacy in market- ing-with the probable exception of the wholesale markets-needs to be examined in the light, first, of the existing legal and regulatory framework, and second, of the economic environment. Cases such as the slaughterhouses, which are usually small, inefficient units, operated by municipalities as sources of revenue and employment, need to be reviewed. Among other factors, the pricing and marketing of agricultural products is af- fected by IDEMA'S price support for staples, OPSA'S price determination or negotia- tion for selected outputs, and the strong function of producer associations in many products. The marketing of fertilizer is dominated by Caja Agraria and FEDERACAFE, which together account for approximately 60 percent of the fertilizer trade, and among producer associations by FEDEARROz and several associations of banana growers, which account for approximately 12 percent of total trade. Out of approx- TECHNOLOGY, INPUT POLICY, AND MARKETING 151 imately forty-two enterprises that participate in the seed trade, Caja controls an average of 18 percent of the national sales. The balance of inputs, primarily pesti- cides, is handled largely by the private sector. Prices paid by producers for inputs, both locally produced and imported, are determined periodically by the Ministry of Agriculture. The public function in marketing has been focused primarily on control, direct intervention, planning, and regulating. As already noted, IDEMA influences prices by purchasing at support prices, selling through the Bolsa Nacional Agropecuaria, S.A. (Bolsa), and wholesale and retail distribution centers, importing or granting licenses for imports, and exporting and maintaining regulatory stocks of a number of basic agricultural commodities-wheat, rice, corn, beans, sorghum, barley, soybeans, and sesame. Except when large imports are made by IDEMA, however, the govern- ment's market stabilization policies have had only a marginal effect because of IDEMA'S limited resources. Aside from its credit role, Caja intervenes in the sale of farm inputs and supplies, handling about 40 percent of total trade through 440 outlets throughout the country. In an attempt to improve marketing efficiency, the government has created and assisted in the creation of several mixed-economy companies. The Union of Agrar- ian Reform Cooperatives, Ltd. (CECORA) is concerned with the development of co- operatives, while the Enterprise for Marketing of Perishable Products (EMCOPER) Is essentially responsible for building and operating cold-storage facilities in competi- tion with the private sector. The Central Wholesale Market of BogotA (cORABAS- Tos) and similar markets in Medellin and Cali were created to build and lease wholesale marketing facilities to private wholesalers; cORABASTOs also participates in day-to-day management of the public markets. The Bolsa, organized in 1979, is involved in commodity trading, processing, and the provision of market informa- tion. FEDERACAFE, although not a marketing institution, is implementing its crop diversification program in the coffee-producing areas, providing technical assistance to some forty multipurpose cooperatives and marketing certain products, most of them perishables. Also, important efforts for improving output marketing of small- scale farmers are being made in specific areas through the DRI/PAN programs. Both Caja and IDEMA are blamed for being inefficient and incurring heavy losses. While these observations are largely true, by and large the underlying problems reflect not only the weaknesses of these two institutions but also extrainstitutional policy decisions regarding their operations. Caja and IDEMA are expected for reasons of equity and political considerations to provide a level of services that is either not in line with their resources and capabilities or most of the time causes net losses to the institutions without compensatory financing for these activities from the government. Any measure to reduce the cost of inputs to agriculture-particularly to small- scale farms-other than through reducing the cost of operation of the ports, changes in taxes and tariffs, and transport and storage costs would need to be ac- companied by substantial improvements in the efficiency of Caja's operations, par- 152 PRODUCTION POLICY ticularly since it controls approximately 40 percent of trade, and, to some extent, determines the market price of inputs. Particular attention will need to be given to the possibility of completely separating the input supply function of Caja from the rest of its activities and to the need to improve its capacity to negotiate its freight needs with the trucking industry. IDEMA'S market-intervention activities could in principle generate sufficient reve- nues to make the institution self-financing, so it does not receive any budgetary or other kind of allocation to finance its recurrent expenditures. The institution is requested, however, to support producer prices and incomes through the purchase of agricultural commodities, mostly in marginal areas, and to maintain regulatory stocks and regulate consumer prices through direct market intervention. On the other hand, the only real source of income to IDEMA is that which is the result of price differentials on commodities imported by the institution. For several reasons, however, including the explicit policy of limiting government intervention in the market, the volume of imports and the size of the price differentials is largely insuf- ficient as a source of income to IDEMA, causing a shortfall of operating funds. Policy Conclusions Research and extension suffered from relative neglect during the I 970s and cur- rent efforts to reverse this trend deserve full support. The decrease in funding for the ICA, the overburdening of the institute with regulatory functions, and the loss of first-rate scientists are problems to be overcome. Continuing efforts to upgrade research need to be quickly linked to extension, which has also suffered in the past from the fragmentation of the agricultural extension services in the country and inadequate funding and policy orientation. Additional irrigation and drainage are of high priority in selected areas in which water is the constraint to higher yields and to the harvesting of two crops a year. In particular, public investment for flood control and drainage could be effective in converting pastures into land suitable for intensive cropping. Rehabilitation of irri- gation is also justified in general by its cost-effectiveness. Expansion of agricultural production on new land with new irrigation facilities is generally a more costly option. The rapid reduction of Colombia's rich forest resources, soil erosion and conser- vation problems, and increasing siltation of water reservoirs are serious policy con- cerns. A greater capacity to execute programs in the use of natural resources, which will clearly require greater public awareness of the problems, needs to be developed. INDERENA'S institutional capabilities for policy execution need to be strengthened, and conservation education programs need to be stepped up. A program of research and baseline studies on critical areas is called for to enable INDERENA to evaluate policies and projects, funding for which might need to be sought externally. The ad hoc approach to forestry development needs to be replaced by a more systematic TECHNOLOGY, INPUT POLICY, AND MARKETING 153 policy. In particular, the execution of pilot projects, such as the Upper Magdalena Pilot Watershed Management Project, and the follow-up to larger-scale projects should be placed within the framework of a national forest and natural resources development program. The most significant element of estimated farm budgets is labor costs, which represent 40-50 percent of variable costs. Policy intervention consists principally of minimum wage legislation, arrived at through negotiations between the govern- ment and the agricultural unions. Negotiations for 1984 had produced a 28 percent increase in the minimum wage by midyear, outstripping the inflation rate adjusted for gains in productivity. Future wage negotiations need to be particularly cogni- zant of the sectoral and macroeconomic implications of excessive work increases. High internal transport and port-handling costs, and to a lesser extent financial costs and import duties, have contributed to the high prices of urea in Colombia. Improvements in the functioning of COLPUERTOS would be vital to the achievement of long-term reduction in the prices of urea to farmers. Efforts to reduce internal transport costs might include measures to reduce high tariffs on agricultural equip- ment and transport vehicles, lower the unduly restrictive entry and low use of ca- pacity in trucking, increase the transport of fertilizer in bulk, and expand the use of port facilities at Buenaventura on the Pacific, which is closer to important farm production centers such as those of the Cauca valley. In the matter of composite fertlilzers, more analysis is needed of the present level of protection and subsidy to the domestic industry, its efficiency, and its pricing policy. If the industry requires protection, ways and means of reducing the costs of its inputs should be explored before domestic output sales prices-that is, of the composite fertilizers-are raised. Long-term improvements in wholesale marketing are evidenced by an estimated decline in the wholesale margin needed as a percentage of the producer price or the consumer price. These margins are nevertheless considered to be large by some recent reviewers, implying that there is scope for improvements in marketing, as has been the experience under the programs of PRODESARROLLo and DRI/PAN. Retail margins seem to have increased during the 1970s. These increases in margins, how- ever, might in part represent improvement in quality and greater shares of process- ing and packaging in the final products in response to gradual increases in real in- come and perhaps to lags in transferring consumer price increases to wholesale and producer levels. Inadequate improvements in the face of growing customer demand for greater marketing services might be another part of the explanation. Proposals have recently been made for marketing projects intended to reduce marketing mar- gins and to bring about institutional improvements; a clear conception of the way these efforts would contribute to agricultural development, however, is yet to emerge. The estimated annual cost of institutional credit is about 11 percent of the pro- duction costs of twelve major crops, including coffee. The rates on institutional credit, while positive in real terms during 198 3-84, are estimated to be as low as 55 percent of the market rate. (As noted earlier, a reason for "high" real market rates is 154 PRODUCTION POLICY the acceleration of the crawling peg.) The 45 percent credit subsidy implies a saving in credit cost to farmers of about 9 percent of their annual production costs. Re- moving the subsidy would increase production costs less than 9 percent if, as ex- pected, it would lead to greater availability of credit, lower market interest rates, and the possibility that farmers could obtain credit from both sources. In compari- son, fertilizer constitutes about 12.5 percent of production costs while machinery rental accounts for about 14 percent. While the subsidy component of the credit cost is not insignificant, other measures-an adequate exchange rate and a reduc- tion in the costs of other inputs (by a further reduction in tariffs and greater effi- ciency in transport and port handling)-could eventually be substituted for it. The present estimates, however, may understate the full credit costs to the farmer, in- cluding nonfinancial costs of the institutional credit and the cost of noninstitutional credit. Additional sector work is proposed to analyze the sources of institutional and noninstitutional credit, the efficiency of existing arrangements and their benefi- ciaries, and the way the system can successfully cope once policy reforms in other areas have led to increased demand for inputs. In the past, Caja Agraria's financial shortfalls have been managed through special rediscount lines from the Central Bank, the payment of relatively low interest rates on savings deposits, and periodic government contributions to replenish Caja's cap- ital. A consensus is emerging, however, that these problems must be dealt with systematically through structural changes. The costs of the social function of Caja need to be identified and quantified separately from those that may arise from any inefficiencies. The government, through specific periodic allocations, suitably mon- itored, would need to assume responsibility for the social costs and try to reduce them. On the other hand, Caja should assume responsibility for improving its oper- ational efficiency: Caja's credit and input functions might be split with a view to achieving greater financial accountability in each of these distinct areas. Notes 1. DNP, "Diagnostico del sector agraro" (BogotA, 1983, processed). 2. Value-added figures from DANE include a combined estimate for coffee (01), other agricultural products (02), and animal production (03), and separate values for processed coffee (08) and sugar manu- facturing (12). These categories are as explained in chapter 1. In disaggregating (01), (02), and (03), approximations of gross output have been used, despite the obvious double counting that results. It should be noted that during the period 1970-81, the value added in the agricultural sector has consis- tently been equal to about 75 percent of gross output. 3. UNico, Feasibility Study Report for an Ammonia and Urea Project, vol. I (Tokyo: UNICO International Corp., July 1984), table 4. 4. These and remaining estimates in this section are from Urrutia, Winners and Losers in Colombia's Economic Growth of the 1970s. and 1982 World Bank calculations using data from DANE. 5 A. Hernandez Gamarra and G. Jimenez Perdamo, "Consideraciones econ6micas y financieras sobre la viabilidad del seguro agrocredito en Colombia" Cuadro 4-3, Estudio de Consultoria para el Proyecto 1ICA-oPSA del Ministerio de Agricultura, September 1982. TECHNOLOGY, INPUT POLICY, AND MARKETING 155 6. The nominal interest rate on Caja loans was 26 percent, while FFAP loans were generally made at a nominal interest rate of 2 1 percent. Because of the Colombian system of deducting quarterly interest payments in advance, the effective interest rates turn out to be 24 percent and 31 percent. It must be noted, however, that some transactions are made at the nominal rate, although to what extent is not known. 7. See World Bank, Colombia: The Investment Banking System and Related Issues in the Financial Sector (Washington, D.C., 1985). Nominal rates were lower, and some transactions are expected to have been carried out at the lower rate. 8. A DNP study carried out in 1977, in which the marketing margins of twenty-four products are reviewed, indicates an average wholesale margin of 15 8 percent and a retail margin of 24.2 percent, for a weighted average of 36.2 percent. A similar study made in the United States by the Department of Commerce indicates a wholesale margin of 6.1 percent and retail margins of 21.4 percent for marketing chains and 18.9 percent for independent retailers; the average total margin was 24 3 percent. 9 Conclusion THIS BOOK has been concerned with policies intended to bring about economic adjustment while at the same time sustaining development. In such a study Colom- bia has offered the opportunity for a joint treatment of macroeconomic and sec- toral issues. While growth options in agriculture are central, the work has also provided a trade-related macroeconomic analysis that goes beyond the concerns of any particular sector. The study has been focused on the links between macroeco- nomic issues and agricultural prospects and has shown how these relations influence performance. The first chapter showed how the dynamics of the principal export item, coffee, affect other agriculture and the rest of the economy and how macroeconomic poli- cies-made partly because of coffee developments-affect agriculture. In many de- veloping countries instability in the external and domestic sectors arises from wide price fluctuations in the world markets for a principal export commodity. Macro- economic policy, particularly with respect to the exchange rate and import restric- tions, poses policy dilemmas in such circumstances. Trade policies have a special significance for agriculture, which is a highly tradable sector, and restrictive trade measures can hurt this sector disproportionately; this factor gave rise to the further analysis of export and import policies in chapters 2 and 3. In the five subsequent chapters developments were considered from the agricul- tural viewpoint, although macroeconomic effects are central to them. Thus, the agricultural incentives discussed in chapter 4 are derived as much from exchange- rate, import, fiscal, and monetary policies as from sector-specific interventions. Sec- toral policies of price supports and credit subsidies are important, but their effects are smaller in Colombia than in some other countries. Issues of price stabilization are also related to overall import policies, even though the proposed measures, discussed in chapter 5, are sectoral. Coffee policy was considered in detail, also from the sectoral viewpoint, in chapter 6, while, as emphasized throughout the book, it is an area of considerable macroeconomic importance. Chapters 7 and 8 156 CONCLUSION 157 dealt with production policies in agriculture that should serve to complement econ- omywide incentive and price policies. Macroeconomic Policy, Performance, and Agriculture While in this work the solid long-term performance of the Colombian economy has been noted, much of the policy discussion has concerned the more recent devel- opments, which have been more difficult. The economy saw a slowing down of real growth rates and generation of employment during 1981-83, after the excel- lent record of the 1970s. By 1984, unemployment surveys in Colombia's four largest cities showed an unemployment rate of about 13.5 percent, almost twice the level registered at the end of 1981. The current account in the balance of payments and the fiscal balances also deteriorated during 1981-8 3, constraining the viability of growth-oriented policies. Significant improvements in the current account of the balance of payments were seen in 1984-85 with the initiation of adjustment poli- cies. Both the public and the private sectors, however, continued to face significant difficulties in tapping the international capital markets. Consequently, the loss of net foreign exchange reserves, which was substantial-$ 1.8 billion-in 1983, con- tinued in 1984. With a deepening of the adjustment measures, the reserves position can be expected to stabilize and improve, and present indications support this view. The influence of external factors in the recent problems has been noted. Exports to Venezuela, for instance, one of Colombia's principal nontraditional trading part- ners, and to Ecuador, Mexico, Argentina, Brazil, Costa Rica, and Chile have de- clined significantly since 1982, following economic problems, major devaluations, and import restrictions in those countries. The external debt problems of other Latin American countries have also been a proximate cause of the pressure on the capital account of the balance of payments. The contribution of domestic develop- ments to the recent difficulties has also been emphasized. The coffee boom of the second half of the 1970s, for example, set in motion an accelerated growth in the money supply and inflation despite the stabilization effort made by the authorities. In the absence of adequate adjustments in the crawling peg, these events led to a significant appreciation of the real exchange rate, reducing the incentives to pro- duce noncoffee exportables. In addition, the fiscal deficit grew during the 1980s, contributing to a higher rate of growth of domestic credit and, thus, of domestic inflation than would have been likely to occur had measures to control the deficit been initiated more promptly. In these circumstances, Colombia needs to adopt significant measures for adjust- ment with attention to creation of employment. If such measures are taken in a timely manner, the country will not need to sacrifice much growth or employment generation. An outward-looking strategy would be consistent with the need for further adjustment with growth and the creation of employment. Such is the present government's economic strategy. A program for fiscal, monetary, and ex- 158 CONCLUSION change rate adjustments that would, among other things, improve the competitive- ness of noncoffee tradables has been initiated, and politically difficult decisions to execute the policy package are being made; these policies for adjustment with growth and for employment generation need to be deepened and sustained. Co- lombia has a long record of responsible and careful economic management. Mainte- nance of the current efforts should once again provide the basis for continued and rapid development in the remainder of the 1980s. The policies being put in place represent a critical phase of medium-term adjust- ment, in which stability would be achieved, permitting continuation of growth as the economy is redirected toward export promotion. Subsequently, policies would be concentrated on further strengthening of the incentives for external trade. In agriculture, medium-term policies would stimulate productivity and increase ex- ports. Part of the macroeconomic adjustments are in place, as of this writing, and some positive results have already been obtained. The 1985-86 macroeconomic policy package concerns further reductions in the expenditure-revenue gap of the public sector, including a review of public-sector investments and improvement of their effectiveness, postponement of large and long-gestating new projects, and maintenance of the prices of public utilities at ap- propriate levels; a slowdown in the expansion of credit to the government by the Central Bank; full correction of the overvaluation of the real exchange rate; liberali- zation of imports that are needed for exports; and a scaling down of external bor- rowing targets, reducing the demand for external funds. These policies are being complemented by measures to strengthen directly the financial and real sectors of the economy, with attention to generation of employment. The authorities recognize that even with domestic policy adjustments, a resump- tion of external commercial bank financing-to, say, the 1981-82 levels-would be essential to the achievement of adjustment with growth. Projections of the bal- ance of payments indicate the need for such a revival in commercial bank financing, to be followed by the prospect of significant expansion of resource-based exports- petroleum, coal, and nickel-and other nontraditional exports during the second half of the 1980s. In view of the government's adjustment efforts, the World Bank has augmented its policy and financial support to the country; an agreement could now be worked out with the commercial banks on the additional financing needed. The projection that is based on the combination of increased macroeconomic pol- icy adjustments and capital inflows, with borrowing kept within prudent limits, envisages a revitalization of sectoral performance contributing to a more rapid eco- nomic growth with balance of payments stability and creation of employment. The post-1975 agricultural expansion has been in good measure based on coffee, a product for which only a modest increase in demand in the world markets is projected for the 1980s. The principal macroeconomic constraints to more vigor- ous noncoffee expansion have comprised a high inflation rate, an overvalued ex- change rate, and import restrictions, which impose a "tax" on a heavily tradable sector such as agriculture. The effects of macroeconomic policy may have been CONCLUSION 159 unintentional, but the net effect from the mid 1970s to 1983 on noncoffee agricul- tural tradables was negative. Through sectoral policies an attempt has been made to increase the incentives for the relatively small group of imported cereals through high levels of import protection and domestic price supports. In addition, credit subsidies to producers and storers, albeit moderate, have provided some sectoral incentives. At the same time, however, agricultural investments by the public sector are estimated to have declined significantly in real terms since the mid 1970s. The adoption of a more neutral macroeconomic policy posture-as envisaged with the changes that are being made-would be essential, beginning with the elim- ination of overvaluation of the exchange rate, reduction of the fiscal deficit, ration- alization of the investment program, liberation of imported inputs needed for ex- ports, and strengthening of the financial sector. Depreciation of the real exchange rate began to take effect in a significant way in 1984, and further substantial prog- ress was recorded in 1985. Such necessary macroeconomic adjustments can also produce difficulties at the financial and real sector levels which need to be addressed by policies in these areas. Once the principal macroeconomic disincentives were eliminated, special sectoral incentives such as price supports and credit subsidies would be less justified; in view of the efficiency losses of the latter, they could be phased out gradually. On the other hand, there is a case to be made for supporting high-priority investments for irrigation, input supply, research, extension, and mar- keting. Coffee Policy and Noncoffee Production Although coffee stocks may have contributed somewhat to strengthening Co- lombia's export share under the coffee agreements, the recent stock trends indicate significant costs to the coffee sector and the economy. At the end of 1984, Colom- bia was estimated to carry accumulated coffee stocks equal to some 122 percent of its exports. The Federation of Coffee Growers is committed to a containment of coffee incentives and is pursuing a mix of policies not only to match production with demand, but also to sustain the long-term yield and health of the caturra vari- ety. An effort is being made to hold the coffee price guaranteed to producers and to eliminate fertilizer subsidies to coffee growers. At the same time, the federation has embarked upon a program to provide income support through direct production and marketing incentives to other products in the coffee zones. The domestic producer prices of coffee increased at a rate lower than that of the consumer price increase during 1980-83. A source of the recent pressure on the producer price, however, has been the significant increase in the rate of the crawl- ing peg, which, together with some increase in the external coffee price, produced a large increase in the peso value of the export surrender price during 1984. Consid- ering that the exchange rate is expected to continue to depreciate rapidly, a contin- ued resolve might be essential to hold the domestic producer price. 160 CONCLUSION Even if the producer price should be maintained in real terms, production might only be stabilized, despite the increases in production costs and the recent outbreak of roya. Unless there is a dramatic increase in demand for Colombian coffee, given the iCA, some decline in production may be needed during the second half of the decade. Several issues need to be analyzed further. Given the life cycle of existing trees, what incentive structure would be needed to bring about the needed adjust- ment in production? Also to be evaluated are the possible mixes of coffee disincen- tives and diversification incentives: how much incentive might be allocated for stumping coffee trees? What additional efforts would be required for diversifica- tion? Sectoral Incentives One of the sectoral incentives reviewed was the price support offered by IDEMA for rice, corn, sorghum, soya, wheat, barley, and sesame, all of which except rice are also imported. The domestic support price of these commodities exceeded the f.o.b. import price-unadjusted for special circumstances such as export subsidies abroad-by 50-100 percent during 1983, although the rapid depreciation of the peso in 1984-85 may have reduced this difference subsequently. Despite the fact that IDEMA has a virtual monopoly on imports and makes a profit on the sale of its imports, the agency has run a deficit on the whole on account of its price support operations. Apart from the fiscal issue, there is also the efficacy question of provid- ing special incentives for crops, such as wheat, in which Colombia does not appear to have a comparative advantage. Technical analysis so far has pointed to potential net gains from phasing out the high protection for some products. Several issues need to be examined in greater detail. What would be the time path for phasing out these output subsidies? How would it be linked to liberalization in the importation of the same commodities? Would the approach produce undesirable income distri- bution effects, and if so how could they be offset? How can some of the price instability concerns be better addressed through selective use of futures markets? The high price of inputs, particularly of fertilizer, is one of the most often men- tioned sectoral disincentives. A suboptimal use of fertilizer in most noncoffee com- modities has been the result of high farmgate prices for fertilizers-reaching a level twice the world market levels, except perhaps for coffee, for which the input price is still less than that for others. These prices are to some extent caused by taxes and tariffs and, more significant, by high port charges and high domestic transport costs. Policies to reduce port and internal transport costs would therefore be highly bene- ficial to the sector. Fertilizer import and price policies themselves have been under government control, but interventions have frequently been the results of pressures from one group or another. There has not been a concerted price policy with clearly spelled out objectives and definitions of the desirable levels of subsidies and CONCLUSION 161 taxes to be borne by the various parties involved. Further analysis of these issues is clearly in order. While subsidized credit has provided some special agricultural incentives, the present system has several inadequacies. During the past decade growth in agricul- tural credit was erratic, and the average annual increase in outstanding loans was below the growth in agricultural output. At the same time an increasing share of agricultural credit has come to depend on the Central Bank's rediscounts, while the resources mobilized by commercial banks and agricultural banks-Caja Agraria, Banco Cafetero, Banco Ganadero-have declined. Furthermore, the existing sys- tem favors short-term investments over long-term, larger farmers over smaller, and agricultural primary production over processing and marketing credit. Finally, the administrative costs of handling agricultural credit seem to be high, especially in the case of Caja Agraria, the main agricultural bank. Related work by the World Bank on the financial sector issues has suggested policy directions for agricultural credit. Issues to be translated into concrete policy recommendations concern interest rates, term transformation, and levels of subsi- dies. On interest rate policies, the existing forced investment requirements that are applicable to resources of financial institutions need to be lowered gradually in or- der to reduce the segmentation between the free and subsidized credit markets. Concerning term transformation, introduction of a floating interest rate might be considered, linking lending rates to an index representing the average cost of raising deposit resources. Turning to levels of subsidy, the Central Bank's rediscounting of funds might replace the current fixed interest rate system by a floating rate system. Furthermore, the proportion of the loans refinanced by the official financial funds needs to be reduced in time, and the financial intermediary should be expected to contribute a larger share from deposit resources mobilized from savers in financing term loans. While the greater part of agriculture-related investments is made by the private sector, it has been shown that public sector investment is significant in selected areas such as infrastructure development, including rural roads (Ministry of Transporta- tion), storage (IDEMA), rural electrification, irrigation, and land development (HI- MAT, INCORA), research and extension (primarily IcA), provision of credit (the Cen- tral Bank through the Agricultural Fund and Caja Agraria), and watershed management (INDERENA and Department Development Corporations). During the past decade, public expenditures in the sector have decreased steadily and signifi- cantly. In 1984 a part of the agricultural investments of the public sector was to be financed out of a Public Investment Fund, which depended on the decreasing in- come of the Special Exchange Account; consequently there were serious im- balances between investment expenditures and recurrent expenditures that are diffi- cult to reduce. In this study a comprehensive review of the public and private sources and use of funds for agriculture has been called for, with a view to ensuring that agricultural 162 CONCLUSION investments are in line with the expected contribution of the sector to growth, exports, and employment. The results of the review would be used to devise pro- grams to support the availability of funds for high-priority agricultural investment in commercial and traditional agriculture, improve the financial discipline of agri- cultural line agencies and their budgetary monitoring and control capabilities, and carry out annual reviews of the investment program and recurrent expenditures, relating performance and expenditures to the agricultural objectives. APPENDIXES A The Interaction of Coffee, Money, and Inflation in Colombia Sebastian Edwards IN THIS APPENDIX a simple model of the effects of changes in the price of coffee on the competitiveness of the rest of the economy is presented. In order to simplify the exposition, a model will first be derived in which the monetary effects of changes in the price of coffee are assumed away; this assumption will be relaxed later by intro- ducing a money market. The model will then be tested, using annual data for Co- lombia. It will trace quite closely the behavior of the price level and of the relative producer prices of noncoffee traded goods in Colombia, which is central to the analysis presented in chapter 1. Changes in the Price of Coffee and Competitiveness Consider a small open economy with a fixed exchange rate that produces three goods: coffee, C, other tradables, T, and nontradables, N.' Assume further, for the sake of simplicity, that the exchange rate is equal to 1; this assumption will be relaxed later. The excess demand for nontradables is assumed to depend on prices and income. Consider first the case in which this excess demand is not affected by the relative price of coffee. As a first approximation, this can be justified by assuming that domestic residents do not consume coffee-or consume negligible amounts in rela- tion to the amount exported-and that factors used in the production of coffee are sector-specific, in both the short run and the long run. These assumptions will be relaxed subsequently. In equilibrium the excess demand for nontradable goods will be equal to zero, and under these assumptions it can be written as: (A.1) N = N(qT, Y) = 0, (+) (+) where qT is the domestic price of tradables in relation to the price of nontradable goods-that is, qT = PT/PN-and Y is real income in terms of nontradables. The 165 166 APPENDIX A signs in parentheses below the arguments of the function refer to the assumed signs of the partial derivatives. The positive sign of qT stems from the assumption of gross substitutability between nontradable goods and tradable goods. Equilibrium in the nontradable sector requires that the excess demand for this type of good is equal to zero, both in the short run and in the long run. In this model it is possible to think of qTas the real exchange rate. The reason for this is that in a country that produces a major commodity export good-that is, coffee in Colombia, or oil in Indonesia- the appropriate measure of the real exchange rate will be the relative price of trad- ables other than the commodity export to the price of nontradables.2 In (A. 1) Y is expressed in terms of nontradable goods and is given by (A.2) Y = H + qTHS + qcC, where H , HS, and qcC are supplies of nontradables, tradables, and coffee, respec- tively, and qc is the relative price of coffee in terms of nontradable goods. The supply of coffee is held fixed in order to simplify the analysis. If it is further assumed that real income, Y, equals expenditure E, equation (A. 1) for nontradable goods equilibrium implies that the external sector is also in equilibrium. Maintaining the assumption of gross substitutability, we can depict the equilib- rium situation in the nontradables market in figure A-1, which has been adapted from Dornbusch.I The NN schedule describes the combination of qT and qc that is compatible with equilibria in the nontradable goods market. The slope of this curve is given by: (A.3) dqT - _ aNlaY __ < 0. dqc [(aN/8qT) + (aN/a Y)HST] The ray OT, on the other hand, measures the price of other tradable goods in relation to the price of coffee, PT/Pc. The initial equilibrium position is given by A, with equilibrium relative prices being equal to qO and qc, respectively. Assume now that there is an exogenous increase in the price of coffee. What will be the effect on the relative price and degree of competitiveness of other tradables' The OT ray will then rotate clockwise toward OT' as in figure A-2. If the (nomi- nal) price of nontradables were constant, the new equilibrium would be given by B, with a constant relative price of noncoffee tradables with respect to nontradables. As long as the slope of the NN is negative, however, at B there will be excess demand for nontradables that will require an increase of the relative price of these goods, with respect to the prices both of coffee and of other tradables. The final equilibrium will then be attained at C. As a consequence of the increase in the price of coffee, there has been a decrease of the relative price of noncoffee tradables, both with respect to coffee-that is, PT/Pc-and with respect to nontradables-that is, from qT to q' . This reduction in the relative price of other tradables, of course, will encourage resources to move out of the other tradables sector into the other sectors of the economy. This phenome- COFFEE, MONEY, AND INFLATION 167 Figure A-1. The Prices of Coffee and Other Tradables in Relation to the Prices of Nontradables TT ( Pr- T K =2 P/ Pc q T N A 'T / /N 00 P,c Source: Adapted from Rudiger Dornbusch, "Tariffs and Non-traded Goods," Journal Of International Economics. vol 4 (May 1974), pp. 177-85. non is similar to what occurred in oil-exporting countries as a consequence of the increase in the price of oil in the 1970s and has been labeled the Dutch disease or the deindustrialization effect.' In figure A-2 the degree of loss of competitiveness of the noncoffee tradables sector-that is, the degree of decline of qT (= PT/Pjj)-depends on the slope of the NN curve. At one extreme, if the NN curve is a vertical line the negative effect on qT of an exogenous increase in the price of coffee will be maximum. On the other hand, if all the additional income generated by the higher price of coffee is spent on tradables, with none of it being spent on nontradables, the NN curve is a horizontal line, and there will be no effects on qTof an increase in the price of coffee. The final effect of a higher-priced coffee on the real exchange rate will depend in a crucial way on the proportion of the higher income generated by the coffee boom that is spent on nontradables. 168 APPENDIX A Figure A-2. The Effect of an Increase in the Price of Coffee on the Prices of Other Tradables /T T (Pr /T ,1 _ T P/ A / / B/ c qT -( Consider now the more general case, in which coffee is also consumed domesti- cally, but in which factors used in its production are still sector-specific. Then, the excess demand for nontradables will be given by: (A.4) N = N( qT, qc, Y) = 0, (+) (Q) (+) where the sign of aN/8qc will be positive if coffee and nontradables are substitutes, negative if they are complements. The slope of the NN curve will now be equal to: (A.5)dqT (8N/qc) + (8N/aY)C dqc (ON/8)Hs + (ON/8qT) This expression can be either positive or negative, depending on the sign of aN/aqc. If coffee and nontradables are complements (aN/89c < 0), it is possible COFFEE, MONEY, AND INFLATION 169 that the numerator of equation (A.5) will be negative and the slope of the NNcurve will be positive. (Notice that aN/aqc < 0 is a necessary-but not sufficient-condi- tion for NN to be positively sloped. The sufficient condition is that aN/aqc + (aN/a Y) C < 0.) In this case an increase in the price of coffee will cause an increase in the relative prices of other tradables and thus in resources that move from the nontradable goods sector into the other tradables sector. If coffee and nontradables are substitutes, however-which is the more plausible assumption, given the level of aggregation considered in this model-the NN curve will be negatively sloped and the analysis presented in figures A- I and A-2, which indicates that a higher price of coffee will reduce the degree of competitiveness of other tradables, will still hold. In the rest of this section it will be assumed, unless otherwise indicated, that the three goods involved are substitutes in consumption, so that equation (A.5) is negative and the NN curve is negatively sloped, as in fig- ure A-1. The preceding analysis shows that under a set of plausible assumptions, increases in the price of coffee will generate an equilibrium reduction in the relative prices of other tradables-in relation to both coffee and nontradables. This movement of relative prices will reduce the level of competitiveness of this sector, noncoffee trad- ables, with resources tending to move out of it. To the extent that this is an equilib- rium result, no particular policy measures should be taken to avoid it.' If the change in the price of coffee is only temporary, however, and the capital market presents some imperfections, there is an argument for implementing policies that will help firms in the noncoffee tradables sector to "survive" this short-run squeeze in their profitability. The more interesting aspect of the model presented here, however, is that it shows that even in the absence of money and inflation, increases in the price of coffee will tend to reduce the degree of profitability of other tradable goods. Coffee, Money, Inflation, and Competitiveness The preceding analysis has been focused on the long-run effect of an exogenous increase in the price of coffee on the real exchange rate and competitiveness of the rest of the tradables industries. The analysis, however, has excluded dynamic as- pects. In this section some dynamic considerations will be introduced into the model. To accomplish this two things will be done: first Harberger will be followed in the explicit introduction of a slowly clearing monetary sector, and second, a crawling peg system will be considered., In order to organize the discussion it will first be assumed that the exchange rate is fixed. Later, the exchange rate assumption will be changed. Under these circum- stances, an increase in the price of coffee, in addition to its real effects, will affect both the supply of and the demand for money. It will increase the supply of money by producing a balance of payments surplus, which the Central Bank will monetize. (It is assumed that the capital account is exogenous and subject to controls.) The 170 APPENDIX A demand for money will increase as well, as a result of the increase in income brought about by the higher price of coffee. Theoretically, the overall result may be either a short-run excess flow or supply of money or an excess demand for it. By Walras's law these situations, respectively, imply an excess demand for goods- both tradables and nontradables-and an excess supply of goods. In the former situation, the excess demand for nontradable goods caused by this short-run mone- tary disequilibrium will create inflationary pressures that will reinforce the effect caused by the real factors discussed earlier-the increase in income caused by the increase in coffee prices. The result of this process will be that the real exchange rate, qT, will decrease in the short run by a greater amount than would be caused by real factors alone. In this case, the nominal price of nontradable goods will tend to overshoot its new long-run equilibrium, and the loss of competitiveness of the non- coffee tradables sector-measured by the decrease of qT-will be greater in the short run than in the long run. If, on the other hand, there is an excess demand for money, qT will decrease in the short run by less than real factors alone would indi- cate. In either situation-excess supply of money or excess demand for money-as the monetary equilibrium is slowly restored through balance of payments surpluses or deficits (under the fixed-rate assumption), qT will move to its new long-run equi- librium value as determined by the real factors in the model discussed in the preced- ing section. This discussion can be formalized. The excess flow supply for money in nominal terms, ME, is given by: (A.6) ME = AMS _ AMD, where AMS is the change in the nominal supply of money, and AMD is the flow demand for money in nominal terms. Assuming that the demand for money equa- tion MD (in nominal terms) depends on the usual arguments-real income, the interest rate, i, and the price level-we can write M' as: (A.7) MD = PL( i ,Y (-) (+) where P is the price level given by: (A.8) P = P7PN and where the domestic price of the noncoffee tradable goods is given by (A.9) PT = ePT, where e is the exchange rate and P* the international price of noncoffee tradables. Notice that in the definition of the price level, P, in order to simplify the exposition the price of coffee has not been included. It is further assumed that ME is equal to zero only in the long run. In particular, an increase of AMS will cause a short-run excess flow supply of money, which, under the assumption of fixed rates, will be slowly eliminated through the balance COFFEE, MONEY, AND INFLATION 171 of payments. It is further assumed that an excess flow supply of money will be reflected in an excess demand for nontradables and an excess demand for noncoffee tradables. Then, equation (A.1) must be modified to incorporate the assumption that in the short run, an excess flow supply of money is translated in part into an excess demand for nontradables. (A.10) N= N(qT,ME Y). (+) (+) (+) In terms of figure A-I an increase in ME will produce a downward shift of schedule NN. The model is completed by specifying the balance of payments and money supply equations. The balance of payments is defined as: (A.11) B = AR = PC - P,E, + CF, where ET stands for excess demand for traditional tradables; C is the amount of coffee exported; CF refers to capital flows, which are assumed to be exogenous; and AR is the change in international reserves. It is also assumed that aB/aPc > 0; that is, an increase of the price of coffee will bring about an improvement in the balance of payments.' The supply of money, on the other hand, is given by (A.12) Ms = M_ + AR + AD, where MS, is the supply of money in the preceding period, AR is the change in international reserves-that is, the balance of payments-and is given by equation (A. 11), and AD is the increase in domestic credit. From (A. 12), of course, AMS AR + AD. This means that AMs could be considered high-powered money. From (A. 11) and (A. 12) it is easy to see that to the extent that an increase in the price of coffee will produce a balance of payments surplus, AMs > 0s Further, assuming that this positive value of AM s produces a short-run excess flow supply for money, ME will increase, and there will be an excess demand for nontradable goods (see equation [A. 10]). In terms of the diagrammatical analysis, this case is captured by figure A-3.' The exogenous increase in the price of coffee simultane- ously causes a downward shift of the NN curve to N'N'-as a consequence of the excess supply of money-and a rotation of the OT ratio to OT'. The NN curve will shift downward, since if there is an excess supply of money at the old relative prices for nontradables, there will be an excess demand for these goods. The new short- run equilibrium will be attained at S. Final equilibrium will be obtained, as before, at C. The dynamics are character- ized by shifts of the N'N' curve to the right towards the NN curve. The speed of this adjustment will depend on how fast the excess flow supply of money is elimi- nated. As can be seen, in this case relative price of noncoffee tradables will under- shoot its final equilibrium level. This means that the loss of competitiveness of the 172 APPENDIX A Figure A-3. The Effects of an Increase in the Price of Coffee on the Money Supply and Equilibrium N T = PC/ qi N', S A T' P q , . LL I\ S SN' qc qc noncoffee tradables sector, as measured by the decline of qT, will be greater in the short run than in the long run. For the sake of simplicity, a fixed exchange rate has been assumed in the analysis presented up to this point. If, however, a crawling peg system is allowed, the results presented will not be altered in any significant way. Specifically, equation (A. 10) on the nontradables market equilibrium condition remains unchanged. Now, how- ever, it will be seen that in the steady state, assuming that P* and P* do not change, the following expression will hold (from [A.8] and [A.9]): (A.13) I = P = P, where (^) refers to percentage change. From (A. 13), it can be seen that the real exchange rate-defined as s = eP*/Pv-wil be constant. As was discussed, as a consequence of an increase in the price of coffee there will be pressures-stemming from both the income effect and the inflationary effect-for P to increase. If the monetary authorities do not alter the rate of devaluation of the crawl, the following will be the result: (A. 14) e < P < PN. COFFEE, MONEY, AND INFLATION 173 In other words, the real exchange rate will decline with the consequent loss of competitiveness in the noncoffee tradables sector. Notice that the foregoing discus- sion can be considered a special case, where j = 0. Of course, the monetary author- ity has the option of accelerating the rate of crawl so that this real appreciation can be avoided, at least in part. This, however, was not what happened in Colombia during the last coffee bonanza. In fact, as discussed in chapter 1, the authority slowed the rate of devaluation of the crawling peg. This reaction of the government seems to be consistent with empirical evidence on the determinants of crawling peg rules. " Empirical Results It has been argued that an increase in the price of coffee would tend to produce a balance of payments surplus, an increase in the quantity of money, and inflation, which, if not matched by an equivalent devaluation, would generate a real apprecia- tion of the domestic currency, squeezing the profitability out of the noncoffee trad- able goods sector. These aspects will be analyzed empirically in this appendix, using annual data for the period 1952-80. Specifically, whether higher (lower) prices of coffee have caused higher (lower) rates of growth of high-powered money in Co- lombia will be investigated. The relation between the growth of high-powered money and inflation will also be analyzed. The model tested in this section is given in semireduced form by equations (A. 15) and (A. 16), where, as before, (^) refers to a percentage change." (A.15) = o + Ea A^fA_ + 'y,DEH, + *Y2P, - U, (A.16) P, = UO - 6,M, + 52j, + 63 PX + 64DUM, + e, where: P = price level (consumer price index) M = high-powered money DEH = ratio of fiscal deficit (in nominal pesos) to high-powered money P, = nominal price of coffee, in pesos y = real income PX = "world" price of tradables expressed in pesos DUM = dummy variable that takes a value of zero between 1952 and 1966 and I from 1967 onward. Equation (A. 15) postulates that the rate of growth of high-powered money de- pends on its past rates of growth (up to three periods), on the magnitude of the fiscal deficit, and on the rate of increase of the price of coffee. Variable DEH -which measures the fiscal deficit in relation to high-powered money-is included, since the fiscal deficit in Colombia is financed in part by the creation of money.12 174 APPENDIX A From an empirical point of view the deficit included in (A. 15) has to be calculated excluding from government revenues those generated by the revaluation of reserves through the Special Exchange Account. On the other hand, the inclusion of P, in (A. 15) captures the hypothesis that changes in the price of coffee will bring about the accumulation of international reserves and a higher rate of change of high- powered money. In the estimation of (A. 15), it is expected that y, > 0 and Y2 > 0. Equation (A. 16), on the other hand, is a traditional open-economy inflation equation." This expression relates the rate of inflation, P, to the rate of growth of high-powered money, the rate of growth of income, and the rate of change of external prices. This equation responds to the notion that the price level is of the following form: P = PUPO-a) T TN (see equation [A.221), with the prices of nontradables, PN, responding to monetary pressures (and the rate of devaluation), and the prices of tradables, PT, being af- fected by external prices (world inflation plus the rate of devaluation). In theory the change in interest rates should also be included in equation (A. 16); lack of the appropriate data, however, made this impossible. In equation (A. 16) it is expected that 6, > 0, 5, < 0, 63 > 0, and 64 > 0. The estimation of (A. 15) and (A. 16) for 1952-80 by ordinary least squares (OLS) yielded the following results, where t-statistics are in parentheses, D.W. refers to the Durbin-Watson statistic, and R2 is the coefficient of correlation. The results for (A.15) are: (A.17) , 0.005 + 0.636 M,-, - 0.128 A,-2 + 0.139 A,- (-0.164) (3.608) (-0.720) (2.276) + 0.219 DEH, + 0.072 fit (3.325) (1.997) R = 0.741 D.W. = 2.232 The estimation of (A. 16) for 1952-80, using OLS, yielded: (A.18) P = 0.033 + 0.458M4, - 0.747j, + 0.200PX, + 0.022DUM,. (0.605) (2.933) (0.764) (1.871) (0.804) R2 = 0.562 D.W. = 1.854 The fit of the equations is quite satisfactory as measured by the R2. More impor- tant, the coefficients of all relevant parameters have the expected signs and are sig- nificant at conventional levels. From equation (A. 17) it can be seen that the rate of COFFEE, MONEY, AND INFLATION 175 growth of high-powered money in Colombia can be well explained by lagged rates of growth of M, by the fiscal deficit, and by changes in the price of coffee. These results show that, with other things given, an increase (decrease) in the price of coffee would cause higher (lower) rates of growth of high-powered money. The estimated coefficient is highly significant at conventional levels. Equation (A. 18), on the other hand, presents the results obtained for the inflation equation. All coefficients have the expected signs, with those corresponding to M, and PX, being significant at the conventional levels. The coefficient of M, indicates that, holding other things constant, an increase of 10 percent in high-powered money will generate a rise in the rate of inflation of approximately 5 percent. This coefficient is lower than the hypothesized unitary value for closed economies. It is perfectly consistent with the case of a semiopen economy, however, in which it can be postulated that an increase in the quantity of money will be reflected in part in prices and in part in a loss of international reserves, a change in the exchange rate, or both. The coefficient of PX, suggests that if money is held constant an increase of 10 percent in the rate of growth of the peso price of tradables-generated by a higher rate of world inflation or a faster rate of devaluation-will produce an in- crease of only 2 percent in the rate of inflation. This result, which may seem some- what surprising, is consistent with the findings by Hanson, who, using different data and a slightly shorter period, found a significant coefficient of 0.25. 14 This result should be taken with caution, however, in the analysis of policy alternatives. In particular, it should be emphasized that a faster rate of devaluation will be trans- lated into 0.2 percent higher rate of inflation only if all other variables are held constant. If, for example, this is not the case, and both the rate of devaluation and the rate of growth of high-powered money are increased 10 percent, the rate of inflation will tend to increase approximately 7 percent. In order to analyze whether the results obtained from the estimation of equation (A. 15) are sensitive to the specifications, an alternative equation was run." In this case P, was replaced with P,-the logarithm of the price of coffee in domestic cur- rency. The following result was obtained: (A.19) M, = -0.146 + 0.498M,_. - 0.637 Mt,-2 + 0.063 M,3 (-2.967) (3.608) (-4.640) (1.005) + 0.190 DEH, + 0.053 P, (2.423) (5.281) R = 0.827 D.W. = 1.429 In order to test whether this result is sensitive to the period, alternative periods were also used; the results are presented in table A-1. The results for the rate of growth of high-powered money and inflation equa- tions reported above were obtained using OLS. The reason for using OLS is that this 176 APPENDIX A Table A-1. Equations for the Rate of Growth of High-Powered Money in Different Periods 3 AMt =UO + Ol,Ait-i + 71 DEHt + 72 fic, + Ut Equation Equation Equation Equation (A.15) (A.16) (A.17) (A.18) Period 1952-70 1952-75 1952-77 1952-73 Uo -0.272 -0.131 -0.110 -0.182 (-2.096) (-1.798) (-2.034) (-2.063) at 0.405 0.433 0.428 0.448 (2.018) (2.443) (2.599) (2.446) U2 -0.722 -0.684 -0.665 -0.712 (-4.041) (-4.541) (-4.730) (-4.511) U3 0.061 0.047 0.044 0.050 (0.804) (0.682) (0.666) (0.706) 7Y 0.200 0.210 0.200 0.209 (1.978) (2.296) (2.427) (2.220) 72 0.079 0.054 0.050 0.063 (3.283) (3.921) (5.006) (3.853) D.W . 1.878 1.727 1.725 1.850 R2 0.723 0.635 0.738 0.637 L 35.02 45.09 49.60 41.20 Note. The values in parentheses refer to t-statistics; D.W. is the Durbin-Watson statistic; R2 is the coefficient of correlation; and L is the log of the likelihood function. system (equations [A. 151 and [A. 16]) is block recursive, with growth of money, M, entering the inflation equation, but with the rate of inflation not entering the rate of high-powered money equation. There is a possibility, however, that equation (A. 15) is still subject to simultaneity bias, since P, is increased in pesos, and changes in M, may affect the exchange rate. For this reason equation (A. 15) was also esti- mated using two stages of least squares. The results obtained were: (A.20) M, = -0.006 + 0.634AM,.-1 - 0.124 Mt-2 + 0.140MI- 3 (-0.117) (3.596) (-0.691) (2.293) + 0.222 DEH, + 0.079 ct- (3.354) (2.023) SEE = 0.037 D.W. = 2.238 As can be seen, the estimation technique does not affect the main finding: changes in coffee prices have affected the behavior of the creation of money in Colombia. COFFEE, MONEY, AND INFLATION 177 Conclusion In this appendix the relation between coffee and inflation in the Colombian econ- omy was analyzed, and a Dutch-disease type of model was tested. As set out in chapter 1, it is shown that changes in the price of coffee will generally tend to cause a higher rate of inflation and a lower real exchange rate. Specifically, the empirical analysis tested the link between the price of coffee, the fiscal deficit, and the rate of growth of high-powered money and also the relation between money and inflation in Colombia. The results confirm some of the main characteristics of the Dutch- disease type of model. In particular, these results indicate that, with other things given, a higher (lower) price of coffee will cause a higher (lower) rate of inflation and a lower (higher) real exchange rate, in the absence of commensurate adjust- ments in the exchange rate. Notes 1. This model is based in part on Sebastian Edwards and Masanao Aoki, "Oil Export Boom and Dutch-Disease. A Dynamic Analysis," Resources and Energy, vol. 5 (September 198 3), and on Sebastian Edwards, "Coffee, Money, and Inflation in Colombia," World Development, vol. 12 (November/De- cember 1984). 2 On this point see, for example, Arnold C. Harberger, "Dutch Disease: How Much Sickness, How Much Boon," Resources and Energy, vol. 5 (March 198 3), pp. 1-20. 3. Rudiger Dornbusch, "Tariffs and Non-traded Goods," Journal of International Economics (May 1974). 4. W. Max Corden and J. Peter Neary, "Booming Sector and De-Industrialization in a Small Open Economy," Economic Journal, vol. 92 (December 1982), pp. 825-48. 5. If externalities are present in the traded goods sector and the accumulation of foreign assets must be restricted, however, intervention may be called for; see Sweder van Wijnbergen, "Dutch Disease: A Disease after All?" Economic Journal, vol. 94, no. 373 (1984), pp. 41-55. There could also be income distribution considerations for intervention 6. Harberger, "Dutch Disease"; see also Sebastian Edwards, "Commodity Export Prices and the Real Exchange Rate in Developing Countries: Coffee in Colombia," in Macroeconomic Adjustment and Real Exchange Rates in Developing Countries, National Bureau of Economic Research Conference Volume, ed. Sebastian Edwards and Liaquat Ahamed (forthcoming). 7. The effect of a coffee price increase on the balance of payments is dB/dPc = C - PrdEr/dPc, where dET/dPc * 0. A sufficient condition for the coffee price increase to produce a dB > 0 is dET 5 0. 8. Actually dMs/dP- = C - PrdEr/dPc, and if, as assumed, (dEr/dPc)/dPc < 0, dMs/dPc > 0. 9. For an alternative diagrammatical definition of the monetary effects of a commodity export boom, see Peter Neary, "Real and Monetary Aspects of the Dutch Disease," in Structural Adjustment in Devel- oped Open Economies, ed. K. Jungenfeld and D. S Hague (London: Macmillan, 1984). 10. For an analysis of the relation between coffee prices and the rate of devaluation of the crawling peg in Colombia, see Edwards, "Commodity Export Prices." 178 APPENDIX A 11. For a fuller discussion, see Sebastian Edwards, "Coffee, Money, and Inflation in Colombia," World Development, vol. 12 (November/December 1984). Some of the results presented here are drawn from that article. 12. J. A. Ocampo and Guillermo Perry, "La reforma fiscal, 1982-1983," Coyuntura econdmica, vol. 13, no. 1 (March 1983), pp. 215-64; Sebastian Edwards, "The Short-Run Relation between Inflation and Growth in Latin America: Comment," American Economic Review, vol. 74 (June 1983), pp. 477-88. 13. Harberger, "Dutch Disease"; James Hanson, "Short-Run Macroeconomic Development and Policy in Colombia" (Washington, D.C.: World Bank, Industry Department, 1982, processed) 14. Hanson, "Short-Run Macroeconomic Development." 15. Estimation of (A. 15) using o.s may cause a simultaneity bias, since PC is in pesos and therefore depends on the world price of coffee and the exchange rate, and changes in M, can affect the exchange rate. We find, however, that the exchange rate is econometrically exogenous to M, and the problem is thus avoided. B The Exchange Rate and Noncoffee Exports Sebastian Edwards IN A NUMBER of studies the determinants of noncoffee exports in Colombia have been investigated econometrically.' The results obtained have generally supported the hypothesis that both the real exchange rate-or the domestic relative price of noncoffee exports-and the level of world economic activity have been important in determining the volume of noncoffee, or minor, exports. As can be seen from table B-1, these studies have generated a wide range of values for the relevant elas- ticities. The elasticities presented in this table, however, have been obtained using different methods and different specifications of the noncoffee exports function, and in that sense, the results are not directly comparable and should be interpreted as providing only approximate orders of magnitude. One of the purposes of the results presented in this appendix is to narrow the range in the previous estimates by providing our own recent results. A central purpose of this appendix is to provide policy support in the area of exchange rate management. Much confusion has surrounded the issue of whether noncoffee exports respond to changes in the real exchange rate. This appendix shows that, while there is little doubt that world economic growth and quantitative restrictions abroad seriously affect Colombia's exports, the level of the real ex- change rate also provides the crucial edge. A Simple Model to Determine Colombia's Noncoffee Exports In general the quantity of a good exported depends on relative prices and levels of economic activity in the rest of the world and in the country under consideration. It would be expected that the quantity exported depends positively on the domestic relative price of exports and on the level of economic activity in the rest of the world. On the other hand, exports may vary negatively with the level of activity in the domestic economy, if a higher domestic level of activity increases the domestic demand for exportable goods and reduces the exportable surplus of that good. With 179 180 APPENDIX B Table B-1. Estimated Elasticities of the Functions of Noncoffee Exports from Colombia Dependent Price Income Study variable elasticity elasticity Period Teigero and Elana Value of minor exports, excluding 1.34 1948-71 (1973) gold and bananas (annual) Teigero and Elan Value of manufactured exports 5.43 1966-71 (1973) (quarterly) Diaz-Alejandrob Change in value of minor exports 0.81-0.87 1955-72 (1976) (annual) Diaz-Alejandro Change in value of minor exports, 0.59 1955-70 (1976) except coffee, bananas, sugar, and (annual) tobacco Diaz-Alejandro Annual changes in value of minor 0.68-1.04 1955-72 (1976) exports (and sub- periods; quarterly) Cardonac Real minor exports 1.36 1967-76 (1976) Carrizosad Noncoffee real exports 0.57 5.34 1960-76 (1979) (annual) Echavarria' Minor real exports, excluding gold 0.90 0.91 1960-67 (1980) (annual) Echavarria Minor real exports, excluding gold 0.94 1.12 1960-67 (1980) and diamonds (annual) a. J. D. Teigero and R. A. Elan, "El crecimiento de las exportaciones menores y el sistema de fomento de exportaciones en Colombia" (Bogota: FEDESARROLLO, July 1973, processed). b. C. F. Diaz-Alejandro, Foreign Trade Regimes and Economic Development. Colombia (New York: Co- lumbia University Press for the National Bureau of Economic Research, 1976). c. Marta E. Cardona, "El crecimiento de las exportaciones menores y el sistema del fomento de exportaciones en Colombia," Revista deplaneacidny desarrollo, vol. 9, no. 2 (April-September 1977). d. Mauricio Carrizosa, "El futuro de la balanza comercial," La economia colombiana en la decada de los ochenta (Bogota: FEDESARROLLO, 1981). e. Juan Josb Echavarria, "La evoluci6n de las exportaciones colombianas y sus determinantes: Un analisis empirico," Ensayos sobrepolitica econdmica, no. 2 (September 1982), pp. 257-94. these premises, if it is further assumed that the long-run export function has a double-log form, the following reduced form for export behavior can be postu- lated:2 (B.1) log X, = ao + a, log PX, + a2 log YW + a3 log Y, + u1, where Xt = long-run volume of exports PX, = domestic relative price of exports YW, = world real level of economic activity Y = domestic real level of economic activity. EXCHANGE RATE AND NONCOFFEE EXPORTS 181 It is expected that a, > 0, a2 > 0, and a3 < 0. Since PX, is the domestic relative price of exports, it will depend on their world price in foreign currency, on the effective exchange rate, and on the peso prices of other goods: (B.2) PX, = E, (PXW,/P1, where PXW, = world price of exports (in dollar terms) E = effective nominal exchange rate-that is, it incorporates the role of export subsidies P, = peso prices of other goods. Since E, in (B.2) is the effective nominal exchange rate it will be given by (B.3) E, = e,(1 + s,), where e, is the nominal exchange rate and s, is the average export subsidy. From (B.3), (B.2), and (B. 1) it can be seen that an increase of s, with other things given, will cause a higher relative price of exports and thus a higher quantity exported. In the case of Colombia's noncoffee exports, subsidies have been quite important since 1967.1 Since that year, the exports incentives system in Colombia has been based on three major tools-Plan Vallejo, PROEXPO credit, and CAT/CERT.' Re- cently, both the implicit subsidy in PROEXPO credit and the CAT/CERT rates have been increased in an effort to compensate in part for the overvaluation of the peso. It is generally accepted that the quantity of any particular good actually exported does not adjust instantaneously to changes in its determinants.' For that reason, the reduced-form equations usually used to describe the behavior of actual exports (X,) include lagged coefficients of its determinants. k k k (B.4) log X, = ao + E 1, log PX,i + Y, log YWri + E bi log Y, + c'. 1=0 z=i=0 Under this formulation, 0o, y0, and 80 can be interpreted as short-run elasticities, while the sums of the Ois, y,s, and bis are long-run elasticities. In the next section results obtained from the estimation of equations of the type of (B.4), using Colom- bian quarterly data for 1970-81, will be presented. Estimation of the Model The first problem faced in the estimation of equation (B.4) is that of finding the appropriate data. The dependent variable is Xc, the quantity, or volume, of noncof- fee exports. Export data, however, are generally available in the form of an index of the value of exports VX,. For this reason, X, was defined as:6 (B.5) X - , PNC,' 182 APPENDIX B where PNC, is the price index of noncoffee exports. A problem with (B.5), how- ever, is that there are no directly available data for PNC,. This index was therefore constructed using data on the total export price index PXT, and on a coffee exports price index PC,, which are available from IFS. Assuming that PXT, has a Cobb- Douglas form PXT, = PC" PNC', PNC, can be computed as: (B.6) PNC, = exp(1 - a)-' [log PXT, - a log PC]. In the actual computation of PNC,, both PNT, and PC, were obtained from IFS. With respect to a it was considered to be variable-that is, c, varies for each t-and in each period it was taken to be equal to that period's ratio of the value of coffee exports to total exports. The relative price variable, PX, was constructed as the effective nominal rate of the peso to the U.S. dollar times the U.S. wei, divided by the Colombian cpi. In that sense, PX can be interpreted as being a measure of the real exchange rate. The rest of the world level of activity, YW, was proxied by the U.S. real GNP, which was taken from IFS. The domestic real level of activity, on the other hand, was defined as domestic real GNP, using data from Montes and Can- delo, which were supplemented for the recent years by the DNP. Equation (B.4) was estimated using polynomial distributed lags-that is, Almon Table B-2. Noncoffee Exports Function: Almon Lags, Quarterly Data, 1971-81 (third-degree polynomial, no end constraints) Lag (i) Constant log REX,_ log YW'_' log y-, 0 -123.063 1.331 2.676 -0.115 (-2.584) (2.503) (1.977) (-0.118) 1 0.690 2.236 -0.661 (2.774) (2.765) (- 1.283) 2 0.217 1.794 -0.981 (1.240) (2.260) (-2.380) 3 -0.088 1.349 -1.075 (-0.378) (1.494) (-2.198) 4 -0.227 0.902 -0.943 (-0.920) (1.064) (-1.949) 5 -0.197 0.452 -0.585 (- 1.157) (-0.816) (-1.792) Sum of lagged coefficients 1.725 9.409 -4.360 D.W. 1.552 R' 20.329 SEE 0.180 Note: Numbers in parentheses are t-statistics; SEE is the standard error of estimate, D.W. is the Durbin- Watson statistic. EXCHANGE RATE AND NONCOFFEE EXPORTS 183 lags.' A problem usually faced when Almon lags are used is that it is not possible to know a priori the appropriate order of the polynomial or the constraints to be imposed on its form. For this reason, and in order to check for the robustness of the empirical results, alternative combinations of the polynomial degree and the con- straints were tried.8 The length of the lag structure-that is, the value of k in equa- tion (B.4)-was varied between four and twelve quarters, and the "best" results, which are those reported here, were obtained when a six-quarters lag structure was used. Tables B-2, B-3, B-4, and B-5 contain the results obtained from the estimation of the reduced form for noncoffee exports for Colombia under alternative formula- tion of the polynomial structure. As can be seen, the results are quite satisfactory. Even though the R's are rather low, all the coefficients have the expected signs, and many of them are significant at the conventional levels. The sum of lagged coeffi- cients, moreover, was always significant for the relative price, REX, and world real income, YW, variables. They, however, were never significant for the domestic real income variable, Y. These results, then, indicate that a strong level of economic activity in the rest of the world, more specifically in the industrial countries, is crucial in order for Co- lombia's noncoffee exports to grow. From a policy perspective, however, the most important finding from these estimations is the significant effect of the real ex- change rate on the behavior of noncoffee exports. Table B-3. Noncoffee Exports Function: Almon Lags, Quarterly Data, 1971-81 (fourth-degree polynomial, no end constraints) Lag (i) Constant log REX,, log YW,-, log y, -, 0 -85.278 1.848 0.379 -0133 (-1.492) (2.136) (0.162) (-0.133) 1 0.459 3.433 -0.846 (0.973) (2.930) (-0.931) 2 -0.221 3.331 -0.942 (-0.360) (2.409) (-1.102) 3 -0.398 1.505 -0.663 (- 1.076) (1.612) (- 1.067) 4 -0.282 -0.614 -0.252 (-0.867) (-0.460) (-0.368) 5 -0.080 - 1.592 0.049 (-0.182) (-1 044) (-.072) Sum of lagged coefficients 1.327 6.442 -2.786 D.W. 1.605 R 20.380 SEE 0.180 Note: Numbers in parentheses are t-statistics. 184 APPENDIX B Table B-4. Noncoffee Exports Function: Almon Lags, Quarterly Data, 1971-81 (third-degree polynomial, far end constraint) Lag () Constant log REX,-i log YWt,- log y'; 0 -116.482 0.793 2.143 -0.528 (-2.382) (2.507) (2.479) (-0.734) 1 0.903 2.782 -0.827 (2.608) (2.735) (-1.025) 2 0.580 2.391 -0.928 (2.546) (2.765) (-1.656) 3 0.076 1.444 -0.865 (0.389) (1.653) (-1.927) 4 -0.356 0.413 -0.668 (-1.174) (0.401) (- 1.037) 5 -0.465 -0.226 -0.369 (-1.611) (-0.258) (-0.603) Sum of lagged coefficients 1.530 8.949 -4.184 D.W. 1.581 R2 0.315 SEE 0.181 Note. Numbers in parentheses are t-statistics. Table B-5. Noncoffee Exports Function: Almon Lags, Quarterly Data, 1971-81 (fourth-degree polynomial, far end constraint) Lag (i) Constant log REX,_ log YW,-i log y, 0 -68.529 2.647 1.706 -0.401 (-1.154) (2.466) (0.632) (-0.383) 1 0.079 1.815 -1.235 (0.095) (0.835) (-1.216) 2 -0.805 2.325 -0.572 (-1.048) (1.398) (-0.596) 3 -0.660 2.249 0.452 (-0.931) (1.413) (0.464) 4 -0.144 0.602 0.702 (-0.179) (0.280) (0.729) 5 0.088 -3.604 -0.958 (0.010) (-1.225) (-0.876) Sum of lagged coefficients 1.206 5.093 -2.012 D.W. 1.736 R2 0.426 SEE 0.182 Note: Numbers in parentheses are t-statistics. EXCHANGE RATE AND NONCOFFEE EXPORTS 185 Notes I See, for example, Fernando Montes, "Principales determinantes del comportamiento de la cuenta coriente durante la dcada," in Ensayos sobre polftica econdmica, vol. 2 (September 1982), pp. 187-255, andJosh Antonio Ocampo, "En defense de la continuidad del regimen," in Coyuntura econdmica, vol. 13, no. 1 (March 1983), pp. 198-214. 2. This equation can also be regarded as a reduced form for the volume of noncoffee exports. 3. See C. F Diaz-Alejandro, Foreign Trade Regimes and Economic Development (New York: Columbia University Press for the National Bureau of Economic Research, 1976). 4. The original scheme, CAT, was revised recently as CERT to provide greater flexibility in the incen- tive. 5. John F. Wilson and Wendy E. Takas, "Differential Responses to Price and Exchange Rates Influ- ences in the Foreign Trade of Selected Industrial Countries," Review ofEconomics and Statistics, vol. 61, no. 2 (May 1979), pp. 267-79; and Morris Goldstein and Mohsin Khan, "Income and Price Effects in Foreign Trade" (Washington, D.C.: International Monetary Fund, 1983, processed). 6. Edward Leamer and Richard Stern, Quantitative International Economics (Boston: Allyn and Bacon, 1970); and Gabriel Montes and Ricardo Candelo, "El crecimiento industrial y la generaci6n de empleo in Colombia," Revista deplaneacidny desarrollo, vol. 12, nos. I and 2 (January-June, 1981). 7. See Goldstein and Khan, "Income and Price Effects," for a discussion on lagged representation in international trade empirical analyses. 8. Gabriel Montes and Ricardo Candelo, "El enfoque monetario de la balanza de pagos: El caso de Colombia, 1968-1980," Revista de planeaci6n y desarrollo, vol. 14, no 2 (May-August 1982), pp. 11-40. C The Rate of Devaluation and the Nominal Interest Rate Sebastian Edwards COLOMBIA'S IS a semiopen economy with a growing, but still partially repressed, capital market.' This appendix derives a model for interest rate determination in a small semiopen economy, and tests it empirically, using quarterly data for the pe- riod 1968-82. The model addresses the relation between the rate of devaluation and the nominal interest rate, which is central to the devaluation issue discussed in chapter 2; the model is also helpful in determining the effect of changes in monetary policy on the rate of interest. Interest Rate, Rate of Devaluation, and Money In a fully open economy, where economic agents are risk neutral and foreign and domestic bonds are perfect substitutes, the internal and external interest rates are rigidly linked through the interest parity condition:' (C.1) it = i' + D' where i, = domestic nominal interest rate i' = foreign (world) nominal interest rate, on instruments that have the same maturity as the domestic papers D = expected rate of devaluation of the domestic currency between period t and the period corresponding to the maturity of the corresponding fi- nancial instruments. The subscript t indicates that this expectation is formed in period t. If in the economy in question there are no impediments to capital movements, equation (C. 1) will tend to hold in both the short run and the long run. The avail- able empirical evidence suggests that a slightly revised version of equation (C. 1)- which replaces D' by the forward premium, incorporates transaction costs, and considers foreign interest rates-holds closely for the case of industrial countries. 186 DEVALUATION AND NOMINAL INTEREST RATE 187 In the case of semiopen or closed economies, however, expression (C. 1) does not seem to hold. The recent experience of the Southern Cone countries-Argentina, Chile, and Uruguay-and of Colombia suggests that in semi-industrialized, se- miopen economies the divergencies from (C. 1) can be substantial.' Equation (C. 1) can be modified in several ways, in order to incorporate the fea- tures of a semiopen economy. In particular, it is possible to write an expression that indicates that the domestic interest rate tends to equate the world rate of interest plus the rate of devaluation and a risk premium in the long run, but that it can differ from it in the short run. First define i, as (C.2) i * = i' + D' + , where /, is a risk-premium term.5 Equation (C. 1) can then be replaced by the fol- lowing expression: (C.3) Ai, = (i - ,), where 0 < 0 < 1. This equation states that movements of the domestic nominal interest rate will respond to discrepancies between i, and the domestic rate in the preceding period. According to (C. 3), in the long run the domestic interest rate will be equal to the foreign rate plus the expected devaluation and the risk premium. In the short run, however, these two rates, i, and it, can differ. The coefficient 0 is a measure of the speed at which discrepancies between i, and i, I will tend to be corrected. For example, if it only takes one period for these interest-rate differen- tials to disappear, 0 would be equal to 1.0. In a semiopen economy, where capital movements are subject to a number of controls, domestic monetary policy could also have some effect on the short-run behavior of the interest rate.' Specifically, it can be postulated that disequilibria in the money market will have an effect on interest-rate movements, with situations of excess liquidity-an excess supply of money-driving the interest rate down- that is, a liquidity effect-and with excess demands for money causing an increase in the domestic interest rate. This possible effect can be captured by the following expression: (C.4) Ai, = 0(i. - i,-,) - X(log m, - log m_ d), where m, is the real quantity of money in t, and where m I is the quantity of money demanded in period t - 1. This equation explicitly allows for internal monetary disequilibria to affect interest rate movements. The parameter X mea- sures the importance of these disequilibria, and the negative sign reflects the hy- pothesis that an excess supply (demand) for real money will generate a decline (in- crease) in the interest rate. An alternative way to write the monetary disequilibrium term would include the contemporaneous value of both the quantity of money and the quantity demanded. In this case the interest-rate equation can be rewritten as (C.5) Ai, = 0(i* - i,-,) - X(log m, - log md). 188 APPENDIX C Estimation In this section results obtained from the estimation of reduced forms for equa- tions (C.3) through (C.5) for Colombia, using quarterly data for the period 1968- 82, are presented. In order to simplify the analysis it is assumed that D, is equal to the actual annualized rate of devaluation in quarter t. This is a plausible assump- tion, since during the period under consideration Colombia followed a crawling peg exchange-rate policy, where the rate of the crawl was altered fairly slowly.' On the other hand, regarding the risk premium (0,) it was assumed that it can be repre- sented as a constant plus a random element. Regarding the demand for money function, it was assumed that it has a conven- tional Cagan form: (C.6) log md = bo + b, logy, - b2 I fory, = real income. Estimation of Equation (C. 3) Equation (C.3) can be rewritten in the following form (where e, is an error term with the usual characteristics): (C.7) it = o + Ceit + C2t-1 + Et. Notice that since a, = 0 and U2 = 1 - 0, 0 is overidentified. Equation (C.7), however, was run without imposing the constraint a = 1 - C2. The result ob- tained was the following, where the values in parentheses are t-statistics. (C.8) it = 0.011 + 0.321 it* + 0.765 i,-, (0.484) (1.671) (7.261) R2 = 0.768 D.W. = 2.422 The coefficients of it and i-_ are significant at the conventional levels. As can be seen, the direct estimate of 0 is 0.32 1, indicating that approximately a third of the discrepancy between it and i,-, is eliminated in one quarter. This means that after one year an interest rate differential of 10 percentage points would be reduced to 2.1 percentage points. This coefficient can also be interpreted as a measure of the effects of an increase on the rate of devaluation or the interest rate. The indirect estimation of 0-as I minus the coefficient of i,- I -gives a value of 0.2 35, suggesting a slightly slower speed at which discrepancies between it and i, I will be eliminated. DEVALUATION AND NOMINAL INTEREST RATE 189 Estimation of Equation (C. 4) Combining (C.6) and (C.4), and adding an error term we, the following reduced form of equation (C.4) can be written: (C.9) it 70 + li* +72 t-1 + 731OgMt + 74Ogyt-I + Wt, where it is expected that 7, > 0, 72 > 0, Y3 < 0, and y, > 0. The expressions for the -ys in terms of the structural equations (C.4) and (C.6) parameters are -Y = 0 7 = I - 0 + Xb2 73 = _ 74 1 X The estimation of (C.9) using OLS yielded the following result for the period 1968:3-1982:4 (t-statistics in parentheses): (C.10) i, = -0.489 + 0.404 i * + 0.383i 3 i, (-1.990) (2.337) (2.847) = 0.815 -0.275 logim, + 0.379 logy,_l D.W. = 2.211 (-1.963) (3.539) N = 58 This result is quite satisfactory, with all the coefficients significant at conventional levels and having the expected signs. The estimated structure coefficients computed from (C. 10) turn out to be the following: = 0.404 X = 0.275 d, = 1.378 d2 = 0.775. As can be seen, the estimated parameters for the demand for money in Colombia are within the plausible range of values.' Also, these results indicate that after one quarter, 40 percent of a unitary uncovered interest-rate differential will have been corrected. After one year, 87 percent of this discrepancy will have been corrected. The coefficient of it, 0, can also be used to simulate the effect of an increase in the rate of devaluation on the interest rate. Assume that the initial-in period 0- domestic interest rate is 40 percent and that the rate of devaluation is 22 percent a year. Assume now that in period 1 the rate of devaluation is increased to 32 per- cent, and maintained at this higher level, with all the rest of the relevant variables remaining constant. The evolution of the domestic interest rate under this case, 190 APPENDIX C using the estimated parameters from equation (C.10), is given in table C-i :1 after six quarters the domestic rate of interest has practically reached its new equilibrium. Equation (C. 10) also provides some information regarding the effect of the quan- tity of money on interest-rate behavior-specifically, this estimate provides a se- mielasticity of the interest rates with respect to real money of - 0.2 75. The corre- sponding elasticity, of course, will be variable and will depend on the level of the interest rate. In table C-2 the corresponding elasticities for some initial values of the nominal interest rate are given. From table C-2 it can be seen that for the case of a nominal interest rate of 40 percent the corresponding elasticity will be -0.688, indicating that, with other things given, an increase of 10 percent in the real quantity of money will reduce the nominal interest rate by 6.9 percent. According to our model, however, in order to reduce the interest rate it is necessary to increase the real quantity of money, requir- Table C-1. Simulation of the Effect of a Higher Rate of Devaluation of the Crawling Peg on the Domestic Interest Rate (percent) Nominal domestic interest rate Rate of Quarter (i) devaluation 0 40.0 22 1 44.0 32 2 46.5 32 3 47.9 32 4 48.7 32 5 49.2 32 6 49.5 32 7 49.7 32 8 49.8 32 Source: See text. Table C-2. Elasticity of the Interest Rate with Respect to Real Money Interest rate Interest rate elasticity (percent) with respect to real money 30 -0.917 35 -0.786 40 -0.688 45 -0.611 50 -0.550 Source. See text. DEVALUATION AND NOMINAL INTEREST RATE 191 ing that an increase in the rate of growth of nominal money not be matched by higher equiproportional inflation. In order to investigate the level of significance of the structural coefficients from the demand for money, equation (C.9) was estimated, using a nonlinear least squares procedure that imposes the respective restrictions across coefficients. The following results were obtained: a, had an estimated value of 1.380, with a t-statis- tic of 1.963, and a2 was estimated to be 0.773, with a t-statistic of 1.368. Estimation of Equation (C. 5) The reduced form of equation (C.5), with an error term, v, added, has the fol- lowing form: (C. 11) i, = 6+ i+ 6-+ log m, + 64 logy, + v,. This expression differs from (C. 10) in that log y now enters contemporaneously. The interpretation of the 6s in terms of the structural parameters, however, is quite different. 66- 1 + Xa2 I 1 Xa2 1-0 Xa 1 + Xa2 I+Na2 In this case it is expected, as before, that 6, > 0, 62 > 0, 63 < 0, and 6, > 0. The estimation of (C. 11), using OLS for the period 1968:3 to 1982:4, generated the following result, where the numbers in parentheses are the t-statistics: (C.12) it = -0.434 + 0.402 it* + 0.363 i1-, - 0.389 logm, + 0.171 logy (-1.832) (2.405) (2.845) (-2.536) (4.171) R2 = 0.840 D.W. = 2.112 N = 58 As can be seen, once again all the coefficients have the expected signs, and now their level of significance is even higher than before. The computed structural pa- rameters are = 0.525 N = 0.298 d, = 1.175 d2 = 0.785. 192 APPENDIX C As can be seen, these numbers are quite similar to those obtained from the esti- mation of equation (C. 10). Now, however, the speed at which discrepancies be- tween i, and i- I are eliminated is faster. Actually, these results indicate that, with other things given, in one quarter more than half of a unitary interest-rate differen- tial will be corrected. An increase in the rate of devaluation of the crawling peg of 10 percentage points-that is, from 22 percent to 32 percent-will produce, in the first quarter, an increase in the domestic interest rate of 5.3 percentage points. After two quarters the increase would have been 7.7 percentage points and after one year 9.5 percentage points. Regarding the estimated coefficient of log m, (- 0.298), it indicates that with a nominal interest rate of 3 5 percent, the elasticity of the interest rate with respect to the real quantity of money will be equal to - 0.851. This means that, with other things given, an increase of 10 percent in the real quantity of money will tend in the short run to reduce the interest rate to 26.5 percent. Forecasting the Interest Rate In order to compare further the relative merits of the three interest rate models tested [equations (C.3) through (C.5)], their forecasting properties are analyzed by reestimating the models for a shorter period, 1968:3-1980:4, and by using the estimated coefficients to make out-of-sample estimates for the seven quarters 1981:1 to 1982:4. Table C-3 presents the actual values of the interest for this pe- riod, and the forecast values obtained from each equation. Table C-4, on the other hand, presents a number of statistics that measure the degree of accuracy of these forecasts. As can be seen in table C-3, for many of the quarters involved the interest Table C-3. Actual and Out-of-Sample Forecast Values of the Interest Rate in Colombia, 1981:2-1982:4 (percent) Year and Equation (C. 3) Equation (C. 4) Equation (C. 5) quarter Actual forecast forecast forecast 1981:1 36.7 - - - 1981:2 60.9 35 4 40.1 406 1981 3 48.6 54 5 48.4 50.0 1981:4 63.7 45 1 43.7 43.8 1982:1 65.7 56.3 45.4 46.2 1982 2 49.5 57.7 48 7 50.0 1982 3 53.8 450 44.4 45.7 1982.4 57.2 49.0 44.8 44.9 - Not applicable. Source: See text. DEVALUATION AND NOMINAL INTEREST RATE 193 Table C-4. Summary Statistics for Comparison of Actual and Predicted Interest-Rate Series: Out-of-Sample Forecasts, 1981:2-1982:4 Equation Equation Equation Statistic (C. 3) (C. 4) (C. 5) Correlation coefficient between actual and predicted 0.604 0.671 0.656 Root mean square error 0 183 0 188 0 186 Mean absolute error 0.152 0.151 0148 Mean error 0 116 0.151 0.144 Trheils U-statistic 0.180 0.193 0.189 Fraction of error caused by bias 0.404 0.641 0 601 Fraction of error caused by different variation 0.225 0.109 0 124 Fraction of error caused by different covariation 0 371 0 250 0 276 Source. See text. rates forecast are quite different from the actual values. It is important to consider that this is an out-of-sample experiment, however, and that during the period through which the forecast was done interest rates were particularly volatile." In order to have a more systematic evaluation of the statistical quality of these forecasts, table C-4 provides some summary statistics from the comparison of actual and forecast values, which indicate that these forecasts are quite satisfactory. The coefficients of correlation between actual and predicted series are fairly high, with the mean errors and Theil's inequality coefficient being on the low side. From these results, however, it is not straightforward to determine which equation provides better forecasts. While some statistics-root mean square error, mean-error, and Theil's inequality coefficient-suggest that equation (C.3), which excludes mone- tary considerations, does a better job of forecasting the interest rate, other statis- tics-mean absolute error, and the correlation coefficient-point toward equations (C.4) and (C.5), respectively, as providing better forecasts. Conclusion For this analysis it was recognized that Colombia's is a semiopen economy and that, as a consequence, open-economy models and closed-economy models are in- appropriate. Three alternative formulations for the determination of the interest rate in a semiopen economy were developed and tested, using quarterly data for the period 1968-82. The results obtained were remarkably good and indicated that the domestic (nominal) interest rate will tend to converge slowly through time toward the world interest rate plus expected devaluation. The estimates indicate that in one 194 APPENDIX C quarter, between a third and half of a unitary discrepancy between the domestic rate and the world rate plus the expected rate of devaluation will be corrected. In six quarters an acceleration of the rate of devaluation of the crawling peg will be almost completely translated into an equivalent increase in the domestic rate of interest. An excess supply of real money will exercise significant negative pressures on the nominal interest rate-that is, there will be a liquidity effect. Finally, out-of-sample forecasts were presented, using the three alternative formulations. The results showed that despite being out of sample, the forecasts were quite satisfactory. Notes 1. On the behavior of the Colombian capital market see World Bank, Colombia: Economic Develop- ment and Policy under Changing Conditions (Washington, D.C., 1984), chapter 5, and Juan Carlos Jaramillo, "El proceso de liberaci6n del mercado financiero colombiano," in Ensayos sobre politica econ6mica, no. I (March 1982), pp. 7-19. 2. This expression abstracts from consideration of taxation. 3. SeeJacob A. Frenkel and Richard M. Levich, "Covered Interest Arbitrage: Unexploited Profits" Journal ofPolitical Economy, vol. 83, no. 2 (April 1975), pp. 325-38; and "Transaction Costs and Inter- est Arbitrage: Tranquil versus Turbulent Periods," Journal ofPolitical Economy, vol. 85, no. 6 (December 1977), pp. 1209-26. 4. See Sebastian Edwards, "Stabilization with Liberalization: An Evaluation of the Years of Chile's Experiment with Free Market Policies, 197 3-8 3," Economic Development and Cultural Change, vol. 3 3 (January 1985). 5. On the existence of a risk premium in interest arbitrage equations see, for example, Lars Peter Hansen and Robert J. Hodrick, "Forward Exchange Rates as Optimal Predictors of Future Spot Rates: An Economic Analysis," Journal of Political Economy, vol. 88, no. 5 (October 1980), pp. 829-53 6. On controls of capital movement in Colombia, see recent issues of International Monetary Fund, Annual Report on Exchange Arrangements and Exchange Restrictions (Washington, D.C.). 7. In order to check the extent to which past rates of devaluation predicted the actual rate of devalua- tion, a regression of the following form, using quarterly data, was run: Dt = a 1 D, I + a2 D, - 2 + u,. For 1968-82 the following result, which indicates that the assumption D, = D, is a fairly good one, was obtained (t-statistics in parentheses): D, = 0.928 D, + 0.001 Dt-2 (15.318) (0.010) D.W. = 2.2 8. All the data were obtained from Gabriel Montes and Ricardo Candelo, "El enfoque monetario de la balanza de pagos," Revista deplaneacidny desarrollo, vol. 14, no. 2 (May-August 1982), pp. 11-40; the DNP; and the International Monetary Fund. 9. Montes and Candelo, in "El enfoque monetario," estimated that for the period 1968-80 the elasticity of the demand for money with respect to real income was 0.955 and that the interest-rate elasticity of the demand for money was -0.20. 10. In obtaining these results it was, in fact, assumed that the monetary authorities manipulate the supply of money in such a way that the money disequilibrium (log m, - log m_ ) remains constant 11. Unfortunately, out-of-sample forecasts are not usually made. In most studies, the forecasts usually reported are those made within the sample, which do not have much value. D The Stability and Predictability of Prices, Producers' Income, and Profitability John Nash IN THIS APPENDIX the way indexes were constructed for several crops to measure the stability and predictability of several economic variables during the period 1970-8 1 (see chapter 5) will be described. The crops are barley, beans, coffee, corn, cotton, potatoes, rice, sugar, and wheat. The variables are the international price, the do- mestic producer price, the domestic consumer price, producers' income, the profit per ton, and the profit per hectare planted-or, for some crops, gross income per hectare. All variables are in real terms of the 1975 peso adjusted by the implicit price deflator of the gross internal product. The methods by which they were com- puted and the method of computing the indexes will be described; then the results will be presented (table D-1) and discussed. Data Sources The international price for each product in each year is the implicit import- export price, that is, it is the total value in pesos of the imports or exports of the product divided by the quantity imported or exported. The prices were provided by the Sociedad de Agricultores de Colombia. For rice, two years of missing data were constructed by taking the preceding year's price and adjusting it in such a way that the price moved by the same percentage as did rice prices on world markets. (This information was taken from International Financial Statistics.) The implicit export price for sugar was for processed sugar. Since it was necessary to make this comparable to producer prices for sugarcane, the export price of sugar each year was adjusted by the overall average percentage markup from cane to processed sugar to give some indication of the way sugarcane prices would have moved had they been governed by movements in international prices. The implicit interna- tional price of beans is not available. 195 196 APPENDIX D Producer prices were taken from a DNP working paper, a Banco de la Republica source, and, for coffee, a FEDERACAFE publication, "Boletin de informaci6n estatis- tica sobre el cafe," no. 48. Consumer prices were taken from a DNP-UEA working paper of February 28, 1983, "Series de precios del sector agropecuario: 1950- 1982." Consumer prices of barley, coffee, and cotton were not available. Output was taken from table 7-1 of the statistical appendix to Colombia: Economic Develop- ment and Policy under Cbanging Conditions, by Jose B. Sokol and others (Washing- ton, D. C.: World Bank, 1984). Yield per hectare was taken from a DNP document, "Indicatores fisicas nacionales del sector agropecuario, 1950-1981." For corn, cot- ton, rice, and wheat, real production costs (per ton) were taken from table 4-11 of "Aspects of Agricultural Development in Colombia," by Jorge Garcia-Garcia (Bo- gotA, April 198 3, processed), a paper prepared for this book. Producer income at domestic prices was computed by multiplying the producer price by output. Like- wise, producer income at international prices was computed by multiplying the international price by output. For the four crops for which production costs per ton were available, the profit per ton at domestic and international prices was com- puted by subtracting the cost from the appropriate price each year. The gross income per hectare at domestic and international prices was computed by multiplying the appropriate price by the yield (in tons per hectare) each year. The profit per hectare at domestic and international prices was computed by multi- plying the profit per ton by the yield. Method of Calculation of the Indexes After the series were computed for each crop, two indexes were calculated for each of the following series for each crop: international price, producer price, con- sumer price, profit per ton at domestic prices, profit per ton at international prices, producer income, and profit per hectare or, for crops for which production cost was not available, gross income per hectare. Index 1 is an index of variability or instability. It is simply the standard error of a linear least-squares regression of the series, using time as the independent variable (to remove any secular trend). For the price series and the profit per ton series, the standard error was divided by the mean price in order to transform it into percent- age terms and make the indexes comparable across crops. After all, an average $1,000 yearly change in the price of coffee, with a price of around $89,000 a ton, would indicate much less instability than the same average change in the price of barley, with a price of around $ 17,000 a ton; to be comparable, the indexes should be in percentage terms. For the same reason, the indexes for the series of producer income were computed by dividing the standard error by the respective means. This was not done for the series profit per hectare and gross income per hectare. The reason is that the indexes from these series were designed to measure the insta- PRICES, PRODUCERS INCOME, AND PROFITABILITY 197 Table D- 1. Indexes of Instability and Unpredictability Economic variable and crop Index I Index 2 1. International price Barley 0 223 0 252 Coffee 0.347 0.371 Corn 0.218 0.286 Cotton 0.163 0.227 Potatoes 0.261 0.377 Rice 0 277 0.388 Sugar 0 692 0.658 Wheat 0.382 0 339 2. Producer price Barley 0.146 0.221 Beans 0.143 0.240 Coffee 0.219 0.286 Corn 0 118 0 277 Cotton 0.167 0.259 Potatoes 0.204 0.342 Rice 0.092 0.305 Sugar 0.081 0.229 Wheat 0.165 0.251 3. Ratio, 1:2 Barley 1.53 1.14 Coffee 1.59 1.30 Corn 1.85 1 03 Cotton 0 97 0 88 Potatoes 1 28 1 10 Rice 3.01 1.27 Sugar 8 55 2.87 Wheat 2 32 1 35 4. Consumer price Beans 0.103 0 238 Corn 0.119 0.261 Potatoes 0.144 0 308 Rice 0.084 0.252 Sugar 0.221 0 399 Wheat 0.161 0.258 5. Producer income at international prices Barley 0.341 0409 Coffee 0407 0415 Corn 0.172 0.315 Cotton 0.265 0.316 Potatoes 0 302 0.465 Rice 0 393 0485 Sugar 0.712 0.701 Wheat 0.423 0.500 (Table continues on tbe following page.) 198 APPENDIX D Table D- 1 (continued) Economic variable and crop Index I Index 2 6. Producer income at domestic prices Barley 0.286 0.386 Beans 0.277 0.218 Coffee 0.262 0 295 Corn 0.085 0.237 Cotton 0.294 0 366 Potatoes 0.179 0.332 Rice 0.163 0 239 Sugar 0.066 0.210 Wheat 0.290 0.478 7. Ratio, 5:6 Barley 1.19 1 06 Coffee 1.56 1 41 Corn 2.03 1 33 Cotton 0.90 0.86 Potatoes 1 68 1.40 Rice 2.42 2.03 Sugar 10.79 3.34 Wheat 1.46 1.05 8. Profit per ton at international prices Corn 0.397 0.396 Cotton 0.221 0.252 Rice 0.319 0.391 Wheat 0.444 0.395 9. Profit per ton at domestic prices Beans 0.162 0.193 Corn 0.225 0.225 Cotton 0.228 0.308 Rice 0.159 0.168 Wheat 0.192 0.181 10. Ratio, 8:9 Corn 1.77 1.76 Cotton 0.97 0.82 Rice 2.01 2.33 Wheat 2.32 2.18 11. Gross income per bectare at international prices Barley 1245.0 2115.2 Coffee 11240.0 12451.8 Potatoes 11970.0 19675.7 Sugar 1726.0 1636.1 12. Gross income per beccare at domestic prices Barley 1012.0 2003.9 Coffee 3065.0 4066.4 PRICES, PRODUCERS INCOME, AND PROFITABILITY 199 Table D- 1 (continued) Economic variable and crop Index I Index 2 Potatoes 6663 0 11223.4 Sugar 224.6 587 1 13. Ratio, 11: 12 Barley 1.23 1.06 Coffee 3.67 3.06 Potatoes 1.80 1.75 Sugar 7.68 2.79 14. Profit per hectare at international prices Corn 1601.0 1597.1 Cotton 12410.0 14494.2 Rice 2442.0 2993.0 Wheat 1857.0 1670.8 15. Profit per hectare at domestic prices Beans 1503.0 1853.2 Corn 1221.0 1243.0 Cotton 9778 0 13214.7 Rice 2599.0 2706.5 Wheat 1056.0 996.7 16. Ratio, 14:15 Corn 1.31 1.28 Cotton 1.27 1 10 Rice 094 1.11 Wheat 1.76 1.68 Sources: See text. bility of return on investment, the investment being in a hectare of land. The return on investment is the profit (or gross income) divided by the cost of the investment (the implicit rental value of the land). Since the implicit rental value of the land itself is not dependent on the crop planted, it would be the same for each crop. Dividing each crop's standard error by the same number would not change the ordering of the indexes so there is really no reason to do so. Index 2 is designed to measure uncertainty or unpredictability. It is important to draw the distinction between instability and unpredictability, since it is concep- tually possible that a variable-price, for example-would be quite unstable but perfectly predictable. If so, the instability would create no risk, in the sense of un- certainty, though it might create other problems, such as destabilization of macro- economic variables. Thus, both indexes are potentially important, each for analyz- ing a different kind of issue. Index 2 was computed as follows. After each series was de-trended by a linear regression against time, the residuals were taken and fitted to a first-order auto- 200 APPENDIX D regressive process of the form X, = XX, + e+ , where e, is "white noise." Index 2 is the standard error of this regression, divided where appropriate by the mean in order to transform it to percentage terms. This index represents the average abso- lute size of the prediction error involved in predicting one year's value of the vari- able from the preceding year's value. It is thus a measure of the degree to which each series is unpredictable. The results are reported in table D-1. That domestic prices are in general more stable than international can be seen in the table, where the ratio of international price instability to domestic price instability can be seen to be greater than unity for most crops. For potatoes, a crop with relatively little intervention, the ratio is rather low, providing some evidence that the government's intervention programs may be a factor in stabilizing price. This cannot be said about some of the other variables, however. In table D- 1, for example, the variability of producer income for potatoes at international prices is larger in relation to variability at domestic prices than is this ratio for several other crops. It is not clear, therefore, that government stabilization programs have stabilized incomes. E The Welfare Cost of Price Stabilization John Nash IN THIS APPENDIX the origin of the efficiency cost estimates of price stabilization, which are presented in table 5-2, will be explained. To illustrate the methodology, the explanation will be phrased in terms of a simple model of an export good whose price in the world market assumes only two values, P, or P2 (P P2), each with probability of 0.5, and whose domestic producer price is stabilized at the mean value, P, by means of a tax-subsidy scheme devised so that the average protection is 0; that is, when the world price is P, there is a tax of P, - Pon the export; when world price is P2, there is a subsidy of P - P2. The results can easily be extended to an import good, a good with multiple possible prices, and a good with a rate of protection which differs from 0, either positively or negatively. In the explanation a linear supply schedule is assumed. To derive the formula exactly, this must be true, at least locally. Consider figure E-1. With a price stabilization scheme, since producers always receive price P, they always produce quantity Q. When the world price is P, the government receives area A in export taxes; when the world price is P2, the govern- ment gives subsidies equal to C + D. When the world price is P,, exporters forgo a producer surplus increase of A + B by selling only quantity Q at a price P. But area A is not a welfare loss to the country because it goes to the government in taxes. The welfare loss from maintaining the producer price at P is area B. Area B is a triangle whose area is /2 (PI - P) (Qj Q). The quantity Q, - Q can be ex- pressed as dQ/dP (P, - P), so area B =/2 (PI - P)2(dQ/dP)E, where E is the export supply elasticity. By the same kind of logic, the welfare loss to the economy from maintaining an internal price of P when the world price is P2 is area D, which is Y12 (P2 - P)2(Q/P) E. So, the average yearly loss is V/2 E(Q/P) var (P), where var (P) is the variance of the world price. By definition, the variance is the average of (P, - P)2 and (P2 - p)2. Also, by similar logic, the welfare loss from stabilization of the price of an im- ported good can be shown to be /2 IN| (Q/P) var (P), where INI is the absolute value of import demand elasticity. 201 202 APPENDIX E Figure E-1. The Effect of Price Stabilization P, A P C P, Q2 QQ To give some idea of the magnitude of such welfare losses, the values of Q, P, var (P) and 12 E(Q/P) var (P) or /2 NJ (Q/P) var (P) are reported in table E-1 for a small sample of crops. The import and export elasticities are computed from esti- mates of domestic elasticities of demand and short-run supply and are thus the elasticities that would prevail in a market with no governmental interference in free trade. The estimates of domestic elasticities were taken from the results of the back- ground study for an article on nutrition in Colombia.' Table E- 1. Annual Welfare Losses from Price Stabilization (1975 pesos) Q P Crop (M T) ($/MT) var(P) N - NI(Q/P)var(P) E 4E(Q/P)var P Import crops Wheat 364,167 3,471 1,642,000 -0.69 59,434,446 - - Corn 57,125 3,072 541,900 - 12.08 60,864,084 - - Barley 50,125 3,898 710,200 -0.19 867,595 - - Export crops Rice 22,467 7,660 4,093,000 - - 38 60 231,694,311 Cotton 50.317 32,961 26,950,000 - - 5.28 108,611,811 Potatoes 6.708 4,571 2,299,000 - - 74.87 126,298,630 - Not applicable. Note, Q, quantity; P, price; N, absolute value of import demand elasticity; E, export supply elasticity Source: Author's estimates. THE WELFARE COST OF PRICE STABILIZATION 203 Note 1. Per Pinstrup-Andersen. Norha Ruiz de Londofio, and Edward Hoover, "The Impact of Increasing Food Supply on Human Nutrition: Implications for Commodity Priorities in Agricultural Research and Policy," American Journal of Agrincultural Economics, vol. 58 (May 1976), pp. 13 1-42. F The Organization and Management of the Coffee Economy John Nash FOR MORE THAN FIFTY years the National Federation of Coffee Growers (hereafter referred to as the federation or FEDERACAFE), a private nonprofit association of cof- fee producers that engages in commercial activities, has been the main body charged by the government with administering coffee policy.' The federation is responsible for the management of the National Coffee Fund (NcF), for the provi- sion of technical assistance to the industry, for the control of domestic and export marketing, and for advice on the setting of certain rates of taxation and prices which affect the industry. The relationship between the government and the feder- ation has been controlled since 1928 by a series of contracts that set out the duties to be delegated to the latter and the remuneration that it will receive in return for its services. The most recent of these covers the ten-year period that began December 31, 1978. Although the federation is allowed considerable freedom of action in running the coffee industry, the government can control its operations in a number of ways. First, the budget of the federation is subject to the approval of the government and, in addition, under the present contract, the federation submits to the government quarterly financial projections. Second, the appointment of the general manager of the federation is subject to the approval of the president. Third, whenever it sees fit the government can convene an extraordinary meeting of the National Congress of Coffee Growers, the supreme authority of the federation. Ministers can present the views of the government to the Congress, although they have no power to vote, and certain major decisions of the Congress are subject to the approval of the presi- dent. Finally, under the present contract the government and the representatives of coffee growers have equal representation on the important National Committee of Coffee Growers, which executes the decisions of the Congress. In addition to this control of the federation, the government determines the rate of taxes in the coun- try, including those specific to the coffee industry, and has a majority in the com- mittee that determines the price at which the federation purchases coffee from growers. 204 ORGANIZATION OF THE COFFEE ECONOMY 205 Apart from administering the NCF and supervising the marketing of the crop, the main activities of the federation are carried out by the departmental committees. In addition to the taxes set aside for specific campaigns, the committees obtain income from their own assets, from various government departments, and from the local community. The activities of the committees are diverse, ranging from the provi- sion of extension services and technical education to improvement of the infrastruc- ture in coffee-growing areas and the provision of social services. The committees, therefore, benefit the community at large as well as growers of coffee and, to the extent that the resources of the committees are drawn from taxation of the coffee sector, their activities bring about a redistribution of income from this sector to the rest of the economy. The NCF was originally established, in 1940, to finance the surplus stocks ex- pected to accumulate as a result of the international export quota arrangement in- troduced in that year. Throughout the life of the fund its administration has been delegated to the federation. With the passage of time the functions of the fund have increased in scope so that it has become the main instrument for regulating the supply and the price of coffee. It has also become an important investor in activities related to the production of and trade in coffee. Coffee Marketing and Export Usually coffee is partially processed on the farm, then sold in the form of dried "parchment." The dried parchment is brought to the nearest village or town, where it is sold either at one of the 500 purchasing points of the federation or to commercial buyers, such as exporters and dealers, who subsequently sell to the federation or to private exporters. The parchment is then bulked and transported to the nearest depot or mill of the federation or the exporter concerned, as the case may be. The federation guarantees to purchase parchment coffee delivered to its agents at the same price throughout the country, provided that the parchment is of federa- tion type-that is, of quality higher than a given standard. This price, hereafter referred to as the minimum price, is established by a committee that includes the ministers of agriculture and finance and the general manager of the federation. Coffee is exported both by the federation and by private traders. The latter ex- port not only coffee purchased from growers and private dealers but also coffee sold to them from the stocks of the federation. The price at which the federation sells for this purpose is set in terms of an ex-dock New York price, expressed in cents per pound, and is varied frequently. The volume of coffee exported by the federation and by private exporters is compared in table F-1 for crop years since 1969/70. All proceeds from the export of coffee must be surrendered to the Central Bank within twenty days of registration for export. After deduction of the ad valorem tax, the bank in the past exchanged these proceeds for currency exchange certifi- 206 APPENDIX F Table F-1. Exports of Coffee by the National Federation of Coffee Growers and by Private Exporters (sixty-kilograrn bags) Proportion by Private federation Crop year Federation exporters Totala (percent) 1969/70 2,963,781 3,910,284 6,874,065 43 1970/71 2,478,130 3,852,543 6,330,673 39 1971/72 2,302,435 4,184,517 6,486,952 35 1972/73 2,589,297 3,665,266 6,254,563 41 1973/74 3,226,002 4,181,856 7,407,858 44 1974/75 3,027,621 4,514,498 7,542,119 40 1975/76 1,489,127 5,533,834 7,022,961 21 1976/77 1,742,338 3,549,892 5,292,230 30 1977/78 4,811,162 2,747,072 7,558,234 64 1978/79 8,838,438 2,592,775 11,431,213 77 1979/80 11,357,071 182,814 11,539,885 98 1980/81 6,106.128 2,924,491 9,030,619 67 1981/82 5,241.000 3,749,000 8,990,000 58 1982/83 5,110,000 4,064,000 9,174,000 56 a. Official registered exports only. Source FEDERACAFE. cates, which could be converted to pesos immediately at a discount of between 6 and 15 percent or after 120 days at their full face value; recently this discount was eliminated. To ensure that the amount of foreign exchange corresponding to the actual earn- ings from exports enters the country, a minimum surrender price is set by the Mon- etary Board of the Central Bank for coffee and certain other commodities. For green coffee the minimum surrender price, the reintegro cafetero, represents the amount of foreign exchange per seventy-kilogram bag that exporters are required to surrender to the bank. As the international price varies, the amount of the reintegro is adjusted so that the reintegro payments to the bank are equal to the foreign ex- change earnings from the export of coffee. In practice there has tended to be a time lag between changes in the international price and in the reintegro. In rising markets the result has tended to be reintegros below the unit values of foreign exchange earnings, thereby allowing exporters to accumulate holdings of foreign currency and increase their margins by virtue of the fact that the export tax is based on the reintegro. When coffee prices fall, the reintegro price tends to be higher than the unit value of export earnings, and exporters have had to purchase foreign exchange on the free market in order to be able to make the full payment of the reintegro to the bank. Changes in the reintegro price, expressed in U.S. cents per pound, and the ico indicator price for Colombian Mild Arabicas ruling on the day of the change are given for the period since 1975 in table F-2. ORGANIZATION OF THE COFFEE ECONOMY 207 Table F-2. Prices and Effective Dates of Coffee Reintegro and the ICO Indicator Price for Colombian Mild Arabicas, 1975-83 Reintegro ICO indicator price for U.S. dollars mild Arabicas Effective per seventy- U.S. cents (US. cents date kilogram bag per pound per pound) 1975 July 22 117 00 75.82 1976 January 15 130.00 84.24 102.50 February 20 143 00 92.66 107.50 April 1 153.50 99.47 118 50 April 7 170.00 110.16 123.25 April 12 193.00 125.06 136.00 May 7 207.00 134.14 147.75 May 18 231 00 149.69 156.50 May 27 245.00 158.76 16800 June 8 259 25 167 99 185.00 November 29 284 65 184.45 19500 December 28 307.60 199.32 223.00 1977 February 11 331.00 214.49 236.00 February 17 354.00 229.39 249.50 February 24 376 50 243 97 275.00 February 28 423.00 274.10 304.00 March 9 440.00 285 12 309.00 March 23 457.00 296 13 325.00 April 14 477.00 309.09 334.00 May 26 466.50 302.29 290.00 June 16 415.00 268.92 229.50 July 12 376.50 243.97 241.00 July 16 361.00 233.93 245.00 August 18 313.75 203.31 202.00 1978 April 8 29000 187.92 193 00 June 16 275 00 178.20 191.50 July 17 259.00 167.83 229 50 1979 January 24 243.00 157.46 160.00 January 31 217.00 140.62 148.00 February 22 188.40 122.08 129.00 April 19 202.00 130.89 142.00 May 5 216.00 139.97 155.00 June 5 251.00 162.65 18600 (Table continues on the following page.) 208 APPENDIX F Table F-2 (continued) Reintegro ICO indicator price for U.S. dollars mild Arabicas Effective per seventy- U.S. cents (U.S. cents date kilogram bag per pound per pound) 1980 May 9 287.32 186 18 205.00 October 3 201.00 130.52 151.00 December 10 181.95 118.15 130.00 1981 April 24 186.55 121.14 141.00 May 6 201.90 13110 157.00 December 14 206.50 134.09 153.00 1982 March 13 217.25 141.07 158.00 May 24 206.50 134.09 146.00 1983 February 19 191.00 123 77 n a. October 15 195.50 126.68 n.a. November 30 204.50 132.52 n.a. n.a. Not avaiable. Source FEDERACAFE. Export Taxes and Contributions Sales of coffee by growers are subject to indirect taxes and contributions. Before receiving permission to export, an exporter of coffee must provide evidence of pay- ment of the first three of these taxes. The Retention Quota Private exporters must contribute to the NCF without compensation an amount of parchment related to the excelso to be exported. The parchment must be deliv- ered to a warehouse of Almacafe, a wholly-owned subsidiary of the federation. Exports by the federation are made on behalf of the fund, and the tax on these exports is an internal transaction within the fund. While its original purpose, in 1958, was to accumulate in public hands the coffee withheld from the market under a retention agreement among Latin American producers, the retention tax has come to be used as a device to manipulate domestic ORGANIZATION OF THE COFFEE ECONOMY 209 prices and to shield domestic producers from the full effects of changes in world prices. In periods when the world price was high, the retention tax was increased to keep domestic prices relatively low, as in 1976, when the tax was at 85 percent; conversely, in periods of low world prices, the retention tax was reduced to prevent domestic prices from falling too much (see table F-3). The Pasilla and Ripio Tax Before a license is issued for the export of a consignment of green coffee, an exporter must provide the federation with evidence of sale of an amount of low- grade pasilla and ripio parchment equivalent to 6 percent of the volume of the consignment. For the delivery, which must comprise eleven parts pasilla to one part ripio and be made to a warehouse of Almacafe, the exporter receives a payment of six pesos per 62.5-kilogram bag of hulled coffee. This rate of payment has been unchanged since 1941 and now represents only a fraction of a percentage point of the value of the coffee. The pasilla and ripio tax is designed to remove low-grade coffee from the export market and provide the federation with stocks for sale to the domestic market. The Ad Valorem Tax A tax equal to 6.5 percent of the reintegro price is currently payable to the Cen- tral Bank in foreign exchange by all exporters of coffee, including the federation. Out of this, an amount equal to 3.2 percent of total export value is paid by the bank to the NCF, and another 0.8 percent of total export value is paid to the departmental committees to be used for projects in the coffee zone. The remainder is paid into the Special Exchange Account of the Treasury and represents an important contri- bution to the national revenues; it averaged about 7.5 percent of total government revenues during the period 1974-81. In the third quarter of 1983 the so-called reintegro anticipado was introduced in an effort to induce early surrender of coffee revenues to the Central Bank. The measure provided exporters with a forward exchange rate of up to sixty days for future coffee sales. The ad valorem tax was established in 1967 at the rate of 26 percent as part of the tax reforms that accompanied the abolition of the special rate of exchange for coffee. The rate of the tax was reduced in steps of 0.25 percent a month until it reached the rate of 20 percent in December 1968, at which level it was held until the end of 1974. The rate was then cut by 1 percentage point each year from 1975 to 1978, when it had fallen to 16 percent. Changes in the rate since 1978 are shown in table F-3. The loss in revenue from these reductions was borne entirely by the Treasury. Of the portion of the tax received by the NCF, a fifth passes directly to the campaign for economic and social progress administered by the departmental committees of the federation. Table F-3. Coffee Price and Tax Variables Reintegro minimo (U.S. dollars Ad valorem Retention Effective per seventy- tax quota date kilogram bag) (percent)' (percent) 1978 January 1 313.75 16 80 April 8 290.00 16 80 June 17 275.00 16 80 July 12 259.00 16 80 1979 January 25 243.00 16 80 February 1 217.00 16 80 February 22 188.40 16 80 February 27 188.40 16 45 April 19 202.00 16 45 May 5 216.00 16 55 June 5 251.00 16 58 1980 May 9 287.32 16 62 October 3 201.00 16 25 December 9 201 00 4 25 December 10 181 95 4 25 December 11 181 95 4 15 1981 April 24 186.55 20 20 September 1 186 55 12 20 September 18 186 55 12 25 November 6 201 90 12 30 December 14 206 50 12 35 1982 March 13 217.25 12 39 May 24 206.50 12 35 October 1 206.50 9 40 1983 February 19 191.00 9 40 September 10 191.00 6 5 45 October 15 195.50 6.5 50 November 30 204.50 6.5 50 December 12 204 50 6.5 58 1984 February 1 204.50 6.5 62 March 22 206.00 6.5 66 a. The ad valorem tax is divided between the government and the coffee sector The National Coffee Fund has always received 3.2 percent, the departmental committees 0.8 percent, and the government the remainder. Of the 6.5 percent tax, the government now receives 2.5 percent. Source: FEDERACAFE, Divisi6n de Investigaciones Econ6micas 210 ORGANIZATION OF THE COFFEE ECONOMY 211 The Discount on Currency Exchange Certificates When surrendering the proceeds of coffee sales in dollars to the Central Bank, the exporter receives a certificate, which in the past-from May 1977 through October 1980-could be redeemed at face value in 120 days or sold immediately at a discount. This system acted as an indirect tax on coffee exports. As noted earlier, the discount has now been eliminated. Table F-4 contains a brief summary of the evolution of the various taxes on coffee since 1950. As can be seen, the retention tax has been gaining in importance in recent years and now provides almost 70 percent of total tax revenues. The Tax System and the Producers' Price By using the various taxes and prices, the government is able to determine the price received by growers, to influence production, and to determine whether growers sell to the federation or to private exporters, as discussed in chapter 6. An exporter pays to the Central Bank the foreign exchange received from the sale of coffee, and the bank pays the equivalent in pesos, converted at the current offi- cial rate of exchange less the value of the ad valorem tax based on the reintegro price. This amount is paid in the form of currency exchange certificates, which the ex- porter values at less than their face value. From this must be deducted the cost of internal transport, grading, and warehousing, and the minimum acceptable amount of profit. The balance is available for payment for the volume of parchment neces- sary for the export order and for the payment of the retention tax. Division of the balance by this volume-including retention and the amount needed to convert pergamino into excelso-gives the maximum price per unit of parchment that the exporter is prepared to pay. In practice an exporter will take other factors into account when determining this price, such as the small payment received for deliv- eries of pasilla. By setting the minimum price of the federation higher than the price that some, or all, exporters can offer, or by increasing taxes, thereby forcing the exporter's price below that of the federation, the government can induce growers to sell to the federation and can reduce or eliminate sales to private exporters. Conversely, by setting the federation price below the price which private exporters can pay, the government can reduce or eliminate sales to the federation. In table F-5 the prices paid by the federation and by exporters are given for each month since January 1975. A grower might be expected to sell coffee to the buyer that offers the higher price. The higher of the prices can therefore reasonably be considered the market price to growers in each month. Table F-4. The Value of Taxes Levied on the Coffee Sector and the Value of Coffee Production, Selected Years 1950-82 (million pesos) Total taxes and Exchange Pasilla contributions as Ad differential General and Total taxes Value a percentage of valorem and exchange export ripio Retention and of the value of Year tax discount tax tax quota contributions production production 1950 . . . . .. 1.0 0.9 1.9 1,070.1 0.2 1955 ... 14.6 1.3 1 2 .. 17.1 1,825.4 0 9 1960 324.0 94.4 1.3 1.3 212.5 633.5 2,573.8 24.6 1965 . 613.7 1.2 1.4 242.6 858.9 4,304.0 20.0 1969 1,563.2 . . I 4 1.4 1,275.2 2,841.2 8,342 1 34.0 1971 1,392.9 .. 1.4 1.4 1,260.1 2,655 8 7,894.5 33.6 1972 1,807.3 . 1.4 1.4 1,630.8 3,441.9 10,922.8 31.5 1973 2,683.9 . . . 1.4 2,424.8 5,110.1 14,497.5 35.2 1974 2,744.0 . . . 1.5 3,319.0 6,004.5 17,229.9 35.2 1975 3,402.5 . . . 1.8 3,587.5 6,991.8 20,397.8 34.3 1976 6,156.0 . . 1.0 10,071.0 16,228.0 39,251.3 41.3 1977 8,964.0 2,5500 . . 4.0 20,545.0 32,063.0 65,928.0 48.6 1978 10,878 3 5,397.0 . . . 9.6 29,707.2 45,992.1 77,099.0 59.6 1979 12,342.0 5,036.0 . . . 4.0 29,846.0 47,228.0 78,764.0 60.0 1980 15,025.0 4,068.0 .. 4.0 31,116.0 50,213.0 99,683.0 50.4 1981 11,468.0 n.a. 4.0 9,597 0 21,069.0' 102,000.0 20.7 1982 10,607.0 n.a. 5.0 23,531.0 34,143.0a 120,0000 28.0 n.a. Not available. . . . Zero or negligible. a. Excluding exchange discount Sources. FEDERACAFE; ICo, Co/tce in Colombia 1979/80 (London, September 1980) ORGANIZATION OF THE COFFEE ECONOMY 213 Table F-5. Prices Paid to Growers by the Federation and by Private Exporters, 1975-83 (pesos per 125 kilograms of federation-type pergamino) Private Year Month Federationa exportersa 1975 Average 2,730 2,934 January 2,500 2,596 February 2,500 2,492 March 2,500 2,406 April 2,435 2,256 May 2,350 2,500 June 2,350 2,561 July 2,560 2,833 August 3,000 4,143 September 3,000 3,551 October 3,065 3,442 November 3,250 3,496 December 3,250 3,500 1976 Average 5,533 5,828 January 3,496 4,131 February 3,845 4,528 March 4,120 4,430 Aprd 4,495 5,356 May 4,495 6,115 June 6,079 6,669 July 6,560 6,221 August 6,560 6,283 September 6,560 6,271 October 6,560 6,225 November 6,619 6,479 December 7,000 7,200 1977 Average 7,179 6,946 January 7,000 7,445 February 7,000 7,371 March 7,000 7,541 April 7,000 7,291 May 7,048 6,589 June 7,300 6,524 July 7,300 6,560 August 7,300 6,828 September 7,300 6,869 October 7,300 6,645 November 7,300 6,839 December 7,300 6,849 (Table continues on the following page.) 214 APPENDIX F Table F-5 (continued) Private Year Month Federationa exportersa 1978 Average 7,300 6,946 January 7,300 7,008 February 7,300 7,001 March 7,300 6,921 April 7,300 6,991 May 7,300 6,923 June 7,300 6,938 July 7,300 6,905 August 7,300 7,009 September 7,300 7,073 October 7,300 6,960 November 7,300 6,849 December 7,300 6,773 1979 Average 7,270 7,179 January 7,300 6,920 February 7,236 6,860 March 6,400 6,150 April 6,400 6,363 May 6,574 6,578 June 6,946 6,955 July 7,143 7,110 August 7,340 7,276 September 7,714 7,759 October 7,900 7,919 November 8,066 8,076 December 8,216 8,185 1980 Average 8,663 8,528 January 8,300 8,305 February 8,300 8,333 March 8,356 8,450 April 8,733 8,761 May 8,733 8,763 June 8,73 3 8,484 July 8,733 8,313 August 8,733 8,345 September 8,733 8,318 October 8,733 8,665 November 8,733 8,675 December 9,140 8,898 1981 Average 9,453 9,271 January 9,200 8,893 ORGANIZATION OF THE COFFEE ECONOMY 215 Table F-5 (continued) Private Year Month Federation' exporters February 9,200 8,861 March 9,200 9,036 April 9.200 9,100 May 9,200 9,000 June 9,200 8,893 July 9,200 9,156 August 9,200 9,380 September 9,460 9,310 October 9,800 9,673 November 10,241 9,956 December 10,330 10,000 1982 Average 11,171 11,003 January 10,330 10,166 February 10,330 10,214 March 10,795 10,619 April 11,050 10,868 May 11,050 10,770 June 11,050 10,930 July 11,050 10,913 August 11,050 10,891 September 11,050 10,923 October 12,100 11,839 November 12,100 11,963 December 12,100 11,935 1983 Average 13,075 n.a. January 12,100 12,095 February 12,100 12,156 March 12,100 12,102 April 12,800 12,460 May 12,800 12,363 June 12,800 12,563 July 12,800 12,543 August 12,800 12,615 September 13,900 13,151 October 14,150 n.a. November 14,150 n.a. December 14,400 n.a 1984 March 14,800 n.a. n.a. Not available. a. Often the higher price paid by the federation than that paid by private exporters is explained by differences in quality. Source: FEDERACAFE. 216 APPENDIX F Table F-6. Proceeds of the Taxes on Coffee and Their Distribution, as Percentage of the Total, 1974-82 Total taxes National Share of the and levies coffee departmental National (million fund committees of government Year pesos) share FEDERACAFE share 1974 6,064 62.9 1.8 35.3 1975 6,992 58.5 2.1 38.4 1976 16,228 68.8 1.7 29.5 1977 32,063 69.3 1.3 29.4 1978 45,992 69.0 1.0 30.0 1979 47,228 68.0 2.0 30.0 1980 50,213 68.0 2.0 30.0 1981a 21,069 59.0 3.0 38.0 1982a 34,143 83.9 2.3 13.8 a. Excluding the exchange discount Source: FEDERACAFE. The Recipients of Coffee Taxes Of the total taxes and levies on the export of coffee, in recent years more than two-thirds have gone to the National Coffee Fund, which obtains its revenues en- tirely from taxes on coffee. The whole of the retention tax is received by the fund, and the share received by the fund from all forms of taxation on coffee has tended to increase as this tax has gained in importance in relation to the ad valorem tax, which, apart from the tax implicit in the system of currency exchange certificates, is the only indirect tax on the coffee sector received by the government.2 Total pro- ceeds and their distribution are shown in table F-6. Notes 1. Much of this appendix, which supports the analysis presented in chapter 6, is drawn from a publi- cation of the International Coffee Organization (Ico), Coffee in Colombia, 1979/80 (London, September 1980). 2 Note, however, that the value of the retention tax does not accrue to the federation or the Na- tional Coffee Fund until the coffee has been sold. The bags of coffee received as retention tax but never sold are of no value. The figures in table F-6 were computed on the assumption that the coffee delivered to the federation as retention tax should be valued at its market value. To the extent that this coffee has a true value less than its market value these figures are an overestimate of the total taxes and the shares of the Ncy and the departmental committees. The government, on the other hand, receives its taxes in cash. STATISTICAL APPENDIX 218 STATISTICAL APPENDIX Table SA-1. Gross Domestic Product, by Type of Expenditure, at Current Market Prices, 1971-84 (million pesos) Economic variable 1971 1972 1973 1974 1975 1976 Gross domestic product at market prices 155,886 189,614 243,160 322,384 405,108 532,270 Gross domestic consumption 131,819 155,877 195,780 258,766 332,095 422.031 Private consumption 114,724 137,733 172,769 230,558 295,919 378,349 Government consumption 17,095 18,144 23,011 28,208 36,176 43,682 Gross domestic investment 30,266 34,371 44,425 69,170 68,838 93,481 Grossfixedinvestment 27,302 30,486 38,416 52,843 62,129 84,571 Change in stocks 2,964 3,885 6,009 16,327 6,709 8,910 Exports of goods and nonfactor ser- vices 19,414 26,392 36,614 48,453 66,961 97,317 Imports of goods and nonfactor ser- vices 25,613 27,026 33,659 54,005 62,786 80,559 Net factor income from abroad - 3,508 -4,286 -5,082 -5,030 -8,134 - 10,859 Gross national product at market prices 152,378 185,328 238,078 317,354 396,974 521,411 Note: Exports and imports of goods and nonfactor services and net factor income are balance of payments figures of the Central Bank, converted to Colombian pesos by World Bank staff members, which differ from DANES figures of national accounts. The exchange rates used in the conversion in this table are the official average rates. Private consumption is a residual item in the account. Table SA-2. Gross Domestic Product, by Type of Expenditure, at Constant Market Prices, 1970-83 (million 1975 pesos) Economic variable 1970 1971 1972 1973 1974 1975 Gross domestic product at market prices 307,496 325,825 350,813 374,398 395,910 405,108 Gross domestic consumption 254,373 275,958 292,327 309,224 322,415 332,095 Private consumption 227,063 241,484 259,481 273,033 287,055 295,919 Government consumption 27,310 34,474 32,846 36,191 35,360 36,176 Gross domestic investment 63,148 65,844 64,244 70,073 83,967 68,838 Gross fixed investment 53,201 55,786 54,687 59,443 64,604 62,129 Change in stocks 9,947 10,058 9,557 10,630 19,363 6,709 Exports of goods and nonfactor ser- vices 47,961 50,036 56,514 58,489 57,889 66,961 Imports of goods and nonfactor ser- vices 57,986 66,013 62,272 63,388 68,361 62,786 Net factor income from abroad -11,258 -10,143 -10,308 -9,473 -6,784 -8,134 Gross national product at market prices 296,238 315,682 340,505 364,925 389,126 396,974 Note: Net factor income is a balance of payments figure deflated by members of the World Bank staff using the manufacturing unit value index (suv). Private consumption is a residual item in the account STATISTICAL APPENDIX 219 1977 1978 1979 1980 1981 1982 1983 1984' 716,029 909,487 1,188,817 1,579,130 1,982,773 2,497,298 3,036,661 3,691,586 556,712 719,187 942,437 1,265,852 1,650,133 2,110,924 2,587,532 3,050,201 501,496 641,367 831,715 1,106,481 1,443,259 1,838,158 2,235,983 2,626,197 55,216 77,820 110,722 159,371 206,874 272,766 351,549 424,004 134,270 166,293 215,782 301,117 408,927 511,625 588,238 698,932 104,041 139,897 183,325 264,894 350,048 436,091 512,780 653,405 30,229 26,396 32,457 36,223 58,879 75,534 75,458 45,507 126,670 157,925 198,226 275,166 254,909 306,719 319,383 457,757 101,623 133,918 167,628 263,005 331,196 431,970 458,492 515,304 -10,004 -11,768 -10,852 -14,893 -30,733 -70,510 -96,682 151,578 706,025 897,719 1,177,965 1,564,237 1952,040 2,426,788 2,939,979 3,540,008 a. Preliminary estimate. Sources: DANE, Central Bank, and World Bank estimates. 1976 1977 1978 1979 1980 1981 1982 1983 424,263 441,906 479,335 505,119 525,765 537,736 542,836 548,055 351,946 367,505 398,487 413,144 439,649 454,808 467,448 476,078 314,398 328,177 355,555 364,892 385,285 398,421 408,442 414,975 37,548 39,328 42,932 48,252 54,364 56,387 59,006 61,103 75,245 87,468 93,516 93,220 103,358 117,037 123,279 117,858 68,039 68,518 74,923 77,775 88,021 93,539 96,307 95,715 7,206 18,950 18,593 15,445 15,337 23,498 26,972 22,143 66,519 62,154 77,452 88,061 90,727 80,770 82,496 72,226 69,447 75,221 90,120 89,306 107,966 114,879 130,386 118.107 -9,503 -7,638 -7,189 -5,477 -6,245 -11,770 -23,399 -27,219 414,760 434,268 472,146 499,642 519,520 525,966 519,437 520,836 Sources: DANE, Central Bank, and World Bank estimates. 220 STATISTICAL APPENDIX Table SA- 3. Gross Domestic Product at Factor Cost, by Sector, at Current Prices, 1970-83 (million pesos) Sector 1970 1971 1972 1973 1974 1975 Gross domestic product at factor cost 122,874 144,591 176,534 226,583 300,100 373,425 Agricultureb 33,515 36,863 46,033 58,961 78.952 97,337 Mining 2,920 3,534 4,544 6,234 6,022 7,407 Manufacturing 21,278 25,052 32,220 44,427 60,523 73,184 Construction 4.953 5,994 7,168 9,960 12,436 12,604 Electricity, gas, and water 1,413 1,685 1,922 2,619 3,104 3,968 Transport and communication 11.701 14,040 16,826 19,375 24,499 34,865 Trade 16,171 19,217 22,941 30,432 43,557 52,442 Banking, finance, rent, and insurance 17,775 21,899 25,140 31,010 41,062 54,613 Public administration and defense' 9,641 12,216 14,644 17,603 22,418 28,281 Other branches' 6,957 8,361 9,902 11,990 15,508 19,964 Less: imputed banking charges 3,450 4,270 4,806 6,028 7,981 11,240 a. Preliminary estimate. b. Includes fishing, hunting, and forestry. c. Equals government services. Table SA-4. Gross Domestic Product at Factor Cost, by Sector, at Constant Prices, 1970-83 (million 1975 pesos) Sector 1970 1971 1972 1973 1974 1975 Gross domestic product at factor cost 280,605 297,892 323,354 348,611 365,334 373,425 Agricultureb 78,379 78,958 85,182 87,096 91,883 97,337 Mining 9,218 8,408 8,911 9,617 7,146 7,407 Manufacturing 50,253 55,019 62,410 69,951 73,137 73,184 Construction 10,012 10,317 10,962 13,256 13,980 12,604 Electricity, gas, and water 2,271 2,588 2,934 3,489 3,684 3,968 Transport and communication 24,210 25,977 28,675 32,039 34,622 34,865 Trade 37,382 40,939 44,022 48,081 51,563 52,442 Banking, finance, rent, and insurance 42,506 46,130 47,215 49,718 52,648 54,613 Public administration and defensec 21,223 23,829 26,108 27,559 28,146 28,281 Other branchesd 13,736 15,110 16,271 17,506 18,846 19,964 Less: imputed banking charges 8,585 9,383 9,336 9,701 10,321 11,240 a. Preliminary estimate b. Includes fishing, hunting, and forestry. c. Equals government services. STATISTICAL APPENDIX 221 1976 1977 1978 1979 1980 1981 1982 1983' 482,578 641,046 810,477 1,065,871 1,420,746 1,815,356 2,282,769 2,785,098 126,061 179,567 209,951 255.905 305,286 381,281 468,770 568,262 10,410 10,617 12,055 17,347 35,764 48,270 63,739 87,047 93,430 115,505 138,741 180,424 267,910 327,768 406,736 483,523 18,366 26,241 35,431 47,989 71,147 98,374 124,829 149,384 6,144 8,229 11,413 16,305 20,802 34,040 48,849 67,426 46,078 61,621 77,332 107,666 143,716 176,267 213,835 252,215 66,594 89,457 119,677 157,301 204,058 262,363 332,926 395,993 69,433 88,063 118,182 163,435 210,969 276,822 352,472 417,968 35,243 45,235 63,051 85,277 120,494 160,715 210,900 272,229 24,972 34,068 47,102 63,193 80,789 108,063 138,470 174,627 14,153 17,557 22,458 28,971 40,189 58,607 78,757 83,576 d Composed of house rentals and personal services. Source: DANE. 1976 1977 1978 1979 1980 1981 1982 1983' 390,285 404,169 429,225 447,501 472,113 485,591 488,249 494,238 99,952 103,107 111,549 117,058 119,144 123,017 120,841 122,911 6,795 5,852 5,463 5,548 6,594 6,941 7,055 7,966 76,547 75,527 76,338 76,610 86,361 85,296 83,084 83,575 14,008 15,093 14,634 14,722 16,833 18,190 19,047 20,045 4,158 4,163 4,377 4,891 5,232 5,511 5,652 5,865 37,216 40,622 45,699 48,681 49,863 52,631 54,872 54,147 55,048 56,979 61,214 63.543 64,804 65,873 67,144 65,743 56,440 59,345 64,198 67,296 71,643 75,741 78,158 80,381 30,647 32,737 34,594 36,857 40,515 42,835 43,891 45,624 20,935 22,209 23,759 24,857 25,219 26,199 26,896 27,497 11,461 11,465 12,600 12,562 14,095 16,643 18,391 19,516 d. Composed of house rentals and personal services. Source: DANE. 222 STATISTICAL APPENDIX Table SA-5. Gross Domestic Product at Market Prices, Agriculture and Total, 1970-83 (thousand pesos) Current pesos 1975 pesos Noncoffee Entire Noncoffee Entire Year Agriculture' agricultureb economy Agriculturea agricultureb economy 1970 36,194 28,216 132,768 86,488 68,056 307,496 1971 39,595 32,092 155,886 88,059 69,931 325,825 1972 49,439 39,692 189,614 93,772 74,709 350,813 1973 65,203 50,752 243,160 96,022 76,599 374,398 1974 84,386 69,585 322,384 100,944 81,131 395,910 1975 108,490 86,703 405,108 108,490 86,703 405,108 1976 147,300 106,444 532,270 108,805 88,589 424,263 1977 211,216 146,487 716,029 109,904 90,142 441,906 1978 240,133 170,801 909,487 123,624 97,579 479,335 1979 285,523 211,026 1,188,817 132,306 101,635 505,119 1980 362,075 261,746 1,579,130 135,499 103,989 525,765 1981 407,649 323,896 1,982,773 136,285 106,095 537,736 1982 503,897 401,423 2,497,298 134,591 103,405 542,856 1983c 612,239 n.a. 3,036,661 137,445 n.a. 548,055 n.a. Not available. a. Consists of pergamino coffee (01), other agricultural products (02), animal products (03), processed coffee (08), and sugar manufacturing (12). b. Consists of other agricultural products (02), animal products (03), and sugar manufacturing (12). c. Preliminary estimate. Sources: DANE, Cuentas nacionales de Colombia, 1970-1982 (Bogott, August 1984), and information not yet published for 1982 and 1983. Table SA-6. Exports (fo. b.) and Imports (c. if), 19 70-83 (million current pesos) Exports (fo.b.) Imports (c if) Processed Rest coffee Sugar Broad Rest of the Sugar of the Year Agriculturea (08) (12) agricultureb economy Total Agriculturea (12) economy Total 1970 2,095 8,749 1,472 12,316 5,303 17,619 769 10 18,545 19,324 1971 2,118 8,279 336 10,733 7,921 18,654 1,324 ... 23,644 24,968 1972 2,781 10,646 672 14,099 11,034 25,133 1,223 ... 23,044 24,267 1973 3,103 15,165 783 19,051 17,239 36,290 2,365 ... 28,429 30,794 1974 4,694 16,703 1,927 23,324 23,551 46,875 3,560 ... 46,830 50,390 1975 7,809 23,622 2,883 34,314 29,763 64,077 2,574 ... 54,188 56,762 1976 8,650 42,329 993 51,972 38,760 90,732 3,639 ... 70,320 73,959 1977 11,229 60,751 106 72,086 48,677 120,763 4,084 22 90,401 94,507 1978 13,132 79,060 1,049 93,241 57,970 151,211 4,618 317 120,561 125,496 1979 14,241 88,762 2,445 105,448 75,448 180,896 5,968 . 153,870 159,838 1980 19,918 116,793 9,273 145,984 110,119 256,103 13,370 1 232,926 246,297 1981 25,618 85,773 4,705 116,096 118,887 234,983 12,130 2 293,575 305,707 1982 26,421 109,330 3,799 139,706 132,820 272,526 18,677 2 360,684 379,368 1983c 30,456 131,087 6,180 167,723 155,276 322,999 21,846 2 381,358 403,206 ... Zero or negligible. Note: Exports and imports for the economy as a whole comprise both goods and nonfactor services. These data differ slightly from the national accounts statistics of the Central Bank. a. Consists of pergamino coffee (01), other agricultural products (02), and animal products (03). b. Consists of 01, 02, 03, processed coffee (08), and sugar (12). c. Preliminary estimate. Sources: DANE, Cuentas naciona s de Colombia, 1970-1982 (Bogota, August 1984), tables 21.1 and 22.1, and information not yet published for 1982 and 1983. Table SA-7. Exports (fo. b.) and Imports (c. if), 19 70-83 (million 1 975 pesos) Exports (f o.b.) Imports (c.i.f) Processed Rest coffee Sugar Broad Rest of the Sugar of the Year Agriculture (08) (12) agriculture economy Total Agriculture (12) economy Total 1970 5,981 18,153 1,565 25,699 20,335 46,034 2,870 .. 50,702 53,572 1971 5,162 18,651 1,685 25,498 22,639 48,137 4,318 ... 60,045 64,363 1972 5,342 18,835 1,992 26,169 27,689 53,858 3,522 .. 52,383 55,905 1973 4,554 19,253 1,779 25,586 32,341 57,927 4,087 .. 53,889 57,976 1974 5,348 19,931 1,843 27,122 28,869 55,991 3,764 60,009 63,773 1975 7,809 23,622 2,883 34,314 29,763 64,077 2,574 54,188 56,762 1976 7,690 20,431 1,436 29,557 32,476 62,033 3,547 60,215 63,762 1977 8,114 15,921 291 24,326 34,916 59,242 3,177 70 66,732 69,979 1978 8,942 27,473 2,141 38,556 35,597 74,153 4,529 1,009 78,940 84,478 1979 7,283 33,991 3,041 44,315 36,032 80,347 3,871 ... 81,268 85,139 1980 8,381 34,753 3,591 46,725 37,725 84,450 6,111 94,994 101,105 1981 8,714 29,018 2,196 39,928 34,529 74,457 4,990 I 101,064 106,055 1982 7,116 28,416 3,825 39,357 33,940 73,297 7,004 1 107,500 114,505 19831 6,682 29,524 3,716 39,922 33,115 73,037 6,787 1 97,078 103,866 Zero or negligible. Note: Exports and imports for the economy as a whole comprise both goods and services. For definitions of agriculture and broad agriculture, see notes to table SA-7. a. Preliminary estimate. Sources. DANEI, Cuentas nacionales de Colombia, 1970-1982 (Bogot s, August 1984), tables 2 1.2 and 22 2, and information not yet publishsed for 1982 and 1983. Table SA-8. Commodity Exports, 1970-83 (million U.S. dollars) Item 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 Major export Green coffee 467.0 400.0 4300 598.0 622.0 672.0 967.2 1,497.9 1,979.0 2,005.0 2,361.0 1,423.3 1,561.5 1,506.2 Minor exports Agrobased products 101.4 105.2 155.0 166.5 225.9 327 4 267.5 353.8 339.3 381.1 643.0 589.6 506.8 420.3 Cotton 34.6 29.7 51.2 38.1 48.6 76.1 59.4 164.0 72.5 52.0 159.3 148.4 66.5 22.5 Cattle and beef 21.8 28.2 37.7 43.4 36.0 56.8 52.1 45.0 46.5 37.2 27.3 54.1 46.1 30.5 Sugar 14.0 15.7 28.4 30.2 68.6 95.1 24.1 2.2 19.5 49.6 165.0 76.9 54.7 68.9 Bananas 18.1 14.7 13.7 15.4 25.4 31.6 40.9 45.6 76.0 84.8 94.0 122.4 131.1 147.7 Tobacco 7.2 9.2 9.9 15,0 18.9 12.8 25.5 19.2 27.5 24.2 25.7 19.6 21.7 22.9 Flowers 1.0 1 8 3.1 8.4 16.0 19.3 21.6 32.6 53 4 79.2 99.4 108.6 111.5 120.6 Rice 0.0 0.0 0 7 3.4 0.5 22.9 21.4 19.9 4.6 8.8 16.6 9.5 32.6 2.8 Cheese . . 0.2 14 3.3 0.4 1.2 2 8 6.8 16 1 17.1 20.4 16.7 10.0 0.2 Fish 4.7 5.7 8.9 9.3 11.5 11.6 19.7 18.5 23 2 28.2 35.3 33.4 32.6 4.2 Manufactured products 83.3 118.4 168.5 2564 479.3 414.1 448.0 469.7 620.5 695.4 889.4 984.7 969.8 951.6 Food productsa 11.1 11.7 15.7 8.1 11 9 19.3 16.1 30.5 31.3 34.0 54.3 94.1 67 9 152 0 Footwear, clothing, and textilesb 18.7 26.7 42.6 81 9 154.0 102.4 139.0 88.0 194.7 1504 180.0 176.2 183.9 91.2 Leather and hides 6.7 7.0 19.0 25.5 16.0 16.5 20.4 30.0 33.8 37.7 32.0 39.7 47.4 30.2 Chemicals and pharmaceuticals 7.6 11.4 16.9 32.2 70.7 53.0 44.5 45.9 12.2 60.7 84.5 78.2 76.5 89.7 Basic metals and products 4.9 6.9 10.9 20.9 28.0 21 5 23.6 32.7 36.4 65.9 47.8 62.4 68.0 75.2 Mechanical and electrical equipment 3.7 5.3 7.3 13.1 23.6 23.1 29 4 43.9 41.1 52.1 62.4 63.8 62.7 31.7 (Table continues on the following page) Table SA-8 (continued) Item 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 Minor exports (cont.) Timber and wood products 5.4 5.6 8.7 22.0 30.6 8.0 13.6 14.7 7.5 15.1 11.3 11.7 16.9 7.5 Paper, cartons, and books 3.6 4.4 10.0 11.0 14.2 16.4 25.1 28.3 72.2 53.0 71.0 90.9 74.0 57.0 Cement 3.3 3.2 5.7 6.9 9.8 11.9 23.3 14.9 21.9 30.7 35.7 31.3 34.3 19.9 Glass 4.2 3.9 4.2 4.8 6.6 7.3 10.0 10.0 9.7 14.3 18.0 14.9 11.3 8.2 Plastics 1.3 2.3 3.5 4.7 5.9 7.6 9.8 10.7 14.4 17.3 25.8 25.4 26.9 26.8 Transport equipment 0.7 9.5 2.3 2.8 5.4 7.1 9.2 15.6 21.1 17.9 25.7 30.2 16.2 10.2 Fuel oil 12.1 20.5 21.7 22.5 102.6 119.9 84.0 104.5 124.2 146.3 240.9 265.9 283.8 352.0 a Other products 83.9 66.4 112.5 156.4 86.1 51.7 62.5 121.8 63.9 218.9 51.6 42.2 34.6 202.8 a Total goods 735.6 690.0 866.0 1,177.3 1,413.3 1,465.2 1,745.2 2,443.2 3,002.7 3,300.4 3,945.0 3,039.8 3,072.7 3,080.9 Balance of payments adjustment 52.4 62.0 113.0 85.7 77.1 281.8 509.8 283.8 267.3 280.6 351.0 440.6 187.0 96.1 Total goods adjusted 788.0 752.0 979.0 1,263.0 1,494.0 1,747.0 2,255.0 2,727.0 3,270.0 3,581.0 4,296.0 3,397.0 3,282.0 3,147.0 Freight and insurance 43.0 49.0 49.0 65.0 98.0 93.0 120.0 144.0 140.0 153.0 142.4 114.1 115.7 103.3 Other transportation 52.0 58.0 55.0 69.0 80.0 82.0 142.0 146.0 165.0 182.0 290.0 366.3 319.6 287.4 Travel 54.0 61.0 59.0 72.0 105.0 141.0 175.0 231.0 260.0 357.0 478.0 437.0 484.4 235.0 Other 63.0 54.0 65.0 79.0 81.0 102.0 113.0 195.0 204.0 385.0 541.0 364.0 583.0 278.0 Total goods and nonfactor services 1,000.0 974.0 1,207.0 1,548.0 1,858.0 2,165.0 2,805.0 3,443.0 4,039.0 4,658.0 5,747.4 4,678.4 4,784.5 4,050.3 . . Zero or negligible. a. Excluding sugar. b. Excluding cotton fiber Source: Customs data. Table SA-9. Balance of Payments, 1971-84 (million U.S. dollars) Item 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984a Merchandise exports f o.b.b 752 979 1,263 1,494 1,747 2,255 2,728 3,270 3,581 4,296 3,397 3,282 3,147 3,658 Merchandise imports f.o.b. 900 848 982 1,510 1,425 1,666 1,980 2,564 2,996 4,283 4,730 5,358 4,464 3,980 Tradebalance -148 131 281 -16 322 589 748 706 585 13 -1,333 -2,076 -1,317 -322 Nonfactor service receipts 222 228 286 365 418 550 716 769 1,077 1,451 1,281 1,503 903 884 Nonfactor service payments 385 388 442 562 605 656 783 861 943 1,210 1,340 1,381 1,350 1,133 Goodsandservicesbalance -311 -29 124 -213 135 483 681 614 719 854 -1,400 -1,954 -1,764 -571 Netfactorincome -176 -196 -215 -193 -263 -313 -272 -301 -255 -315 -564 -1,100 -1,226 -1,504 t, Receipts (11) (10) (26) (67) (61) (71) (72) (132) (267) (494) (647) (510) (280) (124) Payments (187) (206) (241) (260) (324) (384) (344) (433) (522) (809) (1,211) (1,610) (1,506) (1,628) Net private transfers 3 11 12 23 30 39 40 44 98 165 242 169 164 205 Current account balance -484 -214 -79 -383 -98 209 449 357 562 104 -1,722 -2,885 -2,826 -1,870 Officialgrantaid 31 24 23 33 18 12 6 29 3 - - - - - Private capital 59 24 8 27 24 -25 37 35 208 108 631 667 587 n.a. Direct investment 40 17 23 36 32 14 43 67 104 51 228 337 514 411 Loans(net) 19 7 -15 -9 -8 -30 -6 -32 104 57 403 330 73 n.a. Disbursements (109) (104) (81) (88) (58) (44) (55) (65) (152) (70) (690) (428) (307) n.a. Amortization (90) (97) (96) (97) (66) (83) (61) (97) (48) (13) (287) (98) (235) n.a. Public and publicly guaranteed capital 45 61 310 216 9 141 202 79 505 750 981 960 943 1,116 (Table continues on the following page.) Table SA-9 (continued) Item 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984" Disbursements (237) (257) (441) (424) (411) (295) (382) (325) (950) (999) (1,249) (1,285) (1,342) (1,743) Amortization (92) (96) (131) (208) (142) (154) (180) (246) (445) (249) (266) (325) (399) (627) Special drawing rights allocation 17 18 .. . ... 24 24 24 Short-term capital n a n.a. n.a. n.a n.a. n.a. n.a. n.a. n.a. 107 336 370 -395 -390 Public, net' n.a. n a. n.a. n.a. n,a n.a. n a. n.a. n.a. (-83) (38) (41) (-90) (4) Private, net n.a isa. n.a. n.a. n.a. n.a. in a. n.a. n.a. (190) (298) (329) (-305) (-394) Othercapital,net 157 79 -37 -257 -74 225 158 110 -65 75 -138 17 -163 n.a. Capital account balance 409 406 304 19 237 353 403 253 675 1,064 1,834 2,014 972 n.a. 00 Net change in reserves (minusmeansincrease) 75 -192 -225 364 -139 -562 -852 -618 -1,237 -1,168 -112 871 1,854 n.a. CentralBank -19 -178 -180 95 -117 -619 -667 -652 -1,624 -1,310 -214 739 1,812 1,284 Rest of banking system 94 -14 -42 269 - 22 57 - 185 42 387 142 102 132 42 n.a. n.a. Not available. * . . Zero or negligible. a. Preliminary estimate. b. Includes sales of nonmonetary gold. c. Includes liabilities that represent reserves of foreign authorities. Source: Banco de Ia Republica. Table SA-10. Imports, by Economic Category, 1970-83 (million U.S. dollars) Item 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 Consumer goods 86.9 101.1 105.4 161.6 190.3 168 5 204.5 287.6 503.5 451.1 619.6 667.6 690.6 538.9 Durables 43.9 45.1 484 57.5 87.9 78 5 93.7 130.3 187.4 196.8 312.2 336.2 366.7 236.8 Nondurables 430 56.0 57.0 104.1 102.4 90.0 110.8 157 3 316.1 254.3 307.4 331.4 323.9 302.1 Raw materials and intermediate goods 366.1 410.7 405.7 490.1 936.4 780.8 843.0 1,076.5 1,434.8 1,705.3 2,458.8 2,701.1 2,771 2 2,542.8 Fuels 1.2 8.1 5.4 2.1 3.0 14.5 39.9 136.2 204.5 322.2 562.8 724 1 656 7 639.1 Agricultural inputs 8.3 10.1 21 2 34.8 99.4 54.8 22.9 69.7 104.5 95.8 162.2 147.0 189.7 158.3 Industrial inputs 356.6 392.5 379.1 453.2 834.0 711.5 780.2 870.6 1,125.8 1,287.3 1,734.0 1,830.0 1,924.8 1,7454 Capital goods 368.2 397.6 330 8 386.7 464.9 539.3 660.6 664.2 898.0 1,076.8 1,584.2 1,830.5 2,015.9 1,886.4 Construction 20.3 15.8 10 1 18.9 32.4 35.4 42 6 26 1 44.7 63.1 98.3 159.6 213.6 129.3 Agricultural 13.0 8.1 10.6 18.3 24 3 29.0 30.7 44.1 54.2 39.3 63.1 66.1 68.3 66.0 Industrial 1947 246.3 216.7 225.1 260.5 269.1 330.5 387.8 517.0 603.1 955.0 1,113.2 1,148.9 1,112.9 Transport 140.2 127 4 93.4 1244 147.7 205.8 156.7 206.2 282.1 371 2 467.8 491.6 585.1 578.2 Unclassified 15.0 20.0 17.1 23.1 5 6 6.2 . . . . . . Total goods 836.2 929.4 859.0 1,061,5 1,597.2 1,494.8 1,708.1 2,028.3 2,836.3 3,233.2 4,662.6 5,199 2 5,477.7 4,968.1 Balance of payments adjustment -34.2 -29.4 -11.0 -79.5 -87.2 -69.8 -42.1 -48.3 -272.3 -237.3 -379.6 -469.2 -119.7 -504.1 Total goods adjusted 802.0 900.0 848.0 982.0 1,510.0 1,425.0 1,666.0 1,9800 2,564.0 2,996.0 4,283.0 4,730.0 5,358.0 4,464.0 . . Zero or negligible. Notes: Subcategories were calculated on the basis of import registrations as shares of totals for the years 1970-8 3. Figures for total imports of merchandise have been adjusted by the Banco de la Republica. Sources: Customs data and DANE. Table SA- 11. Imports, by Principal Product Group, 1970-83 (million U.S. dollars) Machinery Vehicles and Fuels, Chemicals Paper and electrical transportation mineral oils, and pharma- Iron and materials Rubber Year equipment equipment and products ceuticals steel Plastics and products products Foodstuffs Other Total, 1970 231.2 136.8 8.7 76.6 78.6 22.5 35.9 14.0 31.0 200.9 836.2 1971 275.5 127.7 10.8 87.6 81.6 25.4 35.1 16.9 62.7 206.1 929.4 1972 256.0 117.3 5.4 92.4 64.6 21.0 37.8 17.0 46.7 200.8 859.0 1973 359.3 128.0 3.9 127.0 72.2 21.8 48.4 21.3 80.1 199.5 1,061.5 1974 306.7 191.6 3.7 210.8 139.6 45.1 67.9 35.1 143.5 453.2 1,597.2 1975 329.7 238 4 18.4 199.1 135.3 41.1 68.3 28.7 94.8 341.0 1,4948 1976 406.4 251.4 41.7 200.6 122.9 48.8 66 6 40.5 146.2 383.0 1,708.1 1977 475.5 273.7 136.3 236.8 123.6 61.0 72 2 45.3 156.6 447.3 2,028.3 1978 618.4 380.7 205.1 300.6 180.9 84.0 96.3 55.3 1806 734.4 2,836.3 1979 719.4 455.4 324.3 291.4 251.6 101.2 103.4 69.4 213.6 703.5 3,233.2 1980 1,099.1 626.0 566.5 409.9 316.1 145.3 169.8 85.8 232.6 1,011.5 4,662.6 1981 1,244.9 661.4 729.1 436.3 382.5 135.6 199.7 95.8 294.9 1,019.0 5,199.2 1982 1,310.0 776.9 661.4 457.2 442.5 139.6 213.1 98.2 297.0 1,081.8 5,477.7 1983 1,227.1 647.9 646.9 446.6 315.2 130 6 187.1 84.2 208.5 1,074.0 4,968.1 a. Before balance of payments adjustment. Sources: Customs data and DANE. STATISTICAL APPENDIX 231 Table SA-12. Weighted Average CAT Subsidies, 1978-83 (percent) Section Chapters 1978 1981 1983 1 1-5 3.77 5.58 10.94 2 6-14 2.39 1.61 11.29 3 15 2.00 0.30 0.50 4 16-24 10.05 8.87 13 79 5 25-27 0.63 0.54 1.00 6 28-38 5 11 4.76 10.53 7 39-40 4.93 6.55 14.92 8 41-43 9.97 9.80 12.40 9 44-46 9.77 6.61 9.71 10 47-49 2.63 9.48 14.96 11 50-63 8.81 10.21 14.95 12 64-67 11.97 11 97 14.96 13 68-70 10.52 10 74 14.78 14 71-72 1.00 0.40 0.40 15 73-83 8.72 8.64 11.79 16 84-85 11 93 11.70 14.86 17 86-89 6.69 11.70 14.86 18 90-92 9.96 11.82 15.00 19 93 9.00 9.00 15.00 20 94-98 10.60 11.51 15.00 21 99 0.90 0.00 0.00 Source: Constructed from DNP data. Table SA-13. Index of Real Support Prices Established by IDEMA, 19 70-82 (1975 = 100) Paddy Year Sesame rice Barley Beans Corn Sorghum Soybeans Wheat 1970 81.3 138.9 102 8 89.5 89.1 83.6 73.4 90.4 1971 79.1 1250 995 80.6 95.0 77.1 85.9 81.4 1972 79.2 106.5 84.8 79 0 89.1 69.0 73.2 86.3 1973 65.3 86.9 73.8 82.4 84.2 77.9 61.6 77.3 1974 78.4 123.1 91.0 87.3 100.3 88.4 103.6 96.1 1975 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1976 n.a. 86.9 93.4 99.3 80.9 80.0 86.7 90.3 1977 n.a. 71.4 74.2 81.9 77.0 70.1 73.5 90.3 1978 88.0 86 2 87.1 85.3 91.0 81.1 87.9 68.0 1979 87.7 89.1 78.3 80.6 85.6 77.4 81.9 70.7 1980 91.1 103.0 73.7 75.7 98.5 90.6 80.2 81.2 1981 76.9 104.1 86.5 81.6 111.1 101.1 94.7 85 4 1982 74.7 100.9 92.3 82 5 120.5 114.8 107.9 86.5 n.a. Not available. Sources: DNP; Diagnostico del Sector Agrario, vol. 2, table 76; and IDEMA. 232 STATISTICAL APPENDIX Table SA-14. Ratio of Support Prices to Producer Prices, 1970-84 Paddy Year rice Corn Beans Sorghum Soybeans Wheat Barley Sesame 1970 1.22 0.91 1.13 0.88 0.68 1.09 1.26 0.77 1971 1.17 0.94 0.65 0.87 0.85 1.09 1.14 0.82 1972 1.20 0.81 075 0.61 0.81 1.04 1.05 0.90 1973 0.96 0.66 092 0.67 0.66 1.10 0.81 0.83 1974 1.14 0.96 0.82 0.83 0.99 1.05 1.03 0.73 1975 1.06 0.95 0.84 1.00 1.01 0.92 0 92 1.01 1976 1 13 0.84 0.97 0.90 0.97 1.08 0.96 n.a. 1977 0.85 0.68 0.92 0.79 0.76 1.01 0.96 n.a. 1978 0.99 0.98 0.98 0.95 0.95 1.05 1.10 1.06 1979 1.02 0.76 0.70 0.74 0.86 1.02 0.99 0.99 1980 1.14 0.77 0.68 0.85 0.85 1.11 1.00 1.01 1981 1.14 0.92 0.89 0.94 0.96 1.06 1.04 1.01 1982 1.07 n.a. 0.82 0.98 0 77 1.08 1.07 0.88 1983 1.05 n.a. 0.92 1.00 0.95 1.11 1.05 0.94 1984 0.86 1.28a 0.51 0.99 0.79 1.20' 1.01 0 78 n.a. Not available. a. Support price/producer price of first semester 1984. Source: IDEMA. Table SA- 15. Ratio of Domestic Prices to International Prices of Selected Agricultural Commodities, 19 70-82 Wholesale price! fo.b. international Producer price/fo.b. international price price Year Wheat Corn Sorghum Soybeans Rice Barley Cotton Sugar Beef 1970 1.84 1.38 1 78 1.37 0.82 1.03 0.93 1.4 0.9 1971 1.53 1.46 1.17 1.21 0.78 1.21 0.84 1.1 0.8 1972 1.66 1.77 1.49 1.05 0.62 0.69 0.74 1.3 0.7 1973 0.86 1.44 1.07 0.63 042 0.74 0.76 0.6 07 1974 0 97 0.98 0 87 0 84 0.40 0.86 0.71 0.2 0 7 1975 1.51 1 11 0.94 1 02 0.46 0.90 0.95 0.5 1.4 1976 1.51 1 24 1 02 1.00 0.58 1.08 1.00 0.8 1.5 1977 2.07 2.25 1.59 1.18 0.85 1.12 1.02 2.1 1.6 1978 1.54 1.79 1.43 1.21 0.69 1.38 0.95 2 0 1.8 1979 1.42 2.03 1 54 1.20 0.78 0.93 1 01 0 6 1.2 1980 1.54 2.37 1 39 1 31 0.69 0.96 1 29 0 7 1.2 1981 1.92 2.21 1.49 1.47 0.71 0.86 1 14 1 4 1.4 1982 2.08 1.83 1.93 1.89 1.20 0.98 n.a. 3.0 n.a. n.a. Not available Source. Jorge Garcia-Garcia, "Aspects of Agricultural Development in Colombia" (BogotA, April 1983, processed). STATISTICAL APPENDIX 233 Table SA- 16. Ratio of Production Costs to f o. b. International Peso Prices of Some Agricultural Commodities, 1970-82 Year Rice Corn Sorghum Soybeans Wheat Cotton 1970 1.011 2,402 1.090 0.825 2.663 0.446 1971 0.755 2.551 1.004 0.847 2.534 0.362 1972 0.722 2.459 1.024 0.776 1.977 0.270 1973 0.473 1.479 0.767 0.378 0.914 0.185 1974 0.313 0.780 0.674 0.585 0.866 0.232 1975 0.459 0.872 0.758 0.814 1 273 0.342 1976 0.615 1.013 0.963 0.863 1.495 0.257 1977 0605 1.376 1.243 1.014 1.886 0.389 1978 0.594 1.568 1.240 1.012 1.732 0.700 1979 0.699 1.336 1 000 0940 1.227 0.393 1980 0.745 1.608 1.084 1038 1.463 0.310 1981 0.692 1.853 1.388 1.226 1.954 0.390 1982 1.124 2.138 1.586 1.460 1.767 n.a. na Not available. Source. Garcia-Garcia, "Aspects of Agricultural Development. Table SA-1 7. Ratio of Support Prices to fo.b. International Prices of Selected Agricultural Commodities, 1970-83 Year Rice Corn Sorghum Soybeans Wheat Barley 1970 099 1.25 1.03 0.93 2.00 1.29 1971 0.91 1.37 1.02 1 04 1.66 1.38 1972 0 73 1.44 0.92 0 85 1.73 0.72 1973 0.39 0.94 0.72 0.42 0.95 0.60 1974 0.43 0.94 0.72 0.83 1 02 0.88 1975 0.49 1.05 0.94 1.03 1.40 0.83 1976 0 67 1.04 0.92 0 97 1.63 1.03 1977 0.67 1.54 1.26 0.90 2.09 1.07 1978 0.69 1.76 1.37 1.15 1.62 1 52 1979 0.79 I 54 1.14 1.03 1.44 0.93 1980 0.79 1.81 1.18 1.12 1.71 0.96 1981 0.74 2.04 1.40 1.42 2.03 0.89 1982 1.17 3.93 1.82 1.84 1.96 1.00 1983 1 16 1.86 1 62 2.21 1.68 1.74 Sources. 1970-82. Garcia-Garcia, "Aspects of Agricultural Development"; the 198 3 estimate is based on DNP figures Table SA-18. Index of Real Peso Value off o.b. International Prices of Selected Agricultural Commodities, 1970-83 (1975 = 100) Year Butter Beef Barley Corn Rice Wheat Sugar Palm oil Coffee Sorghum Soybeans Bananas Tobacco Cotton Sisal 1970 64.1 117.6 59.8 67.4 62.5 56.7 25.3 83.6 95.7 68.4 73.4 93.7 106.3 75.3 36.1 1971 88.1 139.3 54 2 65.8 61.7 61.7 30.0 82.0 80.9 64.1 77.2 77.4 95.4 86.1 39.5 1972 98.4 174.5 91.8 61.3 67.5 65.4 47.3 66.3 89.3 66.3 83.3 86.3 101.0 89.5 54.0 1973 72.3 215.5 104.4 96.5 111.6 116.5 55.4 103.6 104.8 104.2 155.8 79.5 94.8 137.9 106.9 1974 76.2 223.5 87.0 1142 137.3 133.1 151.3 161.4 98.6 117.0 127.4 77.7 95.2 127.1 188.3 1975 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 1976 94.9 106.5 77.4 84.1 65.9 79.7 50.0 84.5 172.6 84.0 93.2 94.0 91.1 130.0 72.2 1977 85.7 83 3 64.0 58.4 58.3 50.8 28.9 92.6 215.7 58.0 92.8 81.5 81.4 98.6 64.9 1978 101.2 107.4 49.1 56.1 63.5 157.1 25.4 95.2 151.0 57.3 80.7 77.7 79.7 90.3 56.2 1979 105.9 127.0 67.8 56.5 53.2 62.8 27.5 90.9 131.2 61.5 78.5 77.5 75.9 84.9 71.1 1980 106.9 105.9 59.6 53.4 60.3 58.9 71.0 70.6 111.4 67.3 68.0 77.7 69.9 89.9 67.1 1981 93.4 89.0 73.1 52.2 64.5 56.0 39.3 64.9 74.9 61.6 62.2 78.0 73.9 75.6 54.3 1982 81.5 81.0 65.4 40.7 39.4 48.4 18.4 47.7 77.0 49.5 49.7 68.6 79.3 61.5 46.3 1983a 71.1 84.6 n.a. 52.2 41.6 48.5 18.9 54.9 74.2 n.a. 58.5 80.35 82.3 72.9 44.7 a. Preliminary estimate. Sources: Garcia-Garcia, "Aspects of Agricultural Development"; and DANE. STATISTICAL APPENDIX 235 Table SA- 19. Index of the Ratio off o.b. International Prices in Pesos to Production Costs of Some Agricultural Commodities, 1970-82 (1975 100) Year Rice Corn Sorghum Soybeans Wheat Cotton 1970 45 4 36-3 69.5 98.6 47.8 76.7 1971 60.8 34.2 75 5 96.1 50.2 94.5 1972 63.6 35.5 74.0 104.9 64.4 126.5 1973 97.0 59.0 98.8 215.5 139.3 184.5 1974 146.6 111.9 112.5 139.1 147.1 147.2 1975 100.0 100.0 100.0 100.0 100.0 100.0 1976 74.6 86.1 78.7 94.3 85.2 132.9 1977 75.9 63.4 61.0 80.3 67 5 88.0 1978 77 3 55 6 61.1 80.5 73.5 48.9 1979 65.7 65.3 75.8 86.6 103.7 87.0 1980 61.6 54.2 69.9 78.4 87.0 110.1 1981 66.3 47.1 54.6 664 65.1 87.6 1982 40.8 40.8 47.8 55.7 74.7 n.a. n-a. Not available. Source: Garcia-Garcia, "Aspects of Agricultural Development." Table SA-20. Ratio of Basic Prices to Market Producer Prices of Selected Agricultural Commodities, 1970-81 Product 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 Barley 135.1 114.1 104.5 80.3 89.7 92.4 79 8 86.1 77.8 67 1 86.1 123.3 Beans 113 1 65.4 65.4 50.8 39.7 74.8 67.8 59.7 60.3 39.0 45 8 n.a. Cocoa 92 5 94.1 90.9 74.5 65.0 1062 98 8 67.0 64.0 54.1 77 9 95.9 Corn 87.2 76 7 66.8 73 6 93.6 94.1 82.0 55.7 70.1 49.3 51.2 91.8 Cotton fiber 126.8 106.2 94.1 89.9 82.9 106 8 68.6 83.9 94.8 68.3 102.3 102.0 Cottonseed 131.3 95.0 93.4 105.9 80.2 97.5 80.9 72,1 63 2 50.6 73.6 100 7 Palm oil 97.1 93.9 90.2 61.8 41 1 38.4 34.9 27.0 23.5 18.6 36.6 103.9 Rice (paddy) 109.2 104.6 107 3 80.4 67.7 101 1 101.4 66.7 67.9 58.2 73.5 88.9 Sesame 950 93.0 87.4 74.6 44.2 101.0 87.5 74.6 61.5 49.4 94.5 94.8 Sorghum n.a. 84.9 57.1 42.1 85.0 100.0 87.6 67.0 76.1 58.3 60.3 86.7 Soybeans 88.3 85 2 81.2 64.4 82.4 1009 86 9 57.8 58.7 47.7 60 5 111.9 Wheat 108.6 1086 89.5 859 53.4 89 7 90.4 82.5 876 69.9 73.3 924 n a. Not available. Source: Calculations by the DNP. 236 STATISTICAL APPENDIX Table SA-2 1. Effective Rate of Interest for Banks on Bonos de Prenda, 19 70-81 Market interest Effective rate Year rate' bonos de prendab 1970 13.3 22.0 1971 16.4 22.0 1972 15.6 22 0 1973 20.3 22.0' 1974 304 22.4 1975 23.8 22.6 1976 22.4 22.6 1977 22.9 22.6 1978 25.9 22.6 1979 36.5 22.6 1980 41.5 29.8 1981 52.5 32.1 a CAT, 120-day maturity, average annual rate. b R, = [R,, - (R,)(M,)]/(1 - M), where R, = effective rate of interest, Rm = market rate of in- terest, R, = rediscount interest rate, M, = margin of rediscount Sources: Calculations by the DNP, based on Banco de la Republica, Resoluciones de la Junta Monetaria; Asobancaria. Table SA-22. Value of Discounts and Rediscounts under the Bonos de Prenda System and the Distribution of the Cost of the Implied Subsidy, 1981 (thousand pesos) Value Value re- Total Paid by Paid by Product discounted discounted cost' governmentb banks Beans 6,759 1.660 1,602 476 1,126 Corn 225,875 80,421 53,532 23,081 30,451 Cotton fiber 1,995,175 758,997 472,856 217,832 255,024 Cottonseed 568,298 215,353 134,687 61,806 72,881 Rice(paddy) 1,419,137 475,056 336,335 136,341 199,994 Sesame 45,561 22,382 10,798 6,424 4,374 Sorghum 897,613 325,966 212,734 93,552 119,182 Soybeans 681,131 264,781 161,428 75,992 85,436 Tobacco 1,066,679 238,222 252,803 68,370 184,433 Other agricultural productsd 483,461 168,969 114,580 48,494 66,086 Other products' 2,329,949 115,373 552,198 33,112 519,086 Total 9,719,638 2,667,180 2,303,553 765,480 1,538,073 a. Total cost = value rediscounted (r. - r), where r, = market interest rate and r, = subsidized interest rate for bonos de prenda. b. Amount paid by government = value rediscounted - (r, - r,), where r, = rediscount interest rate. c. Amount paid by banks = total subsidy less amount paid by government d. Barley. cocoa, malt, and wheat. e. Manufactured products and products for export. Source World Bank estimates. STATISTICAL APPENDIX 237 Table SA-2 3. Supply and Distribution of Export-Grade Green Coffee, 1958/59 through 1983/84 (thousand sixty-kilogram bags) Exports Non-ICA Coffee Stocks Domestic ICA quota quota Stocks year' carry-in Production" consumption markets markets Total carry-out' 1958/59 11 7,442 908 6,372 59 6,431 114 1959/60 114 7,648 1,197 5,597 74 5,671 894 1960/61 894 7,500 1,270 5,990 53 6,043 1,081 1961/62 1,081 8,035 1,526 5,536 58 5,594 1,996 1962/63 1,996 7,500 1,416 5,952 104 6,056 2,024 1963/64 2,024 7,800 1,375 6,228 82 6,310 2,139 1964/65 2,139 8,547 1,354 5,612 131 5,743 3,589 1965/66 3,589 8,224 1,202 5,670 195 5,865 4,746 1966/67 4,746 7,507 1,250 5,421 213 5,634 5,369 1967/68 5,369 7,995 1,270 6,344 251 6,595 5,499 1968/69 5,499 7,375 1,290 6,204 330 6,534 5,050 1969/70 5,050 8,266 859 6,467 407 6,874 5,583 1970/71 5,583 6,872 989 6,008 322 6,331 5,135 1971/72 5,135 5,958 1,035 6,198 289 6,487 3,571 1972/73 3,571 8,564 1,046 6,046 209 6,255 4,834 1973/74 4.834 7,066 1,252 6,873 535 7,408 3,240 1974/75 3,240 7,981 1,279 7,102 440 7,542 2,400 1975/76 2,400 7,804 1,369 6,554 469 7,023 1,812 1976/77 1,812 8,939 1,305 4,891 401 5,292 4,154 1977/78 4,154 10,463 1,420 7,144 414 7,558 5,639 1978/79 5,639 12,300 1,638 10,714 717 11,431 4,870 1979/80 4,870 11,848 1,728 10,692 848 11,540 3,450 1980/81 3,450 13,037 1,478 8,310 721 9,031 5,978 1981/82 5,978 12,893 1.592 8,052 938 8,990 8,289 1982/83 8,289 12,810 1,695 8,465 709 9,174 10,230 1983/84 10,230 13,464 1,553 9,130 836 9,966 12,175 a. October through September, b. Series deduced from data on stocks, consumption, and exports. c. Stocks include private holdings Source: FEDERACAFE. 238 STATISTICAL APPENDIX Table SA-24. The Direct Share of Main Agricultural Entities in National Budget Allocations, 1970-83 (million pesos and percent) Partici- pation of public Agriculture sectora Public agencies agencies in agriculture in total Total budget Partici- sector onlyb agricul- pation ture Constant Constant in total Constant sector Current 1970 Current 1970 budget Current 1970 budget pesos pesosc pesos pesos, (percent) pesosd pesosc (percent) Year (1) (2) (3) (4) (3 + 1) (5) (6) (5 +3) 1970 20,644.2 20,644.2 5,186.9 5,186.9 25.1 1,518.7 1,518.7 29 1971 25,522.1 22,446.9 5,413.9 4,761.6 21.2 1,287.3 1,132.2 24 1972 31,279.5 24,191.4 6,388.0 4,940.4 20.4 1,550.3 1,199.0 24 1973 38,492.0 24,753.7 8,103.2 5,211.1 21.1 1,764.3 1,134.6 22 1974 50,726.5 26,255.9 8,688.6 4,497.2 17.1 2,071.8 1,072.4 24 1975 60,719 5 25,384.4 7,124.4 2,978.4 11.7 2,696.0 1,127.1 38 1976 86,185.9 30,050.9 12,852.4 4,481.3 14.9 2,715.6 947.2 21 1977 112,805.7 32,763.8 16,738.9 4,861.7 14.8 3,348.3 972.5 20 1978 174,875.7 40.830 2 16,972.0 3,962.6 9.7 3,993.0 932.3 24 1979 234,160.1 42,404.9 20,693.7 3,747.5 8.8 5,448.0 986.6 26 1980 332,382.3 47,496.8 28,277.4 4,040.7 8.5 7,054.7 1,008.1 25 1981 438,678.7 50,020.4 33,218.3 3,787.7 7.6 8,990.1 1,025.1 27 1982 574,404.5 55,912 4 40,834.2 3,974.8 7.1 10,679.0 919.8 26 Note. Tables SA-24 through SA-31 were prepared by Aichin Wee and Kei Kawabata. a. The agriculture sector includes the Ministry of Agriculture (Direcci6n Superior); its ascribed agen- cies (Entidades Adscritas), the ICA, INCORA, INDERENA, and HiMAT; agriculture-related investments of the autonomous regional corporations, essentially the cvc, Corpouraba, CAR, and Codechoco; and other public enterprises or entities (Entidades Vinculades), mainly IDEMA, COFIAGRO, and EMCOPER, that at various times receive investment funds from budget allocations. The figures include investment funds (Inversion) from both budget allocations (Presupuesto Nacional) and own resources (Recursos Proprios) and recurrent funds (Funcionamiento) from both the budget and own resources. b. Includes only the Entidades Ascritas-CA, INCORA, INDERENA, and HiMAT-which account for about 2 3 percent of total public allocations in the agriculture sector (see table SA-4). c. Deflated, using the implicit price deflator for current government purchases of goods and services, with assumed annual rates of inflation for 1982-86. d. Figures for 1970-81 are derived from the actual figures for 1982 and 1983 in column 6 and from projected allocations for investment and recurrent expenditures for 1984-86 supplied by OPSA. Sources: Based on calculations of the Ministry of Agriculture and OPSA. STATISTICAL APPENDIX 239 Table SA-25. Public Allocations to Agriculture, by Entity, 1970-81 (million constant 1970 pesos and percent) Ascribed agencies (ICA, INCORA, HIMA T, Total agriculture INDERENA) Regional corporations Otber entities Million Million Million Million Year pesos Percent pesos Percent pesos Percent pesos Percent 1970 5,186.9 100.00 1,518.7 29.28 219.9 4.24 3,448.3 66.48 1971 4,814.0 10000 1,132.2 23.52 477.2 9.91 3,204.6 66.57 1972 4,977.0 100.00 1,199.0 24.09 634.7 12.75 3,143.3 63 16 1973 5,211.1 100.00 1,134.6 21.77 839.1 16.10 3,237.4 62 13 1974 4,497.2 100.00 1,072.4 23.85 759.5 16.89 2,665 3 59.26 1975 2,988.4 100,00 1,127.1 37.72 532.9 17.83 1,328.4 44.45 1976 4,481.3 100.00 947.2 21.14 628.7 14.03 2,905.4 64.83 1977 4,315.4 100.00 972.5 22.53 623.0 14.44 2,719.9 63.03 1978 4,081.0 100.00 932.3 22.84 666 7 16.34 2,482.0 60.82 1979 3,856.1 100.00 986.6 25.59 793.6 19.91 2,101.7 54.50 1980 4,214.2 100.00 1,008.1 23.92 1,064.5 25.26 2,141.6 50.82 1981 3,787.7 100.00 1,025.1 27.06 1,179.7 31.15 1,582.9 41.79 Source. Ministry of Agriculture, Oficina de Planeamiento del Sector Agropecuario, on the basis of data from the Contraloria General de la Republica. Table SA-26. Area under Cultivation, by Principal Crop, 1971-84 (thousand hectares) Crop 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983' 1984a Cereals Rice (paddy) 241.8 258.2 291.0 354.5 372.5 365.6 324.4 406.1 442.0 415.8 420.7 446.0 396.5 364.1 Barley 55.6 63.5 52.4 59.1 75.6 68.0 46.6 68.4 73.9 62.8 36.0 34.9 17.6 17.4 Wheat 46.9 60.7 56.5 45.1 30.1 32.8 33.0 29.7 307 37.6 44.0 45.3 46.3 42.9 Corn 666.5 624.5 580 3 570.1 572.7 647.5 580.5 670.9 615.6 614.4 629.0 636.0 582.3 593.0 Sorghum 92.1 84.0 135.4 151.2 134.0 173.6 189.5 224.8 221.2 206.0 231.3 291.1 271.9 237.6 Other food crops Yuca 248.8 251.3 249.8 250.8 256.7 223.3 218.3 216.9 221.7 207.7 207.0 170.0 172.7 183.0 Potatoes 88.3 89.5 98.6 92.0 110.0 125.0 130.0 141.6 148.0 142.0 159.5 165.2 160.5 160.7 Plantain 324.9 324.8 326.7 327.9 341.0 340.1 386.3 400.1 412.1 432.6 433.0 358.2 356.8 366.9 Panela 183.0 188.0 194.0 196.9 173.5 171.5 178.9 197.8 200.0 209.0 187.0 170.7 176.5 186.2 0 Beans (common) 68.0 84.6 87.0 90.7 120.7 101.0 115.8 110.9 112.4 115.4 117.3 112.2 112.5 109.6 Cocoa 49.0 52.6 54.9 57.9 52.6 54.5 57.5 60.5 62.7 64.1 68.0 77.3 84.6 89.3 Cotton and oilseeds Seed cotton 219.0 242.3 250.8 258.4 280.7 285.6 377.2 327.9 186.5 216.9 221.1 99.2 76.7 140.8 Soybeans 55.1 54.0 54.0 57.0 87.8 37.6 56.7 69.0 71.3 78.1 43.9 49.4 59.5 50.6 Sesame 470 43.2 37.0 32.2 41.6 36.1 23.7 24.9 27.7 24.2 19.4 12.3 9.0 8.3 African palm oil 13.8 15.0 16.5 18.2 15.7 16.0 17.4 19.0 21.8 24.6 25.2 31.6 35.4 38.3 Other export crops Sugarcaneb 64.0 72.9 78.6 75.1 75.7 83.0 76.5 86.5 91.1 93.2 92.1 92.9 94.8 93.4 BananaSb 14.0 15.7 13.3 14.9 14.2 16.3 19.5 20.8 22.0 20.9 21.0 21.8 22.5 23.1 Tobaccob.c 23.0 26.3 26 2 25.5 34.1 29.7 33.3 28.8 30.6 28.5 30.1 29.0 29.8 27.3 a. Preliminary estimate. Ib Refers to calendar year. c Combination of white and dark tobacco. Sources Ministry of Agriculture, Oficina de Plancamiento dcl Sector Agropecuario, and the oN. Table SA-2 7. Production of Principal Crops, 1971-84 (thousand metric tons) Crop 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983a 19841 Cereals Rice (paddy) 851.9 997.5 1,151.1 1,540.4 1,614.0 1,560.0 1,307.0 1,714.7 1,932.5 1,797.9 1,798.7 2,018.2 1,779.8 1,695.8 Barley 107.2 98.0 81.5 96.9 121.8 71.4 81.3 118.9 136.6 109.5 56.4 55.6 27.8 28.2 W heat 53.2 69.2 72.4 58.8 38.9 45.3 38.5 37.7 420 45.7 62.3 70.7 77.8 59.3 Corn 818.5 806.2 739.1 791.5 722.6 883.7 752.8 862.2 870.2 853.6 880.0 898.5 863.8 864.3 Sorghum 239.6 210.0 280.2 336.6 335.0 427.7 406.2 516.7 501.3 430.5 532.0 568.4 595.2 589.6 Other food crops Yuca 1,990.4 2,010.4 1,998.4 2,125.9 2,021.1 1,845.7 1,972.6 2,044.1 1,908.9 2,150.4 2,150.1 1,552.3 1,554.8 1,674.5 Potatoes 868.9 823 4 1,030.5 1,012.0 1,320.0 1,515.8 1,608.5 1,995.6 1,966.1 1,726.7 2,006.1 2,149.0 2,186.7 2,462.9 Plantain 1,517.3 1,562.3 1,653.1 1,678.9 1,791.7 1,852 0 1,844.0 2,192.0 2,235.8 2,348.0 2,400.0 1,993.0 2,247.9 2,277.4 Panela 457.0 508.0 524.0 557.2 805.6 833.6 837.6 965.4 9847 987.8 802.6 734.8 779.6 824.6 Beans (common) 35.6 61.1 56.9 67.1 89.9 67.6 74.9 74.8 74.7 83.6 79.3 72.8 81.8 80.1 Cocoa 19.0 20.0 22.0 23 0 21.2 29 2 27.0 31.0 32.3 35.7 38.3 39.4 38.6 41.6 Cotton and oilseeds Cotton 322.4 412.1 344.8 420.3 400.9 408.6 4804 330.3 281.6 353.2 366.2 153.6 130.4 243.3 Soybeans 100.7 104.6 97.2 1140 168.9 75.1 102.9 1308 143.6 1545 89.0 98.8 122.4 94.1 Sesame 31.4 28.3 18 1 17.2 20.7 20.7 13.0 13.7 15.6 12.9 11.6 7.2 4.9 4.9 African palm oil 36.2 41.4 44.0 50.8 39.2 38.6 48.1 52.6 59.6 70.0 79.9 85.2 101.9 118.2 Other export crops Sugarcaneb 744.0 823.7 809.9 894.8 969 7 934.6 853 3 1,025.9 1,096.0 1,188.6 1,148.1 1,302,9 1,330.0 1,177.6 BananaSb 351 0 282.0 301.0 469.7 559.0 521.5 593.1 719.0 800.5 944.3 1,109.6 1,146.6 1,043.4 1,106.1 Tobaccob., 39 3 36.1 39.7 41.1 57.6 38.6 58.3 46.6 51.4 46.3 49.4 48.7 47.9 43.8 a. Preliminary estimate. b. Refers to calendar year c. White and dark tobacco. Sources: Ministry of Agriculture, Oficina de Planeamiento del Sector Agropecuario and the DNP. Table SA-2 8. Yields of Principal Crops, 1971-84 (metric tons per hectare) Crop 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983a 1984a Cereals Rice (paddy) 3.5 3.9 4.0 4.3 4.3 4.3 4.0 4.2 4.4 4.3 4.2 4.5 4.5 4.6 Barley 1.9 1.5 1.6 1.6 1.6 1.1 1.7 1.7 1.8 1.8 1.6 1.6 1.6 1.6 W heat 1.1 1.1 1.3 1.3 1.3 1.4 1.2 1.3 1.4 1.2 1.4 1.6 1.7 1.4 Corn 1.2 1.3 1.3 1.4 1.3 1.4 1.3 1.3 1.4 1.4 1.4 1.4 1.5 1.5 Sorghum 2.6 2.5 2.1 2.2 2.5 2.5 2 1 2.3 2.3 2.1 2.3 2.0 2.2 2.5 Other food crops Yuca 8.0 8.0 8.0 8.5 7.9 8.3 9.0 9.4 8.6 10.4 5.5 9.1 9.0 9.1 Potatoes 9.8 9.2 10.5 11.0 12.0 12.1 12.4 14.1 13.3 12.2 13.2 13.0 13.6 15.3 Plantain 4.7 4.8 5.1 5.1 5.3 5.4 4.8 5.5 5.4 5.4 5.5 5.6 6.3 6.2 Panela 2.5 2.7 2.7 2.8 4.6 4.9 4.7 4.9 4.9 5.0 4.3 4.3 4.4 4.4 Beans (common) 0.59 0.62 0.61 0.64 0.7 0.7 0.6 0.6 0.7 0.7 0.6 0.6 0.7 0.7 Cocoa 0.39 0.38 0.40 0.40 0.35 0.54 0.47 0.51 0.5 0.5 0.6 0.5 0.5 0.5 Cotton and oilseeds Seed cotton 1.5 1.7 1.3 1.6 1.4 1.4 1.3 1.0 1.5 1.6 1.7 1.5 1.7 1.7 Soybeans 1.8 1.9 1.8 2.0 1.9 1.0 1.8 1.9 2.0 2.0 2.0 2.0 2.1 1.9 Sesame 0.67 0.66 0.49 0.53 0.5 0.6 0.5 0.6 0.6 0.5 0.6 0.6 0.5 0.6 African palm oil 2.6 2.8 2.7 2 8 2.6 2.4 2.5 2.5 2.7 2.8 3.2 2 7 2.9 3.1 Other export crops Sugarcane 11.6 11.3 10.3 11.9 12.8 11.3 11.2 11.9 12.0 12.8 12.5 13.5 14.0 12.6 Bananas 25.1 18.0 22.6 31.5 39.4 32.0 30.4 34.6 35.6 45.0 52.8 52.7 46.4 47.9 Tobaccob 1.7 1.4 1.5 1.6 1.7 1.3 1.8 1.6 1.9 1.6 1.6 1.6 1.6 1.6 a. Preliminary estimate. b. Dark tobacco. Source. Ministry of Agriculture, Oficina de Planeamiento del Sector Agropecuario. Table SA-2 9. Gross Domestic Product and Gross Output, by Main Economic Activity, 1970-83 (million current pesos) Gross domestic product Gross output Pergamino Processed Processed coffee coffee Sugar Rest of the coffee Sugar Rest of the Year Agriculturea (01) (08) (12) economy Total Agriculture' (08) (12) economy Total 1970 32,052 4,417 3,561 581 96,574 132,768 37,321 10,116 1,358 171,694 220,489 1971 35,396 3,919 3,584 615 116,291 155,886 41,677 9,850 1,594 208,863 261,984 1972 44,034 5,192 4,555 850 140,175 189,614 51,908 12,404 2,142 248,388 314,842 1973 56,460 6,690 7,761 982 177,957 243,160 67,638 17,354 2,532 313,716 401,240 1974 75,441 7,841 6,960 1,985 237,998 322,384 92,345 19,281 4,573 432,722 548,921 1975 93,164 8,971 12,816 2,510 296,618 405,103 113,991 26,629 5,949 537,773 684,342 1976 120,908 17,017 23,839 2,553 384,970 532,270 145,451 45,723 6,078 693,691 890,943 1977 172,900 30,898 33,831 4,485 504,813 716,029 202,674 64,530 8,803 900,252 1,176,259 1978 201,111 35,426 33,906 5,116 669,354 909,487 239,178 84,955 10,706 1,176,242 1,511,081 1979 244,477 41,098 33,399 7,647 903,294 1,188,817 294,841 96,222 14,599 1,570,314 1,975,976 1980 292,524 47,269 53,060 16,491 1,217,055 1,579,130 361,411 125,627 26,815 2,107,969 2,621,822 1981 364,760 57,489 26,264 16,625 1,560,120 1,982,773 452,292 112,216 28,875 2,707,970 3,301,353 1982 445,387 63,256 39,218 19,292 1,923,401 2,497,298 553,008 121,790 35,585 3,407,525 4,117,908 1983 539,379 n.a. 47,233 25,627 2,424,422 3,036,661 666,824 146,565 46,972 4,104,222 4,964,583 n.a. Not available. a. Consists of pergamino coffee (01), other agricultural products (02), and animal products (03) b. Preliminary estimate. Sources: DANE, Cuentas nacionales de Colombia: 1970-1982 (BogotA, August 1984), and unpublished information for 1982 and 1983 Table SA-30, Gross Domestic Product and Gross Output, by Main Economic Activity, 1970-83 (million 1975 pesos) Gross domestic product Gross output Pergamino Proccssed Processed coffee coffee Sugar Rest of the coffee Sugar Rest of the Year Agriculture' (01) (08) (12) economy Total Agriculture' (08) (12) economy Total 1970 75,3 38 8,370 10,062 1,088 221,008 307,496 91,119 20,850 3,824 400,178 515,971 1971 76,195 7,784 10,344 1,520 237,766 325,825 92,155 21,434 4,070 437,141 554,800 1972 81,565 8,595 10,468 1,739 257,041 350,813 98,650 21,691 4,566 464,142 589,049 1973 83,354 8,636 10,787 1,881 278,376 374,398 100,814 22,190 4,546 498,935 626,485 1974 87,918 8,793 11,020 2,006 294,966 395,910 108,250 22,911 5,009 536,693 672,863 1975 93,164 8,971 12,816 2,510 296,618 405,108 113,991 26,629 5,949 537,773 684,342 1976 95,839 9,386 10,830 2,136 315,458 424,263 116,801 23,552 5,074 567,940 713,367 1977 98,946 10,853 8,909 2,049 332,002 441,906 120,587 19,288 4,842 595,366 740,083 1978 107,088 11,852 14,193 2,343 355,711 479,335 130,157 30,728 5,583 639,547 806,015 1979 112,379 13,482 17,189 2,738 372,813 505,119 136,736 37,3 35 6,547 671,800 852,418 1980 114,849 13,945 17,565 3,085 390,266 525,765 140,256 38,140 7,458 703,254 889,108 1981 118,667 15,391 14,799 2,819 401,536 537,821 144,316 31,943 6,767 724,446 907,472 1982 116,149 16,081 15,045 3,397 408,245 542,836 142,401 32,188 7,951 728,095 910,635 1983b 118,153 iisa. 15,801 3,491 410,610 548,055 144,603 33,508 8,086 727,175 915,372 n.a. Not available. a. See note a, table SA-29. b. Preliminary estimate. SO'UTCeS: DANE, Cuentas nacionales de Colombia. 1970-1982 (Bogot6, August 1984), and unpublished information for 1982 and 198 3 and extraction of pergamino coffee (0 1). Index Adjustment, economic. See Macroeconomic pol- Agricultural Development Bank. See Caja Agraria icy (adjustment and) Agricultural equipment, 47, 136, 137, 153; rental Ad valorem export tax, 101, 110, 111, 112, 113 of, 147, 154 AGD. See Almacenes Generales de Dep6sito Agricultural fund. See Fondo Financiero Agro- Agricultural commodities: domestic consumption pecuario of, 9, 62, 63; domestic demand for, 7, 22, 62, Agricultural income, 61, 63, 66, 69-70, 81, 82, 81, 83; domestic prices of, 12, 46, 54, 66, 67, 83, 108, 143-44, 147, 152 69, 73, 74, 76, 77, 82, 83, 87, 97, 98; domestic Agricultural investment, 14, 61, 81, 82, 123-34, supply of, 22; employment share of, 5; export 136, 141, 159, 161-62. See also Infrastructure earnings share of, 5; export prices of, 5, 12, 34, (agricultural) 67, 69, 84, 123; export quotas of, 62; exports Agricultural loans, 132-33, 145-49, 161 of, 3, 4, 5-6, 7, 9-10, 22, 23n8, 28, 30, 33- Agricultural Marketing Institute. See Instituto de 35, 38, 41, 43, 62, 67, 72, 77, 162; export sub- Mercadeo Agropecuario sidies for, 62, foreign demand and, 5, 9, 22; for- Agricultural practices, 140. See also Coffee pro- eign exchange and, 3, 5, 10, 86; growth rate duction technology and, 3, 5, 6, 7, 10, 144, 156, 158, 161; import- Agricultural trade policies. See Decree, Plan Va- competing commodity price of, 67; import lejo; Resolution prices of, 63, 68-69, 70, 71; imports of, 9-10, Almacenes Generales de Dep6sito (AGD), 88, 93 14, 46, 47, 54, 61, 62, 70, 72, 137, 160; import Almagrario (agricultural storage), 93, 99n8 trade restrictions on, 62; incentives and, 4, 61, Almapopular (agricultural storage), 93, 99n8 77, 156, 161; inputs to, 46, 47, 61, 63, 69, Andean Group, trade with, 28, 48 136-40, 151, 152, 154; processing costs of, 69; Andean Pact, 44, 48 producer prices of, 14, 62, 70-71, 74, 82, 149; Asociaci6n Colombiana de Productores de Flores production of, 4-5, 9, 10, 14, 21, 22, 28, 43, (ASOCOLFLORES), 131 46, 47, 61, 66, 68, 70, 72, 73, 76, 81, 136-37, Asociaci6n Colombiana de Productores de Semi- 141-42, 161; production costs of, 14, 15, 22, Ilas(ACOSEMILLA), 131 47, 62, 63, 72, 83, 136, 145, 147, 154; produc- Asociaci6n de Cultivadores de Cafia de Azicar de tion incentives and, 71-73, 77, 84, 157, 158, Colombia (ASOCARA), 62, 134 161; relative prices of, 13, 61, 73-76, as share Asociaci6n Latino Americana de Desarrollo y ofGDP, 5, 6, 9, 10; stocksof, 151;surplusof, 41, Integraci6n (ALADI), 48-49 62, 77; value added in, 46, 137; yields of, 14, Auctioning. See Licenses (auctioning of) 15, 52, 137, 141, 152. See also Marketing; Research and extension; Storage (agricultural); Balance of payments, 20, 43, 44; current account Technology; specfic commodities of, 3, 19, 26, 157; factors affecting, 9, 10, 16, 245 246 INDEX Balance of payments (continued) Coffee: boom in (1976-80), 3, 7, 21, 29, 61. 104, 19, 26, 47; projection of, 22, 158; surplus in, 5, 108, 157, 158; demand for, 5, 22, 117, 160; 15 disease of, 109, 118, 160; diversification m, Bananas, 9, 28, 34, 62, 67, 72, 77, 137, 140, 150 103-04, 114-16, 117, 118, 132, 151, 159, Banco Cafetero, 107, 146, 161 160; domestic prices of, 11, 74, 80, 82, 106, Banco Ganadero, 146, 161 107, 108, 117, 159; export quotas on, 62, 106, Banking, 9, 91, 107, 132-33, 158; storage and, 109, 110, 116, 117, 118; exports of, 7, 9, 10, 88-90, 92 11, 34, 105, 108, 109, 110, 113, 118, 156, ex- Barley, 88, 89, 91, 131; imports of, 46, 137; port share of, 26, 102, 105; export taxes on, 62, prices of, 46, 67, 80, 82; price supports for, 62, 101, 110-13; guaranteed price of, 110, 111, 63, 160; producer prices of, 71; production of, 117, 159; input price of, 145, 160; labor force 72, 137; stocksof, 151 and, 101, 103-04; land area in, 101, 108, 115; Basic price, 92-93, 94, 97 marketing of, 106; prices of, 11, 12-13, 15, 17, Beans, 62, 63, 71, 80, 88, 93, 151 20, 21, 22, 26, 29, 34, 61, 67, 72, 74, 102, Beef, 34, 67, 72, 78 104-05, 107, 108, 109, 110-12, 117-18, Bolsa Nacional Agropecuaria, S.A., 87, 151 159-60; price stabilization for, 82; production Bonos de prenda (BP), 88, 89, 90, 91, 92, 94, 97, of, 11, 72, 101-02, 105, 108, 109, 113, 114, 98, 124 115, 117, 118, 137, 160; production costs of, Booming sector syndrome, 10 115, 116; research and, 140; retention quota Border prices, 47 for, 110, 111; revenues from, 17, 19, 101, 107, Buffer-stock scheme for stabilization, 83 113; as share of GDP, 101; stocks of, 22, 101, Butter, 67, 72 105, 107, 108, 109-10, 113, 116, 117, 159; storage of, 88, 108, 115; value of, 102, 104, Caja Agraria, 124, 132, 134, 145-46, 147-49, 110; varieties of, 102-03, 104, 108, 109, 159, 150, 151-52, 154, 161 yields of, 102, 108, 137. See also Coffee produc- Capital costs, 37, 78 tion technology Capital flows, 26, 158 Coffee Congress (1984), 118 Capital goods, 45, 54 Coffee Federation. See Federaci6n Nacional de Capital-market constraints, 4, 157 Cafeteros de Colombia Castillo Nifio, Alvaro, 96 Coffee leaf rust. See Roya CAT. See Certificado de Abono Tributario Coffee production technology, 102-04, 106, Caturra (variety of coffee tree), 102, 103, 108, 109, 113-14, 117, 160 113, 114, 117, 143, 159 COFIAGRO. See Corporaci6n Financiera de Fomento Cauca Valley, 126-27, 141, 153 Agropecuario y de Exportaciones CD. See Certificado de Dep6sito Colombia, 4, 5; economy of, 3-7, 17, 19, 21, 36. Central Bank, 29, 38, 67, 88, 90, 91, 92, 110, 157-59; Ecuador and, 44, 157; fiscal year of, 111, 113, 124, 127, 133, 148, 154, 158, 161 15; recession in, 7, 17, 29; Venezuela and, 11, Central de Cooperativas de la Reforma Agraria, 44, 157 Ltda. (cECORA), 151 Colombian Port Authority (CoLPuERTos), 145, 153 Central WholesaleMarketofBogoti (CORABASTOS), Competitiveness, 19, 21, 22, 28, 30, 53, 61, 68, 151 72, 73, 77, 140 CEREALS, 46, 68, 70, 76, 159 Concertation (approach to legislation), 132 CERT. See Certificado de Reembolso Tributario Consejo Nacional de Politica Econ6mica y Social Certificado de AbonoTributario (CAT), 27, 28, 31, (CONPES), 38 32, 68, 83, 125 Conservation of natural resources, 141, 142, 143, Certificado de Dep6sito (CD), 88 152 Certificado de Reembolso Tributario (CERT), 31, Constant prices, 19, 30 32, 62, 83, 84, 85, 87, 97 Consumer goods, 45, 54 Clearing price, 56 Consumer price index, 31 Coal, 20, 39, 158 Consumer prices, 71, 80-81, 83, 94, 97, 98, 149, Cocoa, 62, 68, 78, 80, 88, 91, 115, 133 153 INDEX 247 Cooperatives, 132, 151 tural commodities (domestic prices of); Coffee Corn, 88, 140; domestic consumption of, 63; im- (prices of); Sugar (prices of) ports of, 46, 47, 80, 137; prices of, 67, 78, 82; Dmd. See Programa de Desarrollo Rural Integrado price supports for, 62, 63, 151, 160; producer Drugs, illegal, 5, 10, 143 prices of, 71; production of, 68, 72, 141; pro- Dutch disease, 17, 23-24n16 duction costs of, 72; storage of, 93 Corporaci6n de Abastos de Bogota (CORABASros), Ecuador, and Colombia, 44, 157 151 Effective rate of protection (ERP), 50, 69, 70 Corporaci6n Financiera de Fomento Agropecuario Efficiency, 61, 68, 70, 77, 84, 91, 154 y de Exportaciones (cOIAGRO), 124 Elasticity: of demand, 54, 82; of exports, 35; of Costs: of capital, 37, 78; of coffee production, supply, 70 115, 116; of corn production, 72; of cotton pro- Employment, 44, 53, 55, 143, 161; boom in, duction, 72; of distribution, 63; of fertilizer, (1970s), 5; growth of, 5, 143, 157 147, 150, 154; and inflation, 37; of labor, 78, Enterprise for Marketing of Perishable Products 137, 153; of rice production, 72, social, 147, (EMCOPER), 151 148-49; of sorghum production, 72; of soy- ERP. See Effective rate of protection bean production, 72; of wheat production, 72. Excelso(grade ofcoffee), 111, 115 See also Agricultural commodities (production Exchange rate, 4, 10, 15, 35, 36, 39, 61, 67, 68, costs of); Port-handling costs; Transport costs 76, 77, 154, 156, 158; competitive, 25, 26; Cotton, 34, 62, 72, 88, 91, 93, 140; CERT and, 62; domestic policies and, 4; equilibrium, 25, 26, export subsidies for, 62, 68; labor force for, 143; 33, 55; fluctuations, 25; management, 25; mul- prices of, 67, 68, 82; processing of, 69; produc- tiple, 26; nominal (NER), 30, 36, 40, 144; real tion costs of, 72; research and, 140; yields of, (RER), 6, 14, 15, 16, 17, 18, 19, 20, 21, 22, 26, 137 28, 30-31, 32, 33, 35, 36, 37, 38, 39, 40, 41, Crawling peg, 13-14, 15, 20, 25, 27, 30, 33, 35, 44, 47, 48, 56, 72, 76, 157, 158; real effective, 36, 37, 39-40, 154, 157 53. See also Crawling peg Credit, 123, 124, 131, 133, 142, 158, 161; allo- Export Promotion Fund (PROEXPO), 27, 28, 31, cation of, 92; countercyclical, 83; demand for, 38-39, 48, 53 15; domestic, 29; eligibility for, 146; and Ley Exports, 19, 27, 28, 29-30, 33, 35-36, 40; ag- 5a, 114, 115, 127, 133; as percentage of agri- gregate demand and, 7; boom in, 10; capacity cultural GDP, 146; PROEXPO and, 27, 38, 62; social for, 108; developing countries' share of, 38; costs and, 147, 148-49; subsidies for, 4, 14, 76, development of, 5, 10-11, 158; diversification 88-91, 93, 98, 136, 145, 146-47, 153-54, of, 26; European share of, 38; foreign demand 156, 159, 161; tax, 125 and, 25; incentives for, 15, 21, 22, 27, 31-32, Decree (Colombia): 444/67, 53; 637/84, 83; 33, 40, 48, 54; internal relative prices of, 12- 13, 22; labor-intensive, 5, 28, 44, 45, 143; loans 68m/6d7 53;rege 308, 4; ffor, 25; long-term elasticity of, 35; nontradi- Demand: aggregate, 18, 55; for agricultural com- tional, 27, 45, 53, 158; North American share modities, 5, 7, 9, 10, 22, 62, 81, 83; for coffee, o 7, rrm ce, 5, 30, 75ericeshr1 5, 22, 117, 160; for credit, 15; for exports, 7, 12 25; for imports, 21, 54; for labor, 45, 143, 144 1, 19, 28, 83; promotion of, 22, 25, 38-39, Departamento Administrativo Nacional de Esta- 41, 44, 45, 53, 76, 78; subsidies for, 26, 28, 32, distica (DANE), 131, 144 33, 53, 56, 61, 62, 83, 84; taxes on, 53, 54, 62, Departamento Nacional de Planeaci6n (DNP), 51, 67, 68, 83, 101; values of, 7, 11; worldwide 127, 145 volume of, 11. See also Agricultural commodities Department Development Corporations, 161 (exports of); Manufacturing (exports of) Development plan: 1966-70, 44; 1970-74, 28, External debt, 11, 33, 127 External sector, 3, 4, 5, 62 Discount margin, 89 Farmgate price, 67, 68, 69, 71, 107, 111, 145, Distribution costs, 63 160 Domestic prices, 6, 16, 21, 30, 35, 46, 47, 80, 82; Federaci6n Colombiana de Productores de Papa determinants of, 4, 43, 54, 63. See also Agricul- (FEDEPAPA), 131 248 INDEX Federaci6n Nacional de Arroceros (FEDEARROz), ICA. See Instituto Colombiano Agropecuario; 131, 134, 150 International Coffee Agreement Federaci6n Nacional de Cafeteros de Colombia ico. See International Coffee Organization (FEDERACAFE), 47, 80, 82, 102, 106, 107, 108, IDEMA. See Instituto de Mercadeo Agropecuario 110, 111, 112, 113, 114, 116, 117, 118, 131- HCA. See Inter-American Institute for Agricul- 32, 134, 150, 151, 159 tural Sciences Federaci6n Nacional de Cooperativas (FENALCO), Import deposit. See Prior deposit requirements for 132 imports Federaci6n Nacional de Cultivadores de Palma Import duty drawback scheme. See Plan Vallejo Africana (FEDEPALA1A), 131 Import policy, 25, 29, 43-56, 63, 76, 145, 156 Fertilizer, 47, 48, 103, 114, 136, 137, 153, 160; Imports, 19, 39; capacity for, 5; constant prices of, consumption of, 137, 144, 145; costs of, 147, 19; controls on, 4, 13, 14, 19, 26, 39, 41, 43, 150, 154; marketing of, 150; price controls on, 44, 53-56, 61, 76, 77, 83, 86, 98, 150, 156, 62; prices of, 145, 153, 160; subsidies on, 118, 158; decline of, 43, 48; demand elasticity of, 54; 159 domestic demand for, 21; domestic prices of, FFAP. See Fondo Financiero Agropecuario 54; duty-free, 25; exchange rate and, 26; growth Fiscal deficit, 159; growth of, 16-17, 18, 29, 76, rate of, 19, 45; incentives for, 15, 21, 159; inter- 157; as share ofGDP, 15, 16, 20, 21-22 nal relative prices of, 12, 13; price elasticity of Flowers, 9, 28, 34, 62, 69, 77 demand of, 54; prices of, 67, 68-69, 71, 152, Fondo Financiero Agropecuario (FFAP), 114, 127, 160; of raw materials, 63; as share of GDP, 9, 19, 133, 145, 146, 161 29, 45; smuggled, 29, 45; subsidies for, 84; taxes Food: demand for, 22; exports of, 10; imports of, on, 52, 53, 84, 158; values of, 7; volume of, 54. 46, 51; prices of, 34, 80; production of, 5, 10; See also Licenses; Tariffs self-sufficiency in, 67. See also specificfoods Import substitution, 7, 9, 27, 45, 47, 53 Foreign demand. See Agricultural commodities Incentives. See Agricultural commodities (incen- (foreign demand and); Exports (foreign demand tives and); Exports (incentives for); Imports (in- and) centives for) Foreign exchange, 5, 11, 22, 28, 44, 47, 51, 97, Income: disposable, 13, 15; real, 35, 149. See also 157. See also Agricultural commodities (foreign Agricultural income; Wages exchange and) Income distribution, 53, 56, 98, 144, 160 Foreign investment, 39, 44 INcOMEX. See Foreign Trade Institute Foreign Trade Institute (NcomFx), 38, 51, 56, 1NcORA. See Instituto Colombiano de la Re- 62, 71, 110 forma Agraria Forestry, 133, 142, 143, 152, 153; products, 35, INDERENA. See Instituto Nacional de los Re- 47, 142 cursos Naturales Free trade zones, 39 Industry, 3, 19, 21, 44; exports by, 7; import- Futures market, 85-87, 97, 160 competing, 33, 45, 48, 53; inputs in, 45; value General Agreement on Tariffs and Trade (GAT), added by, 28 48,84 Inflation, 4, 13, 19, 20, 21, 22, 25, 29, 30, 37, 40, 61, 77, 94, 95, 96, 97, 99, 144, 153, 157, Government inepeniues See Public ser 8158; cost-push, 37; price supports and, 63, 66. Government intervention, 14, 61, 62-66, 82, 83 See also Macroeconomic policy (inflation and) 97-98, 123, 136, 151, 152, 153 Ifaintx 19 Gross domestic product (GDP), 3, 7, 9, 10, 15, Infrastructure, agricultural, 14, 101, 106. 123, 16, 71-72, 77, 101, 146; growth in, 5, 6, 19 124,r132, 133, 1 4, 12,6150, 16 Growers' associations. See Producers' associations 124,132,133,134,141-42,150,161 Growth, economic, 20, 28, 45, 157-58. See also Input prices, 4, 136, 137-40, 145, 151, 160 Gross domestic product (growth in) Input supply, 14, 39, 61, 159 Instituto Colombiano Agropecuario (ICA), 124, Hedging, 85-86, 97-98 125, 126, 127, 131, 133, 140, 152, 161 HIMAT. See Instituto de Hidrologia, Meteoro- Instituto Colombiano de la Reforma Agraria logia, y Adecuaci6nes de Tierras (iNcoRA), 125, 126, 127, 133, 161 INDEX 249 Instituto de Comercio Exterior. See Foreign Liberalization of trade, 19, 21, 29, 43. 44, 48, 50, Trade Institute 51, 55, 76, 77, 158, 160 Instituto de Hidrologia, Meteorologia, y Ade- Licenses: auctioning of, 55, 56, for importing and cuaci6nes de Tierras (HIMAT), 126, 133, 161 exporting, 19, 38, 43, 44, 46, 48, 50, 51-52, Instituto del Transporte, 150 53, 54-55, 56, 62, 63, 110, 151; price supports Instituto de Mercadeo Agropecuario (IDEMA), and, 62, 71 46, 62, 63-66, 69, 70-71, 76, 77, 80, 81, 85, Liquidity, 29, 37, 128 87, 88, 98-99, 124, 134, 150, 151, 152, 160; Livestock, 34, 133, 137; CERTand, 62 storage and, 88, 93-97, 161 Livestock products, 9, 63; exports of, 34, 72; im- Instituto Nacional de los Recursos Naturales ports of, 46; protection of, 67, 78 (INDERENA), 126, 127, 133, 142, 143, 152, 161 M p Interatd RralDevlopmnt lan(DR), 27, Macroeconomic policy, 61, 77, 156, 1 58-59; ad- 149, 151, 153 justment and, 15, 20, 21, 157, 158, agricultural investment and, 134; and economic growth, Integrated Rural Development Program (m1p), 157-58; inflation and, 15-18; management of, 143 .4, 20-21; price instability and, 81, 83, trends in, Inter-American Institute for Agricultural Sciences 5, 21 (llCA), 146 52 (11CA, 146Manufacturing: decline of, 7; exports of, 28, 38, Interest rate, 17, 94, 95, 133, 146, 147, 161; de- Manufactrg dine , 7; ep.oto 28, 38, ~55~5 a sar o op ad,9115; xpcaton fr,4 lotig 44; growth rate in, 6, 7, 28, production in, 44; posits and, 91, I54; expectations for, 4; floating, a hr fGP 161; nominal, 33, 36-37, 40, 146, real. 44, 88;. subsidies for, 8 8, 90, 92, 93, 161 Marketing: of agricultural products, 14, 35, 98, sntersedieood8, 4,4 9124, 132, 136, 149-52, 153, 159, 161; of cof- Intermediate goods, 45, 48 fee, 106 Internal prices, 49, 73, 107, 118 fee. , 10 Market price, 66, 93, 94, 100nl17, 152 International Coffee Agreement (icA), 11, Microeconomic production, 4-5, 83 107, 110, 118, 160 Middleman margins and prices, 82 International Coffee Organization (ico), 62, Milk, 63, 67, 88 105, 108, 109, 113, 116, 117 1train econom9 , 35,6,4 7 Ministry of Agriculture (MoA), 62, 63, 71, 125, International economy, 35, 36, 41, 72, 77 .126, 127, 132, 146, 151. See also Oficina de International prices: for agricultural commodities, Planeacion del Sector Agropecuario 47, 62, 63, 66, 71, 72, 77, 82, 140; for coffee, Ministry of Finance, 128 74; for Colombian exports, 5, 11, 12, 21, 41; Monetary Board, 52, 88, 89, 92 and futures markets, 97. See also Agricultural Monetary policy, 17, 21, 37-38, 92, 158 commodities (export prices of) Money supply, 29, 37, 40, 91-92, 98, 157 International reserves, 15, 16, 19, 20, 21, 43, 48, 55, 56 National Apprenticeship Service (SENA), 127 Investment, 16, 29, 38; return on, in land, 82. National Budget Statute (1973), 128 See also Agricultural investment; Foreign invest- National Coffee Fund (NCF, 1940), 106-07, ment 109, 112-13, 118; loans by, 115 mDP. See Integrated Rural Development Pro- National Council of Economic and Social Policy gram (CONPES), 38 Irrigation and land development, 14, 123, 127, National Food and Nutrition Plan (PAN), 127, 134, 141, 142, 152, 159, 161 149,151,153 National Forestry Research Plan (PLANIF), 47, Labor, 136; costs of, 78, 137, 153; demand for, 143 45, 143, 144; deployment of, 10; migrant, 103; National Institute for Renewable Natural Re- productivity of, 144; shortages of, 143 sources and the Environment. See Instituto Na- Latin America, 5, 10, 11, 35, 157 cional de los Recursos Naturales Latin American Association for Development and Natural resources, 39, 127, 142-43, 152-53 Integration (ALADI), 48-49 NER. See Exchange rate (nominal) Ley 5' program (for agricultural credit), 114, 115, Net capital inflow, 19, 20, 22, 29, 158 127, 133 Nontradables, 21, 40 250 INDEX Oficina de Planeaci6n del Sector Agropecuario Price supports, 4, 14, 46, 47, 61, 62, 63-66, (opsA), 80, 150 70, 71, 76, 77, 81, 85, 150, 151, 156, 159, 160 Output. See Gross domestic product; specific crops Prior deposit requirements for imports, 44, 52-5 3 (production of) Private sector, 14, 29, 39, 50-51, 116, 151, 157; agricultural investment and, 123, 130-3 3, Palm oil, 72, 88, 133 134, 142, 161; agricultural research and, 140; PAN. See Plan Nacional de Alimentacion y Nu- storage and, 95, 97, 98 trici6n PROCA&A. See Productores de Caiia Pasilla (grade of coffee), 110, 111 Processing: of agricultural products, 131, 132, Pergamino(processed coffee), 110, 111, 113, 116 149, 150, 153, 161; and exports, 35, 39, 69 Peso (Colombian), 10, 159; appreciation of, 11, PRODESARROLLO. See Programa de Diversifica- 18, 21, 26, 30, 36, 43, 53, 67; depreciation of, ci6n y Desarrollo de Zonas Cafeteras 13, 17, 21, 26, 29, 32, 33, 37, 40, 55, 107; Producer prices, 6, 11, 14, 47, 62, 67, 80, 98, possible devaluation of, 36, 37, 38, 40; real 152, 153, 159-60; relative, 21, 22 value of, 72 Producers' associations, 91, 93, 106, 123, 127, Peso-dollar rate, 30, 31, 32, 35, 39 131-32, 150 Pesticides, price controls on, 62, 151 Productores de Cafia (PROCArA), 62 Petroleum, 20, 26, 39, 44, 46, 50, 158 PROEXPO. See Export Promotion Fund Plan Nacional de Alimentaci6n y Nutrici6n Profitability in agriculture, 61, 82, 96, 140 (PAN), 127, 149, 151, 153 Programa de Desarrollo Rural Integrado (DRI), Plan Nacional de Investigacion Forestal 127, 149, 151, 153 (PLANiF), 47, 143 Programa de Diversificaci6n y Desarrollo de Zonas Plan Vallejo (import-export regime), 27, 28, 39, Cafeteras(PRODESARROLLo), 114, 149, 153 41 Protectionism, 35, 41, 44, 48, 50, 53, 56, 61, 62, Policies, government. See Agricultural trade pol- 67-71, 77, 78, 159, 160 icies; Development plan; Import policy; Macro- Public Investment Fund, 128, 161 economic policy; Monetary policy; Stabiliza- Public sector: accounts, 16, 20, 22; agricultural tion policy; Trade reform investment and, 123-30, 133, 134, 136, 154, Port-handling costs, 47, 67, 68, 69, 71, 145, 159, 161; expenditures, 16, 29, 40, 127, 128, 153, 154, 160 158; imports and, 50-51 Ports, 68, 153 Pulp and paper, imports of, 47 Potatoes, 82, 88, 131, 137 Purchasing power parity (Ppp), 25-26 PPP. See Purchasing power parity Price controls, 62 Quantitative restrictions. See Quota system Price distortions, 78 Quota system: for exports, 77, 83; for imports, Price fixing, 62 54, 56, 62, 63, 76. See also Coffee (export quotas Price interventions, 61, 63, 66-71, 76-78, 107, on) 110, 145, 160 Prices: basic, 92-93, 94, 97; border, 47; clearing, Recession, international, 10, 22. See also Colom- 56; constant, 19, 30; factor, 36; import parity, bia, recession in 71; input, 4, 136, 137-40, 145, 151, 160; in- Reforestation, 47, 142, 143 ternal, 49, 73, 107, 118; market, 66, 93, 94, Refrigeration of agricultural products, 150, 151 100n17, 132; output, 4, 62, 137, 140; relative, Regional corporations for development, 126-27 12-13, 15, 18, 53; retail, 69; wholesale, 69; Reintegro minimo (minimum surrender price for World Bank projections of commodity, 34. See coffee), 111 also Consumer prices; Domestic prices, Farmgate Relative prices, 12-13, 15, 18, 53. See also Agri- price; International prices; Producer prices; prc cultural commodities (relative prices of) entries under specif co wodities RER. See Exchange rate (real) Price stabilization, 4, 11, 77, 80, 81-82, 97-99, Research and extension: agricultural, 14, 61, 123, 156; annual and seasonal aspects of, 80-81, 127, 131, 134, 140, 142, 144, 152, 159, 161; 83-97 in forestry, 143; stations, 124 INDEX 251 Resolution: 2/82, 52; 16/82, 52; 39/82, 48; Storage, agricultural, 80, 87-97, 98-99, 150, 99/83, 53; 015/83, 48; 030/83, 48 161. See also Coffee (storage of) Resource allocation, 28, 43, 77 Stumpingofcoffee trees, 113-14, 117, 118, 160 Retail margins, 149, 153 Subsidies: on agricultural loans, 114-15; con- Retention quota for coffee, I10, 111 sumer price, 97; for fertilizer, 118; production, Rice, 34, 62, 72, 78, 142, 150; CERT and, 62; 69, 160. See also Credit (subsidies for); Exports export subsidies for, 62, 68; prices of, 46, 67, (subsidies for); Imports (subsidies for) 80, 82; price supports for, 62, 63, 70, 76, 160; Sugar, 34, 72; price controls on, 62; prices of, processing of, 69; production of, 137, produc- 62, 67, 68, 82; processing of, 69; producer tion costs of, 72; stocks of, 151; storage of, 93, prices of, 80; production of, 137; research on, 95; subsidies for, 88; yields of, 141 140 Riio (grade of coffee), 110, 111 Sugar Exporters' Association, 62 Roya (coffee leaf rust), 109, 118, 160 Sugar Growers' Association. See Productores de Rural-urban migration, 131, 143 Caiia Sugar Mills Association. See Asociaci6n de Culti- vadores de Caifa de Azucar de Colombia sAc. See Sociedad de Agricultores de Colombia Supply: aggregate, 18, 20; input, 14, 39, 61, SEA. See Special Exchange Account 159; money, 29, 37, 40, 91-92, 98, 157 Seasonality: of food prices, 80-81, 87-88, 100; of money supply, 92 Secretarias de Agricultura, 127 Tariffs 1, 13, 3, 43, 44, 47, 48, 54, 55, 56, 62, Seeds, 131, 136, 151; certified, 140 68, 71, 145, 150, 154, 160; nominal, 46, 48- Self-sufficiency: in food, 67; in petroleum, 26, 39 T0, 53, 6 Servicio Nacional de Aedizj (sENA), 127 Tax: import, 52, 53; inflation, 91, 92; stamp, 53 prenizaje Tax reimbursement certificate. See Certificado de Sesame, 88; price supports for, 62, 63, 66, 71, Reembolso Tributari 160; producer prices of, 71; stocks, 151 Tax revenues, 17, 29, 111. Se also Exports (taxes sF. See Situado Fiscal Tax revenus 17, 2 e o Shade trees, in coffee cultivation, 103, 114,1I17 on); Imports (taxes on) Shsal, 62, , 8 e cTechnology, 7, 127, 136, 137, 140, 151; trans- Sisal, 62, 80, 88 fer of, 61. See also Coffee production technology Situado Fiscal (sy), 128 Terms of trade, 11, 12, 21, 26, 73-76 Smuggling, 1 18-19n5. See also Imports (smug- gled) Textiles: exports of, 28, 62; smuggling of, 45 Social costs and benefits. See Coffee (diversifica- Tipica (variety of coffee tree), 102 ialon s Tobacco, 9, 34, 51, 62, 67, 69, 72, 77, 88, 91 tion in) Taedfct,1,4 Sociedad de Agricultores de Colombia (SAC), Trade deficits, 19, 47 132 Trade reform (1967), 26, 27-29 132 Trade surplus, 10 Soil erosion, 142, 152 Sodrhum, 8;d etcf 6; Transport costs, 47, 53, 67, 68, 69, 77, 113, 145, Sorghum, 88; domestic consumption of, 63; im- 150, 153, 160 ports of, 46, 47; prices of, 46, 67, 80; price sup- Transport sector, 9,107, 127 ports for, 62, 63, 160; producer prices of, 71; production of, 68; production costs of, 72; re- search on, 140; stocks of, 151; storage of, 93; Unemployment rate, 19, 157 yields of, 141 Union of Agrarian Reform Cooperatives, Ltd. Soybeans, 88, 140; domestic consumption of, 63; (cECORA), 151 imports of, 47, 137; prices of, 67, 80; price sup- Upper Magdalena Pilot Watershed Management ports for, 62, 160; producer prices of, 71; pro- Project, 142, 143, 153 duction of, 68, 72; production costs of, 72; U.S. dollar, 32; appreciation of, 30, 31, 39; and stocks of, 151; yields of, 141 Colombian peso, 30, 31; depreciation of, 40 Special Exchange Account (SEA), 29, 128, 161 U.S. import duties, 34 Stabilization policy, 13, 83, 157, package (1977), U.S. market for Colombian exports, 40 29. See also Price stabilization U.S. milk prices, 67 252 INDEX Value added, 9, 46, 50, 144 prices of, 46, 67, 80, 82; price supports for, 62, Vegetables, imports of, 46 63, 66, 70, 71, 160; production of, 28, 68, 72, Venezuela, 10, 11; and Colombia, 11, 44, 157 77; production costs of, 72; stocks of, 151; stor- age of, 93 Wholesale margin, 149, 150, 153 Wages: expectations of, 4; minimum, 144, 153; Wholesale markets, 150, 151, 153 nominal, 40; real, 5, 16, 18, 36, 40, 47, 143, Wholesale prices, 69 144. See also Agricultural income World Bank, 43, 128, 137, 161; price projections Welfare effect of agricultural policies, 70, 84 by, 34; projects of, 141 Wheat, 88, 131, 140; domestic consumption of, 46, 63; ERP of, 70; imports of, 46, 47, 69, 137; Yuca, 137 The most recent World Bank publications are described in the annual spring and fall lists- The latest edition is available free of charge from Publications Sales Unit, Department B, The World Bank, Washington, D.C. 20433, U.S.A. Vinod Thomas is a senior economist in the Country Programs Department of the Latin America and the Caribbean Regional Office of the World Bank. 0 8018 3121 0