Short-term Responses to Trade and Incentive Policies in the Ivory Coast Comparative Static Simulations in a Computable General Equilibrium Model Gilles Michel Michel Noel WORLD BANK STAFF WORKING PAPERS Numnber 647 f . .. - A r @ q , .J I ;'V1' fei0, Vfi -'- ' l~Sdf l1i S . w - S . | .1 WORLD BANK STAFF WORKING PAPERS Number 647 Short-term Responses to Trade and Incentive Policies in the Ivory Coast Comparative Static Simulations in a Computable General Equilibrium Model Gilles Michel Michel Noel The World Bank Washington, D.C., U.S.A. Cc-pyright © 1984 The International Bank for Reconstruction and Development S THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. First printing June 1984 All rights reserved Manufactured in the United States of America This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The views and interpretations in this document are those of the author(s) and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting on their behalf. Any maps used have been prepared solely for the convenience of the readers; the denominations used and the boundaries shown do not imply, on the part of the World Bank and its affiliates, any judgment on the legal status of any territory or any endorsement or acceptance of such boundaries. The full range of World Bank publications is described in the Catalog of World Bank Publications; the continuing research program of the Bank is outlined in World Bank Research Program: Abstracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Publications Sales Unit of the Bank in Washington or from the European Office of the Bank, 66, avenue d'Iena, 75116 Paris, France. Gilles Michel is an industrial economist in the World Bank's Industry Department, and Michel Noel is an economist in the West Africa Programs Department II. Library of Congress Cataloging in Publication Data Michel, Gilles. Short-term responses to trade and incentive policies in the Ivory coast. (World Bank staff working papers ; no. 647) Bibliography: p. 1. Ivory Coast--Economic conditions--1960- -- Mathematical models. 2. Ivory Coast--Economic policy-- Mathematical models. 3. Equilibrium (Economics) 4. Statics and dynamics (Social sciences) I. Noel, Michel. II. Title. III. Series. HC1025.M53 1984 330.9666'8'00724 84-7519 ISBN 0-8213-0371-6 ABSTRACT The Ivory Coast experienced a series of external shocks in the last seven years which led to a severe deterioration of its financial position and to negative growth since 1981. This occurred at a time when the growth potential of the country had declined and the ability of the economy to adjust was limited by distortions in the system of macroeconomic and sectoral incentives. To remedy this situation, the Government has undertaken a program of stringent financial recovery and of structural adjustment, articulated around substantial policy reforms. This paper analyzes the short-term effects of various trade and incentive policies that can be envisaged within this program. The analysis is carried out through a series of comparative static experiments done in the framework of a computable general equilibrium model. Simulations are performed to assess the impact of exchange rate adjustments, tariff reform, the introduction of an export subsidy and, finally, removal of quantitative import restrictions. Extracto La Costa de Marfil experiment6 una serie de conmociones externas en los ultimos siete afios que produjeron un grave deterioro de su situaci6n financiera y un crecimiento negativo desde 1981. Esto ocurri6 en un momento en que el potencial de crecimiento del pais habia declinado y la capacidad de ajuste de la economia estaba limitada por las distorsiones del sistema de incentivos macroecon6micos y sectoriales. A fin de remediar esta situaci6n, el Gobierno ha encarado un programa riguroso de recuperaci6n financiera y ajuste estructural, articulado en torno de considerables reformas de la politica. En este estudio se analizan los efectos a corto plazo de varias politicas comerciales y de incentivos que pueden considerarse dentro de dicho programa. El analisis se realiza a trav6s de una serie de experimentos estaticos comparativos realizados en el marco de un modelo de equilibrio general computable. Se realizan simulaciones para evaluar el efecto de los ajustes cambiarios, la reforma arancelaria, la implantaci6n de un subsidio a la exportaci6n y, por uiltimo, la eliminaci6n de restricciones cuantitativas para las importaciones. La C6te d'Ivoire a subi, au cours des sept dernieres annees, plusieurs chocs d'origine exterieure qui ont amene une deterioration grave de sa position financiere et une croissance negative depuis 1981, au moment meme oa le potentiel de croissance du pays se trouvait r6duit et ou l'aptitude de 1'6conomie a s'ajuster 6tait freinee 1ar des distorsions du syst&rPf d'incitations macro6conomiques et sectorielles. Pour rem6dier a cette situation, le Gouvernement a lance un programme de redressement financier et d'ajustement structurel tr&s strict, centre sur d'importantes reformes des politiques d'action. Le present document analyse les effets a court terme des diverses mesures d'incitations et d'actions en matiere de commerce exte- rieur susceptibles d'etre envisagees au sein dudit programme. L'analyse comprend une s6rie d'experiences statiques comparatives r6alis6es dans le cadre d'un modele d'6quilibre gen6ral quantifiable. Des exercices de simulation ont 6t6 effectues en vue d'evaluer les effets qu'auraient des ajustements des taux de change, des r6formes des tarifs douaniers, l'in- troduction de subventions a l'exportation et, enfin, l'elimination de contingents d'importatioIIs. ACKNOWLEDGEMENTS We are indebted to Sherman Robinson for his continuous support and advice throughout this study. We also bene- fitted from many helpful discussions with Kemal Dervis. We thank Timothy Condon and Benoit Morin for valuable research assistance. Table of Contents Page Introduction 1 I. Background 3 Chapter 1: Overview of the Ivorian Economy 3 1. Structure of the Economy 3 2. Recent Economic Developments and Policy Issues: 1976-1982 13 Chapter 2: The Model 18 1. General Structure of CGE Models 18 2. The Treatment of Foreign Trade in CGE Models 20 3. The Ivory Coast Model 23 II. Policy Experiments 35 Chapter 3: Exchange Rate Adjustments 35 1. The Exchange Rate Issue 35 2. Economic Mechanisms Involved 38 3. Analysis of the Experiments 41 Chapter 4: Tariffs and Export Subsidies Reforms in the Industrial Sector 52 1. Policy Issues 52 2. Economic Mechanisms 54 3. Analysis of the Experiments 56 Chapter 5: Quantitative Import Restrictions 65 1. Background 65 2. Economic Mechanisms 67 3. Analysis of the Experiments 68 Conclusion 72 Annex 1: Equations of the Model 75 Annex 2: The Ivory Coast SAM 85 Annex 3: Sectoral Tradeability Indicators 89 References 90 Introduction After two decades of continuous growth, the Ivory Coast economy experienced a series of external shocks over the last six years: the coffee and cocoa boom in 1976 and 1977; a 40 % decline in the terms of trade between 1978 and 1982; and sharp increases in real interest rates on international ,,. markets. These shocks have led to a rapid deterioration of the country's financial position and to two consecutive years of recession in 1981 and 1982. These dramatic changes have affected the Ivory Coast at a critical phase of its development. Since the mid-seventies, the growth potential of the country had progressively declined due primarily to low returns on public investment, inefficiencies in puJ'ic enterprises, and inadequate support for agricultural development. At the same time, the ability of the economy to adjust was severely hampered by rigidities and distortions in the system of macroeconomic and sectoral incentives, particularly with respect to the exchange rate, tariffs, quantitative restrictions, and export incentives. In order to remedy this situation, the government has embarked on both a stringent financial recovery program and a structural adjustment program that calls for substantial policy reforms, in particular in the system of incentives in industry and agriculture. The purpose of this paper is to analyze in a general equ:ilibrium framework the short-term effects of various trade and incentive policies that could be envisaged within the framework of these programs. This is done through a series of comparative statics simulations carried out with a computable general equilibrium (CGE) model of the Ivorian economy, which has been built along lines already followed in an increasing number of countries. In Chapter 1, we present an overview of the Ivorian economy and discuss the most salient economic developments over the last six years. -2- Chapter 2 is devoted to a general discussion of CGE models, with a special emphasis on the treatment of foreign trade, and to a detailed pre,-entation of the Ivory Coast model. Turning to policy simulations, an analysis of experiments on exchange rate adjustments is presented in Chapter 3. After stressing the importance and the scope of the exchange rate issue for the Ivory Coast today, we briefly discuss the economic mechanisms involved, and then turn to an in-depth discussion of the base experiment and the sensitivity analysis0 Chapter 4 contains an analysis of the experiments on tariff and subsidy reforms, as well as a detailed presentation of the base experiments and a series of alternative experiments. This policy issue is directly relateu to the reforms envisioned by the government in the structural adjustment program. In Chapter 5, we examine the impact of the removal of quantitative import restrictions on the economy when exchange rate adjustments are allowed. I. BACKGROUND CHAPTER 1: OVERVIEW OF THE IVORIAN ECONOMY This chapter is devoted to a broad description of the structure of the Ivorian economy, and to a discussion of the current economic situation and policy issues facing the country, which constitute the background and the rationale for the present study. 1. Structure of the Economy In the first two decades following independence, the Ivory Coast's GDP grew 7.5% annually, a rate of growth which ranks among the highest in Africa and among the top fifteen in the world. This performance has been based on a rapidly growing agricultural sector (primarily coffee and cocoa), which generated surpluses for in. stment, and on a liberal and pragmatic economic policy that made the productive use of foreign factors possible. Out of a total population of 8.3 million, 1.5 million are immigrants who responded to employment opportunities in the Ivory Coast and to the growing income differentials with its neighbors. With an annual population growth of 4.3% per year (including 1.5% due to migration), per capita income increased 3.5% annually until the end of the 1970s. Although a severe slowdown has since ocCurred, the Ivory Coast had a per capita GDP of US$1,200 in 1981, ranking first among Sub-Saharan developing countries-I/ This compares with US$870 for Nigeria, US$880 for Cameroon and US$430 for Senegal. The following tables provide some basic macroeconomic indicators on the structure of the economy. 1/ World Bank. World Development Report 1983. [1983]. -4- Table. 1.1 Structure of the Ivorian Economy in Three Sectors (in 1980) Gross Domestic Value Added (at Output Exports Imports Demand factor prices) Agriculture 23.3 49.4 5.1 19.1 33.1 (inc. integrated industries: sugar, rubber, palm oil) Industry 43.1 28.4 78.2 47.8 29.1 (inc. energy, utilities, arnd construction) Services 33.6 22.2 16.7 33.1 37.8 (inc. transportation) Total 100.0 100.0 100.0 100.0 100.0 Source: Ministry of Finance, Budgets Economiques. Table 1.2 Structure of Trade (in 1980) Share of Ratio of Share of imported exports in imports to intermediate inputs production domestic supply in total int. inputs Agriculture 3.5 39.0 22.6 Industry 30.6 15.4 33.0 Services 11.1 15.4 26.6 Source: Ministry of Finance, Budgets Economiquese Agriculture is a major component of the Ivorian economy and the basis of its development. It contributes one-third to GDP, provides around 50% of exports, and employs an estimated 75% of the labor force, of which 13% are immigrants.l/ The Ivory Coast is the world's third largest exporter of coffee, and the most important producer of Robusta. It is also the world's largest producer and exporter of cocoa. There are two main agricultural regions in the country. The Southern Forest and the Northern Savannah have different climates, natural endowments, culture and income levels. Traditional cash crops (coffee, cocoa, palm oil and, more recently, rubber) grow in the rich Southern Forest. The more arid and much poorer North is the area of development of cotton and sugar production. Food crops are grown throughout the country. Table 1.3 presents details of the structure of agricultural production in the Ivrory Coast. Food crops and cash crops contribute almost equally to nominal output and value added, while acreage appears to be larger for cash crops, especially coffee and cocoa. However, the acreage data should be regarded only as indicators, because many plantations are intercropped, and part of the area under food cultivation remains unaccounted for due to the absence of extension services for these crops, (except rice). 1/ See Ph. Fargues [1981]. - 6 - Table 1.3 Structure of Agricultural Production Food Cash Main Cash Crops crops crops Coffee Cocoa Cotton Rice Source - - - -(Percentages)-- ---------- Share in total cultivated acreage 1/ 36 64 30 21 1 8 (1) Share in the value of agricultural production 56 44 12 20 2 8 (2) Share in total value added 15.8 13.1 3.6 6 .6 n.a. (3) Share in agricultural value added2/ 54 46 12 20 20 2 (2) Share in total exports2/ -- 36.3 15.1 17.6 -- (3) Share in agricultural exports2/ -- 100 41.6 48.5 (3) Sources: (1) Recensement National de l'Agriculture (ENA) [19811 (2) Ministry of Finances, Budget Economiques (3) Social Accounting Matrix in G. Michel and M. Noel, [1983] 1/ The acreage is the theoretical one, defined in the RNA as the acreage needed to growing a solestand, as opposed to growing in association with other crops. 2/ Excluding forestry. Since the early 1960s, the government has emphasized the development of cash crops, through direct intervention in production consisting of the provision of extension services by public enterprises, or through market regulation.1/ Cash crops under regulated prices include coffee, cocoa, cotton 1/ Except when explicitly mentioned, the figures on public enterprises presented in this chapter are taken from the joint report of the Caisse Centrale de Cooperation Economique (CCCE) and the World BaTnk [1982). -7- and palm oil. For these products, official producer prices are fixed at the beginning of each season, and a Stabilization Fund makes up the difference between the domestic price and the price obtained in export markets. This mechanism provides producers with more stable incomes and has also generated substantial surpluses for the government. Most coffee and cocoa production comes from smallholders, who typically benefit from the extension services of SATMACI, the regional development agency for the Center Zone. Cotton is produced almost exclusively by small farmers under the auspices of CIDT, the regional development agency for the north, which provides extension services and controls the marketing of the crop. Seventy percent of palm oil originates from the industrial plantations of PALMINDUSTRIE, with village plantations accounting for the rest. Rubber culture is also dominated by public enterprises, and smallholders represent less than 8% of total production. Finally, the production of sugar is entirely controlled by SODESUCRE, which runs six sugar complexes in the north and the center of the country. In the mid-seventies, the government started to give more priority to the development of food crops. However food crops have not been subject to direct intervention by the public sector, in sharp contrast with the situation prevailing for most cash crops. The only exceptions is rice. Producer and consumer prices for rice are regulated through the Caisse de Perequation, and assistance is provided by extension services of regional development agencies. Public enterprises are not limited to the agricultural sector, but play a role in other key parts of the economy as well, with the important exception of industry. Although it favors a liberal type of economy, the Ivorian Government has always relied partly on public enterprises to implement its development policies, particularly in areas where private initiative -8- remained inadequate. Public enterprises are important because of their contribution to national production and investment. It has been estimated that one-third of agricultural production is directly dependent on public enterprises. They are also involved in land transportation, shipping, services, construction and electricity. Overall, 15-20% of GDP can be attributed to parastatal enterprises in the Ivory Coast.1/ The contribution of public enterprises to investment is even more important. According to the Budgets Economiques [1982], their share of total investment was 29% in 1980, 26% in 1981 and 22% in 1982, after having reached a peak of 37% in 1978. Table 1.4 gives the sectoral allocation of public enterprise investment in 1980. Table 1.4 Sectoral Allocation of Public Enterprise Investment in 1980 Agriculture 18.1 Mining-petroleum 29.3 Manufacturing industry 0.9 Utilities 17.7 Construction 5.7 Services 28.3 100.0 Source: Ministry of Finances, Budgets Economiques. 1/ Estimates of the share of public enterprises in GDP and investment are based on the Budgets Economiques definition of public enterprises. -9- A large part of these investments have been financed by foreign capital, which has been attracted by publicly guaranteed investment opportunities. As a result, public enterprises have been instrumental in the opening of the economy to foreign investors. Accordingly, they account today for 35-40% of the external public debt. Finally, since public enterprises investments are ultimately controlled at the government level, they represent an important policy variable for the government. With both public enterprise and central government investment, the government controlled as much as 70% GZ total investment in 1980. Industry has been one of the most dynamic sectors in the Ivorian economy.l/ Since independence, manufazturing activity has grown at the rate of 13% per year, and its share in GDP rose from 4% in 1960 to 13.8% in 1980. This rapid expansion has been primarily based on import substitution and on the processing of local agricultural raw materials for exports. Until the early 1970s, industrial development took place in the framework of a simple system of incentives, which were essentially based on a moderate tariff schedule that applied equal protection among industrial activities, and without quantitative import restrictions. The 1973 tariff reform and, more recently, the proliferation of quantitative import restrictions, have introduced increasing distortions in the industrial incentives structure. As a result, inefficient import substituting enterprises have been established and a strong bias against exporting activities has appeared.2/ 1/ See Appendix B: "The Industrial Sector" of B. den Tuinder [1978] for a more detailed description of the evolution of this sector since independence. 2/ See M. Noel [1982] for a comprehensive study of the system of industrial incentives. - 10 - This sector developed without the direct involvement of public enterprises. Instead, it relied exclusively on private initiative, which was mainly foreign, especially in the first stages of industrialization. Foreign firms were attracted by the stable and liberal framework provided by the government as well as by the incentives provided under the Investment Code. Table 1.5 Structure of the Manufacturing Industry Share in Share of Share of Share of Share of Share of Imports industrial total exports in total import in total over GDP (in %) GDP gross output exports intermediate imports domestic input supply Food process- ing 28.3 3.9 28.7 10.1 21.7 8.8 54.4 Textiles 25.2 3.5 20.2 4.0 29.2 4.1 40.6 Chemicals 9.3 1.3 14.5 1.4 47.1 8.8 62.8 Construction material 15.5 2.1 32.5 3.7 26.7 2.7 54.1 Engineering industries 21.7 3.0 11.8 2.8 51.7 35e5 269.7 Total 100.0 13.8 -- 22.0 -- 59.9 Source: G. Michel and M. Noel [1983], Table 1.5 summarizes some key indicators for the structure of the manufacturing sector. The sector consists of two distinct types of activities: those which are based on the processing of local agricultural or forestry raw materials (food processing, textiles, and construction materials); and those which import raw materials intensively (chemicals and plastics and engineering industries). The former were the leaders in the early industrialization process of the Ivory Coast and are relatively export- - 11 - oriented. The second type of industry is much less export-oriented, and relies heavily on imports for its operations (50% of intermediate inputs are imported). With the exception of fertilizers, these industries fall far short of satisfying domestic demand, as is shown by the ratio of imports over domestic supply. This is because local production is concentrated on specific product lines and imports in these areas are essentially complementary to domestic production. Moreover, engineering industries alone represent more than a third of the imports of the country, and 60% of its total imports are manufactured goods. Energy dependence is rapidly decreasing due to the development of hydroelectric production in the 1960s and 1970s and to the current expansion of the oil sector. Although the start of power production by several new dams in the late 1970s has not made the Ivory Coast energy self-sufficient, the consumption of fuel oil needed to generate electricity has declined by 57.7% since 1979 as a result. The prospect of the huge Soubre dam, expected to become operational in the late 1980s, makes self-sufficiency a reasonable prospect. Moreovpr, petroleum production is increasing, and one-third of Ivorian consumption of crude oil was produced domestically in 1981. Although the projections were to be lowered recently, the outlook is still for the Ivory Coast to become a net oil exporter by 1986. Finally, an oil refinery with a 4 million ton capacity already meets domestic requirements as well as exports to neighboring Upper Volta and Mali. The Monetary arrangements for the Ivory Coast play an important role in development. The country is a member of the West Africa Monetary Union (UMOA)1/ (with Benin, Niger, Senegal, Togo and Upper Volta), which centralizes foreign currency reserves, issues a single currency (the CFA Franc) through a 1/ UMOA: Union Monetaire Ouest Africaine. - 12 - common Central Bank, has a common interest rate structure and allows free transfers of funds within the Union. The CFAF is linked to the French Franc at a parity which has never changed since independence. It is fully convertible with the French Franc, and the French. Treasury is responsible for the necessary operations to maintain parity. Although this limits the member countries' autonomous monetary policy and prevents the adoption of exchange rate policy, convertibility with the FF and the backing of the French Treasury provide advantages that appear to have been beneficial for their development. The Ivory Coast is also a member of the West Africa Economic Community (CEAO), a customs union with Mali, Mauritania, Niger, Senegal and Upper Volta. As a partner in the Lome agreements, the Ivory Coast receives duty-free treatment for its exports to the EEC. These external ties are reflected in the trade structure of the Ivory Coast. Approximately half of the country's trade is with the EEC, while about 10% of exports are to CEAO countries. Thus almost two-thirds of exports are to countries with which the Ivory Coast has special relationships and arrangements. Table 1.6 Trade Structure in 1981 Share of total exports Share of total imports EEC 54.7 48.4 of which France (18.6) (31.1) Other European Countries 5.1 4.6 Sub-saharan Africa 12.3 7.1 (of which CEAO) (9.9) (2.3) USA 11.5 5.4 Asia 2.9 7.7 Others 13.5 26.8 100.0 100.0 Source: Customs statistics - 13 - 2. Recent Economic Developments and Policy Issues: 1976-1982 Over the past five years, the Ivory Coast economy has been subject to a series of external shocks: the coffee and cocoa boom in 1976 and 1977, a 40% decline in the terms of trade following the sharp drop in coffee and cocoa prices in 1978, the increase in the price of imported oil, and, more recently, sharp increases in real interest rates on international financial markets. These shocks have led to rapid financial deterioration and to a severe slowdown in economic activity. The Ivory coast experienced two consecutive years of recession in 1981 and 1982, in sharp contrast to the continuous record of economic growth since the early sixties. During the coffee and cocoa boom, when prices for these two crops were multiplied by 3.6 and 2.2 respectively between 1975 and 1977, large surpluses for the Stabilization Fund were generated and the balance of payments situation substantially improved. Under these favorable circumstances, the Government embarked on an ambitious investment program. The share of public investment in GDP increased from 15 to over 25% between 1976 and 1978. This program was also financed through external borrowing, which was possible because balance of payments pressures had declined. The share of investment financed by national savings thus fell from 50% to less than 20% between 1976 and 1978, while the gross inflow of public capital increased from CFAF 56 billion to about CFAF 200 billion between 1975 and 1977. GDP grew by almost 10% in real terms in 1978. However, between 1977 and 1978 coffee and cocoa prices declined by 31% and 10%, respectively, and remained constant at these low levels until the first half of 1980, while the public investment program was maintained at a level corresponding to 20% of GDP. This created several concerns. On the one hand, the Ivory Coast's terms of trade declined by 15% between 1977 and 1978, mainly due to the decline of export prices and by an additional 17% between - 14 - 1978 and 1980, due to sharp increases in import prices. On the other hand, the goverment increasingly relied on foreign borrowing to finance its investment program. As a result, a public sector deficit of around 10% of GDP appeared in 1979, rising to 12.7% in 1980, and the balance of payments turned negative in 1979. The abrupt decline in coffee and cocoa prices that started in the second half of 1980 further aggravated the situation. In 1980, the current account deficit represented 16% of GDP, and the net foreign assets of the Central Bank were exhausted despite continued heavy borrowing abroad, while correspondingly, a CFAF 118 billion overdraft was registered in UMOA's operations account with the French Treasury. In 1981, the government initiated a drastic financial stabilization program within the framework of an Extended Facility agreement with the IMF. The program called for a reduction in the public sector deficit to 8.6% of GDP in 1982 and 6.3% in 1983; a decline in the current account deficit to 11.8% of GDP in 1982 and 8.3% in 1983; and the restoration of overall balance of payments equilibrium by 1983. Although the fiscal targets of the stabilization program were met by the end of 1982, a series of unfavorable international economic developments resulted in a significantly larger current account deficit than was originally targeted. Indeed, the Ivory Coast's terms of trade declined again by 16% between 1981 and 1982. Simultaneouisly, the country became increasingly dependent on nonconcessional loans to finance the public investment program and the public sector deficit, at rapidly increasing interest rates. Finally, oil production did not expand as expected, reaching a level of 1.3 million tons in 1982 against the expected 3.5 million tons. At the same time, a surge in gross foreign borrowings by the public sector from CFAF 277 billion to CFAF 407 billion between 1981 and 1982 contributed to a reduction in the overall - 15 - balance of payments deficit within the program target. However, it presaged rising debt service ratios in future years, and more deficits of the overall balance are projected at least until 1987. The following table summarizes the main macroeconomic indicators for the period 1978-1982. Table 1.7 Macro Economic Indicators 1978 1979 1980 1981/1 1982/1 Growth rate of GDP at constant prices 9X9 5.2 6.3 -1.3 -1.8 Revenue of Stabilization Fund (% of GDP) 10.1 7.8 3.8 1.3 2.4 Public investment (% of GDP) 25.5 22.8 20.8 15.8 12.7 Gross public savings (% of GDP) 13.0 9.3 6.5 6.1 2.0 Public sector deficit (% of GDP) 8.5 10.1 12.8 9.3 8.9 Exports f.o.b. (% of GDP) 33.1 29.0 29.4 32.4 32.3 Imports f.o.b. (% of GDP) 25.9 23.8 24.6 24.5 23.0 Source: World Bank Data Base. 1/ Preliminary figure. This sharp deterioration in the internal and external financial situation has affected the Ivory Coast at a critical phase in its development. Two decades of growth were essentially based on the exploitation of the most obvious areas of comparative advantage, particularly in agriculture, and on the development of a dynamic, although still limited, industrial base. The growth potential of the country progressively deteriorated during the later part of the seventies primarily because of the low economic returns on the expanded public investment program launched in 1977; the rigidity of the system of macroeconomic and sectoral incentives in the face of rapid changes in the country's external environment and inadequate - 16 - support for agricultural development. The rapid expansion of public investment toward the end of the seventies was accompanied by a sharp deterioration in the quality of investment projects. Unlike the 1971-75 plan, the increase in public investment for the 1976-80 program was mainly allocated to large projects with high unit costs. In agriculture, for example, most of the increased allocation was for six sugar complexes which had unit operating costs two to three times above world market prices. Overall, it has been estimated that the incremental national income from the US $8 billion spent on public investme-nt from 1976-80 was approximately 40% less than it would have been if earlier criteria had been maintained. This figure translates into an annual cost to the economy in terms of foregone earnings of around 5% of GDP in the early 1980se./ The rigidity of the system of macroeconomic and sectoral incentives has imposed more constraints on growth and on the country's capacity to respond flexibly to rapidly changing external conditions. The sharp acceleration of domestic inflation after 1975 and the sharp deterioration in the terms of trade between 1978 and 1982 led to a large overvaluation of the CFA Franc vis-a-vis the currencies of the Ivory Coast's main trading partners. This situation revealed fundamental deficiencies in the incentive system. Although major agricultural exports were not immediately hurt by the overvaluation because of their considerable comparative advantage, this was not the case in the industrial sector, where comparative advantage was not as strong. Expansion of the industrial sector has been directly affected by the strong bias against exports in the system of industrial incentives. At the 1/ See World Bank [1979 . - 17 - same time, the most obvious opportunities for exports within the protected CEAO markets have been exploited. As a result, the Ivory Coast will have to turn to non-regional markets to sustain the growth of its industrial base and" to generate the foreign exchange earnings required to balance its external accounts. Finally, the growth potential of the country has suffered from inadequate support of agricultural development. Although agriculture has been a key to the country's successful economic performance, agricultural development projects have remained a relatively sriall part of the public investment program, apart from the large expenditures for the six sugar complexes. More recently, however, the Government has reinforced planning capabilities in the agricultural sector, with the objective of designing a coherent development strategy based on an adequate system of incentives reflecting the medium to long-term comparative advantage of the country in particular crops. - 18 - CHAPTER 2: THE MODEL Computable general equilibrium (CGE) models have generated a large literature since the pioneering works of Johansen [1960] and, more recently, of Adeilnan and Robinson [1978]. The objective of this chapter is to provide an overview of this class of models and to focus subsequently on the characteristics of the Ivorian model.1 1. General Structure of CGE Models Computable general equilibrium models simulate the functioning of an economy by explicitly capturing the behavior of the various agents (households, firms, government, rest of the world), the institutional framework (fiscal system and transfer mechanisms), and the market clearing processes (price and quantity). As a result, they can be considered "structural" as opposed to "reduced form" models., CGE models are in the tradition of economy-wide multisector models that have been used for development planning over the last two decades. In contrast with traditional input-output and programming models that are best- suited for planning in centralized economies, CGE models have been developed for policy simulation in mixed economies, in which relative prices vary in response to supply and demand decisions by individual agents, and where the government can affect the outcome of these decisions, either indirectly through fiscal and incentive policies, or directly through intervention in some sectors of the economy. This class of models has recently been used for planning and policy analysis as well as for the evaluation of the performance 1/ For a comprehensive analysis of CGEs, see Dervis, de Melo and Robinson [1982], hereinafter referred to as DMR. Since this textbook pro-vides the description of a general type of CGE on which the model presented in the paper is based, it should also be considered general background for this study. - 19 - and prospects of an increasing number of developing countries. - The structure of CGE models is Walrasian in spirit, stressing interaction among economic agents and the workings of market clearing processes. At the same time, these models can formally incorporate rigidities and distortions that are particularly strong in developing economies. The treatment of foreign trade, for example, can substantially differ from a standard neoclassical story through the introduction of various foreign exchange rationing schemes (see Lewis and Urata [1983] or Bhattachyara, Grais and Pleskovik [1983]) or through quantitative import restrictions (see below). A CGE model is designed to provide a description of the evolution of an economy over a number of periods, given the value of exogenous parameters and policy instruments. Its structure can be separated into a static general equilibrium model, which solves within each period, and a dynamic between periods model, which links two successive periods. Within each period, the model provides a set of equilibrium prices which leads to a balance of supply and demand on each market, given the behavioral rules of the agents, exogenous parameters and the way marks s operate. Supply and demand for factors, commodities and foreign exchange are derived from each agent's optimal choice and from government decisions. Basically, it is assumed that households maximize their utility, given their budget constraint, while firms maximize their profits, given their technical constraints. Once supply and demand on each market have been determined, a price system has to be found to clear those markets. At this stage, the definition of the way each market operates is crucial, as it is often necessary to impose particular restrictions on markets that represent strong 1/ See, for example, Lewis and Urata [1983], Dervis and Robinson [1982], Armarand and Grais [1983], or Taylor et al. [1980]. - 20 - departures from the pure neoclassical paradigm (such as fixed nominal prices or fixed supplies). After the equilibrium solution is found for a given period, the between-period part of the model updates all the exogenous variables entering thie static part of the model. This operation is done using: (1) the equilibrium values of the foregone period, such as the amount of investment by sector that is added to existing capital to yield new capital stocks; (2) government policy decisions, such as a new level of tariffs or public investment; (3) simple exogenously specified trends, such as the growth of the population; and (4) pure exogenous variables, such as world prices. The model can then find a solution for a new period on the basis on the new set of exogenous data. CGE models are therefore basically simulation models, designed to investigate the impact of policies that work through the market system, such as taxes, tariffs or subsidy changes, or to analyze the impact of direct government intervention in the economy, such as the profile and composition of the public investment program, pricing policies or import restrictions.l/ 2. The Treatment of Foreign Trade in CGE Models The treatment of foreign trade has been given particular attention in CGE models. Two different approaches have been proposed in the literature for the treatment of foreign trade. On one hand, classical trade theory is built on the small country assumption, so that each country is a price taker on international markets. This assumption makes a strong distinction between traded goods, which are considered perfectly substitutable for imports and 1/ For a critical view on CGE's see Bell and Srinivasan [1983]. - 21 - whose prices are fixed on the international market, and non-traded goods, whose prices are entirely determined on the domestic market. Along with constant returns to scale in production, this theory leads to extreme specialization among countries, and it rules out two-way trade. On the other hand, the structuralist school assumes that imports are non-competitive, so that the degree of substitutability between domestic goods and imports is zero (this is, for instance, the assumption in the two-gap models). Imports are treated as perfect complements for domestic goods. This approach results in a rigid framework in which trade policy has no role to play in closing the foreign exchange gap. The assumption made in this class of CGE models is that sectoral imports are neither complete substitutes nor full complements of domestic production. In other words, two-way trade at a sectoral level as well as price differentiation between domestic and import prices is allowed. Both features are ruled out by traditional models. The original formalization of these ideas was made by Armington [1969], who introduced the notion of a composite good, that is, an aggregate of imported and domestic goods. In this approach, agents do not demand either domestic or imported goods, but an aggregate of the two. The composition of the aggregate good depends on the relative price of domestic and imported goods, which may differ. This specification could hardly be sustained at the ideal most disaggregated level where all goods would truly be perfect substitutes with imports, but is is of considerable relevance at the high degree of aggregation usually taken in CGE models, and supported by many empirical evidence (see Armington [1969]). Instead of the strong distinction between traded and non- traded goods this approach implies that sectors are characterized by different degrees of tradability depending on their trade substitution elasticities. - 22 - Similarly, the extent to which a particular good is complement or a substitute to its import equivalent varies according to the relative size of trade substitution and own-price elasticities. 1/ Exports can be treated in various ways in CGE models, following the type of country under study. The modelling of exports can be based on assumptions ranging from the small country assumption with fixed export output ratios to product differentiation with downward sloping demand on the export market. In the Ivory Coast model, we used a third option in which the ratio of exports to output in a particular sector is assumed to be an increasing function of the ratio of export to domestic prices. This specification, which is retained for the modelling of manufacturing sectors, can be thought of as a formalization of the arbitrage open to the domestic producers, who would choose between offering their output on the domestic market or on the export market depending on the relative incentives perceived in both markets. This approach allows to model phenomena which cannot otherwise be properly handled at the level of aggregation used in this type of model. Finally, it is possible to consider various import rationing situations in the framework of CGE models. A quantitative import quota mechanism, which is used in the Ivory Coast model, is described in Section 3 below. Premia rationing, corresponding to the case in which a parallel market for import licenses exist in a country, or simple quantitative rationing by direct allocation can also be analyzed, but were not used in our model. 2/ 1/ Goods whose trade substitution elasticity exceeds (is smaller than) their own-price will be regarded as import substitutes (complements). 2/ See Lewis and Urata [1983] for a model where these two specifications are taken into account. - 23 - 3. The Ivory Coast Model This section first presents the data sources and the principal characteristics of the model in terms of aggregation level, behaviour, important assumptions and market clearing mechanisms. It then turns to a more detailed discussion of some particular specifications, which have been designed to reflect some important features of the Ivorian economy. A presentation of the way effective protection can be defined and measured in these models follows. This is a key discussion in this study as it deals with concepts which are at the center of the ongoing reforms in the country as well as of the experiments presented in the paper. Finally, we discuss the choice of trade substitution and export supply elasticities. (i) Data Base The data sources used for the model come from Budgets Econaniques (Economic Budgets), Comptabilite Nationale (National Accounts), Centrale des Bilans (Census bureau for corporate societies) as well as from a 1979 Consumption Budget Survey and Ministry of Agriculture data. Data from these various sources have been made consistent in the frameTork of a Social Accounting Matrix (SAM), presented in Annex 2. A detailed description of its construction is given in Michel and Noel [1983]. (ii) Principal Characteristics The main features of the model, i.e. the behavior of the various agents, the principal assumptions, and the market clearing mechanisms, are described in this section. The mathematical formulation of the model is given in Annex 1. Annex 2 provides, together with the SAM, other statistical information such as production structure, input-output matrix, and consumption shares. The production side of the economy is disaggregated into fifteen - 24 - sectors. Of these, six are agricultural sectors, six industrial, two non- traded, and one services sector. Firms are assumed to be price-taking agents who maximize their short-run profit to determine their labor demand and output supply. Capital stocks are fixed for a given period. Intermediate consumption is required in proportion to output according to a fixed structure by product. Some sectors have a fixed output in the short run and, accordingly, fixed labor demand and intermediate consumption. This is the case of "forestry" and "mining and petrl'eum" sectors, both of which have been assumed not to be price-responsive in the short-run. This latter situation must not be confused with that of sectors whose prices are set by the government and which therefore also appear as having an essentially fixed output in the short run (see iii-(a) below). Public and private investment levels result from the choice of the "closure" rule (see (iii) below). They are broken down according to two different structures in order to determine investment by sector of destination. Investment by sector of destination is in turn converted into investment by sector of origin, which forms part of the final demand. The model distinguishes five household categories (two rural and three urban) corresponding to five labor categories. Each household category's income is the sum of its labor renumeration, its profit share, its transfers and share of remittances from abroad. The allocation of profits, transfers and remittances between labor categories is exogenous (see Michel and Noel [1983]). Labor remumeration is the outcome of the labor market mechanism (see below). The share of consumption and savi-ngs into post-tax income is constant and real consumption is broken down into consumption demands by product through a Cobb-Douglas consumption function. All savings are collected into a "savings pool" (see (iii)). - 25 - Labor markets are, at this stage, modeled in a very simple way. Labor demand for each labor category is the result of firms optimizing behavior. Nominal wages are assumed to be fixed in the short run, reflecting minimum wage policies in effect in the country. Labor supply is always in excess of labor demand at the prevailing wage rate in the short run, and unemployment is therefore endogenous. The government collects income taxes, tariffs, indirect taxes, and the Stabilization Fund surplus (or deficit), as well as a fixed share of the exogenous capital inflows. Part of this revenue is saved and the rest is spent in transfers, subsidies, wage remumeration, and public consumption. The balance of the government budget is done either on public consumption or on public savings, according to the closure rule (see (iii) below). All goods are composites of domestic and imported goods, aggregated through a constant returns to scale GES function, following the Armington assumption of imperfect substituability (see section 2 above). The relative share of imports and domestic production of any particular good is thus a fumction of their relative prices. The allocation of domestic production between export and domestic market is a function of the relative prices on these markets, derived through a logistic function (see also section 2 above). In general, product markets are price clearing, although some domestic prices may be set by the government. In such cases the markets are cleared by quantity adjustment, with the volume of exports as the endogenous variable (see iii (a) below). Factor markets are solved for unemployment since nominal wages are fixed and do not adjust to match demand and supply. When a flexible exchange rate regime is chosen, the exchange rate then adjusts to generate a given level of deficit or surplus of the balance of trade (or, equivalentely, of the balance of payment since transfers with the rest of the - 26 - world and capital rmovements are exogenous) . The model is solved following the "product market strategy" described in Annex B of DMR. In this method, factor marlcets are solved Eirst and products markets are then solved based on the solution for factor markets, after having been ordered in a particular and convenient way. (iii) Particular Specifications The Ivory Coast model contains three particular specifications which have been designed to take into account important aspects of the Ivorian economy. These are: the existence of fixed prices in the agricultural sector and the Stabilization Fund mechanism; the presence of quantitative restrictions on the imports of some goods; the large share of public investment in total investment. We describe in this subsection how these characteristics have been taken into account in the model. (a) Fixed Prices in the Agricultural Sector and the Stabilization Fund The Stabilization Fund is essentially an institution designed to shield domestic producers of certain agricultural products from the world market fluctuations. It is taken into account in the model by allowing for the possibility of exogenously setting the nominal price received by producers. The positive (or negative) difference is then received (or paid) by the Stabilization Fund account which, in turn, transfers its surplulses to (or has its deficit financed by) the government. This captures the essence of the stabilization mechanisms of the country. It should be noted that the implementation of such a mechanism has no consequence on the specification of exDort behavior (such as arbitrage between the two markets) and is compatible with any price determination mode on the world market of exports (such as a downward sloping curve for exports). It is also possible to set certain domestic prices in the model, as - 27 - is the case in the Ivory Coast for several agricultural products. However, it is then also necessary to specify how the market clears for excess supply or demand since price clearing is by definition excluded. The possible solutions range from adjustment of capacity utilization to adjustment through competitive imports or changes in exports. Since exports crops are the goods whose domestic prices are fixed in this model, we have assumed that excess supply (demand) in the domestic marke.t of these goods was just balanced by an additional (smaller) volume of exports. In such a case, the ex-ante arbitrage decision which may have been made (see Section 2 above) will be superseded. Export as well as domestic prices have been assumed constant for coffee, cocoa and "other agricultural products," which account for products mainly with fixed prices (palm oil, rubber), (b) Import Quotas The limitation in the number of import licenses became a common practice for a growing number of imported goods in the Ivory Coast as a way of protecting the domestic industry (see Chapter 5 below). Although this has not necessarily led to binding quotas because the license ceiling has sometimes been larger than requested, the volume of imports is effectively limited in some key sectors, such as textiles. Quantitative import restrictions amount to a fixed supply of imports in the economy and reverse the logic of the treatment of imports in the model. Indeed, if the availability of imported goods is limited and binding and if there is no mechanism of direct allocation of imports (i.e., quantity rationing with a rationing scheme), thi. only possibility of clearing the demand for imports is to increase their domestic price. Since world prices and tariffs are supposed to be fixed and exogenous, this results in a pure rent for the importers, at a given exchange rate. Therefore, whereas in the - 28 - usual case (small country assumption), the domestic price of imports is exogenous and its volume is endogenous, world supply being infinite, the volume of imports is exogenous when quotas are binding and their domestic price is endogenous. The resulting rents are separately accounted for in the model and distributed to "urban high incorAe" households, which are assumed to include all license holders. (c) Treatment of the Investment-Savings Problem The balance between investments and savings, which is an ex-post necessary condition for equilibrium, is not ex-ante warranted given the specifications usually retained in CGE models. This is the so-called "closure" problem, which requires an ad hoc choice based on the way the modeler thinks the economy works to eliminate this dilemma ("closure rule").1/ In the particular case of the Ivory Coast, public investment as a whole (i.e. central government plus public enterprises) represented more than three-quarters of total investment in 1980 (see Chapter 1). This suggests a specific treatment for the investment savings problem. The government savings rate and overall public investment can be realistically regarded as policy variables, whereas private investment and private savings rates can be safely considered fixed and exogenous in the short run. This means that the problem of matching savings to investment is in fact an implicit policy choice in the Ivory Coast. This choice is integrated explicitly in the model. In this context, an investment-driven model should be interpreted as accounting for a policy whereby the government decides to implement a given public investment program 2/ and accepts the consequent adjustment of its savings rate. 1/ See Lysy [1982], Dewatripont and Michel [1983]. 2/ Investment levels can be set either in nominal or real terms. - 29 - Conversely, a savings-driven model must be regarded as a model in which the government decides to adapt the level of public investment to the available savings of the economy. Accordingly, we distinguish in the model between public and private investment. Public investment is split into central government and public enterprises, with different structures for origin and destination, 1/ and is scaled in order to make overall investment match available savings in a savings-driven model (with proportional adjustment on the two components). Private investment has another structure of origin and destination and a.s broken down for accounting purposes into households, oil and non-oil investment. (iv) Effective Tariff Protection and Nominal Tariff Structure In the framework of its program of incentive reforms in the industrial sector, the Government is preparing a comprehensive revision of the Customs Code, with the objective of equalizing effective tariff protection rates across subsectors and, within each subsector, between the various degrees of processing. Given the importance of such a reform for the economy, it is fundamental to concentrate on the treatment of the concept of effective protection in the context of the CGE model before turning to a discussion of the corresponding policy simulations. In this section, we first recall the definition of the effective rate of protection in a partial equilibrium framework, and briefly discuss the problems associated with the treatment of non-tradeables in that context. We then turn to the definition of the effective rate of protection in the framework of the CGE model, and discuss I/ The structure of origin of investment concerns its capital goods composition. Its structure of destination indicates the allocation of these goods to sectoral capital stocks. - 30 - the relationships between the partial and general equilibrium approaches to the estimation of effective protection. Finally, we discuss the treatment of the equalization of effective tariff protection rates across subsectors in the context of the CGE model. In particular, we emphasize the impact of the aggregation problem arising from the estimation of ERPs in the framework of the CGE model, and we present the approach retained in the Ivory Coast model. By definition, the effective rate of protection (ERP) for a given sector or product measures the excess of value added per unit of output for that sector or product over its reference value added per unit of output under free trade, expressed as a percentage of the latter value. The main difficulty in estimating this indicator arises from the need to estimate of value added tnder free trade, short of a full general equilibrium model which seldom exists at the desired level of disaggregation. In a partial equilibrium context, it is possible to obtain a simple expression of ERPs under the following restrictive conditions: (i) all goods are tradeables and are perfect substitutes for imports; (ii) no substitution is allowed between primary factors and intermediate inputs, nor among the latter; (iii) there are constant returns to scale in production; and (iv) the elasticity of supply of imports is infinite.-1/ The existence of non- tradeables in the economy complicates the matter since, as long as their elasticity of supply is non-zero, the imposition of nominal protection on a traded good will yield an indirect protection on some non-traded goods 1/ The ERP for a given sector or product i then writes p. - Ea,. p. ERP = J 1+t. - ji 1+t where a. are the domestic input-output coefficients; p the lomestic prices and t the tar ffs. - 31 - through the set of input-output relationships. It then becomes difficult, in this partial equilibrium setting, to separate the direct from the indirect effects of protection. Three methods have been proposed to deal with this problem. The first method is due to Corden [1971], who proposes to measure jointly the direct and indirect effects of protection. This is done by applying a decomposition procedure, in which non-traded intermediate inputs are decomposed into their traded intermediate inputs and value added components through the input-output chain. The traded intermediate inputs used in the production of non-traded inputs are then combined with intermediate traded inputs used directly in the activity under study. Value added generated in the production of non-traded inputs is combined with direct value added in the same activity.!' Conversely, Balassa [19711 assumes that the elasticity of supply of non-traded goods is zero, and therefore that they do not benefit from any indirect protection. The Balassa ERP measure therefore takes into account only the direct effect of protection on traded goods. The third method is that of Little, Scitovsky and Scott [1970]. Assuming a single non-traded good in the economy, these authors propose to assign to this good a fictive tariff equal to the degree of overvaluation introduced by the structure of protection, and to treat the non-traded good as traded in subsequent calculations. As shown by Ray [1973], however, these three methods are limited because they assume that non-traded goods are "pure" intermediates, i.e. that they are not directly consumed. Once this restriction is removed, none of the three methods holds in a general equilibrium sense and the only solution left is to solve a general equilibrium model to obtain the reference free trade value-added prices of non-traded goods. 1/ The ERP for a given sector i then writes exactly the same way as before with, however different input-output coefficients (aji), reflecting the effects of the decomposition. - 32 - The problem is still more complex in the CGE framework since we assume imperfect substitutability between domestic and imported goods. In this context, the assumption that the domestic price of tradeables is entirely determined by their world prices plus the impact of tariffs and other protective measures no longer holds. Therefore, the estimation of the effective rate of protection must be carried out in two steps. First, the model is run with all tariff or non-tariff barriers removed to yield the reference value-added prices. Second, the model is solved with the existing tariff and non-tariff barriers in order to yield value-added prices under protection. The general equilibrium ERP (ERPG) is then deduced by comparing these two sets of prices. The ERPG indicator is not equivalent to the rate of net effective protection derived under the partial equilibrium methodology. This is because, following the imperfect substitutability hypothesis, ERPG is no longer computed with respect to the import price of the perfect import substitute, but with respect to the reference price of the composite good actually consumed, which is a bundle of domestic goods and actual imports. The imperfect substitutability hypothesis has therefore a strong implication for the link between effective protection and nominal tariff structure at the sectoral level. In sectors where the composition of imports and domestic goods differ substantially and where substitution elasticities between the two types of goods are very low, a given tariff on actual imports will have relatively less impact on the net price of the domestic good and therefore less protective effect on domestic production, by contrast with the protective effect obtained in a sector where the composition of imports and domestic goods is similar and where substitution elasticities between the two types of goods are high. For example, imagine an automotive sector where imports are dominated by luxury cars and domestic production consists of - 33 - assembly of low-price cars. In such a situation, the Armington specification will entail a very high tariff on imports of luxury cars to attain a given level of protection for the local production of low-price cars. By contrast, the partial equilibrium methodology computes the level of protection by estimating the impact of tariffs and other protective measures on the price of the local product, taking the border price of the perfect substitute as a reference. At the level of aggregation usually retained in CGE models, the level of effective protection resulting from the Armington specification will therefore be different from the one estimated through the partial equilibrium methodology. Furthermore, within the latter approach, different results will be obtained depending on the treatment of non-tradeables, Accordingly, the structure of nominal sectoral protection generated as a result of the equalization of effective protection rates across sectors will vary with the concept of effective protection used as a benchmark in the calculations. The ERPG indicator is clearly the adequate measure of effective protection from the point of view of consistency with the general equilibrium framework of the CGE model. However, it suffers from an aggregation bias, which is essentially due to differences in composition, and therefore to the presence of complementarities, between the bundles of domestic and imported goods that comprise the composite good at the sez_toral level. In a fully disaggregated CGE model, this bias would disappear, as the ERPG would be estimated with respect to close import substitutes only. However, this property does not hold in the context ot the 15 sector Ivory Coast model, and the ERPG indicator will in general differ substantially from the standard ERP that would result from the averaging of ERPs estimated on a product by product basis. For these reasons, and despite its known weaknesses in a general equilibrium context, we decided to retain the Corden ERP masure as an indicator - 34 - of effective protection in the following experiments (see Chapter 4). (v) Trade Substitution and Export Supply Elasticities We discuss in this section the determination of the estimates retained for the trade substitution elasticities and for the elasticities of the export supply functions. A plausible range of trade substitution elasticities was first determined from cross-sectional evidence based on CGE modelling experience in other developing countries. The protected sectors in the model were then grouped into three different categories corresponding to high, medium and low trade substitution elasticities, on the basis of our knowledge of the product composition within each individual sector in the Ivory Coast, Under the central set of estimates, which has been retained for all base experiments, the medium estimate for trade substitution elasticities is 0.8, indicating that a 1.0% change in the relative price of imported to domestic goods induces a 0.8% change in the ratio of imported goods to domestic goods in the composite commodity demanded by domestic agents. Low and high estimates were set at 0.6 and 1.1, respectively. Two alternative sets of estimates were retained for the purposes of the sensitivity analyses, corresponding to the high end and the low end of the range of realistic values. A similar approach was followed in order to determine the elasticities of the export supply functions. A central estimate of 0.5 was applied to the base experiments across all sectors where export arbitrage occurs, corresponding to the middle range of plausible values. This elasticity indicates that a 1.0% change in the relative price of exports to domestic goods induces a 0.5% change in the ratio of exports to sales on the domestic market by exporters. Two alternative estimates were used for the purposes of sensitivity analyses, corresponding to the high end (1.0) and the low end (0.25) of realistic values. - 35 - II. POLICY EXPERIMENTS CHAPTER 3: EXCHANGE RATE ADJUSTMENTS In the first chapter devoted to policy experiments, the short-run effects of exchange rate adjustments are considered. In Section 1, we discuss the significance of the exchange rate issue in the current economic situation of the Ivory Coast. We describe in Section 2 the economic mechanisms involved in the experiments where tke exchange rate is adjusted. Finally, in Section 3, we present the analysis of the experiments. 1. The Exchange Rate Issue As a member of the Franc Zone, the Ivory Coast has had a fixed nominal exchange rate vis-a-vis the French Franc since independence. In purchasing-power-parity (PPP) terms, the exchange rate of the CFA Franc (CFAF) as against the currencies of the Ivory Coast's tradling partners remained relatively stable until the mid-1970s. Taking 1973 as a base, the index of the PPP exchange rate averaged 97.6% from 1960-1974, fluctuating within a range of about 6.0%. However, it appreciated sharply after 1975, as a result of the acceleration in the rate of domestic inflation, especially following the coffee and cocoa boom in 1974. In comparison to 1973, the rate of appreciation reached 28.7% in 1977 and remained at about 25% until 1980. A depreciation in PPP terms with respect to the 1977-1980 situation then started in 1980. This was mainly the result of the slowdown in the rate of domestic inflation after 1980, and of currency movements within the European Monetary System and between those currencies and the US dollar. Accordingly, in the first quarter of 1983, the degree of overvaluation of the CFAF was estimated at only 17.5% as opposed to its 1960-1974 value. Nonetheless, the PPP index, which estimates the degree of appreciation of the exchange rate resulting from differences in inflation - 36 rates at home and abroad, is not an adequate indicator of the overvaluation of the domestic currency as it does not take into account the effects of changes in terms of trade and transfer payments on the country's foreign exchange position. The adequate measure is the equilibrium exchange rate, defined as the exchange rate that equals the supply and demand of foreign exchange during a certain period for a given set of trade taxes and subsidies. Changes in the equilibrium exchange rate can be decomposed into several factors. Among the most important are differential inflation, changes in terms of trade, and changes in transfer movements with the rest of the world. 1/ Several factors suggest that the CFA Franc has in fact remained largely overvalued from the point of view of the Ivory Coast, in spite of the apparent improvement shown by the PPP index. The reduction in inflation differentials with the country's trading partners, which led to the depreciation of the PPP index starting in 1980, was accompanied by a sharp deterioration in the terms of trade, estimated at 48.6% between 1977 and 1982, of which 16.9% during the 1980-82 period. Between 1977 and 1982, the current account balance deteriorated rapidly, as a result of both rising interest payments on external public debt and continued increases in salary remittances abroad. The former increased from 4.1% of exports of goods and services in 1977 to 21.0% in 1982, while the latter grew from US$ 344.2 million in 1977 (or 14.3% of exports of goods and services) to US$ 436.1 million in 1982 (or 19.2% of exports of goods and services). These factors also led to a deterioration of the overall balance, which turned into deficit starting in 1979. By the end of 1982, continued deficit of the overall balance of 1/ See DMR, Chapter 6.3 and 10.3. - 37 - payments resulted in a negative net foreign assets position of US$ 1.2 billion. During the rest of the decade, current price projections indicate no further deterioration of the Ivory Coast's terms of trade. Moreover, reductions in imports following the increase in domestic oil production, and the possibility of oil export surpluses of 500,000 tons a year after 1987, are likely to help the current account situation. However, it is not clear whether this improvement will result in a corresponding reduction of the degree of overvaluation of the CFA Franc. The availability of oil surpluses could presumably lead to an increase in the inflation differential with respect to the country's trading partners. Furthermore, despite these increases in domestic oil production, continued deficits of the overall balance are projected over the rest of the decade. The debt burden in comparison to exports of goods and services will continue to increase until at least 1985, and lower inflows of external public capital resulting from sharp cuts in the public investment program until 1985 will not be matched by corresponding reductions in the current account deficits, at least under the present set of trade policies. Given the position of the country within the Franc Zone, many difficulties arise in addressing the exchange rate issue. The exchange rate policy of the Ivory Coast must be understood within the framework of the Western Africa Monetary Union, and any modification of the parity of the CFA Franc with respect to the French Franc should be discussed in that context. However, alternative arrangements are possible within the existing institutional framework. In particular, it should be recognized that there is no conceptual incompatibility between the continuation of the convertibility agreements and the introduction of periodical parity modifications within the Franc zone. - 38 - Ultimately, the analysis of flexible exchange rate policies should be carried out from a medium to long-term perspective in a dynamic framework. However, in our opinion, comparative statics experiments carried out with a general equilibrium model can provide a useful first contribution toward a better understanding of this issue from several points of view. First, there is a need to address the issue of overvaluation beyond simple estimates of the PPP index. Second, there is a need to quantify the impact of flexible exchange rate policies on the economy as a whole. Third, it may be useful to concentrate on the analysis of the impact of such policies in the context of a single period model, before expanding the analysis to take into account dynamic effects. 2. Economic Mechanisms Involved The analysis of flexible exchange rate policies presented in the next section consists of simulating the response of the economy to a tighten- ing of the foreign exchange constraint. This is done under alternative hypotheses concerning key behavioral parameters of the model, such as the response of exporters to changes in relative prices at home and abroad, or under alternative foreign trade and incentive policies, particularly concerning quantitative import restrictions in key sectors. Starting from a position of an overall balance of payments deficit, a tightening of the foreign exchange constraint calls for a devaluation, which primarily results in a change in the relative price of imported to domestic goods, and in a reduction in the available savings to the economy. The overall net effects will depend on "sectoral tradeablility" characteristics and on the pattern of demand adjustment following the reduction in the inflow of foreign savings. A devaluation raises the relative domestic price of close import - 39 substitutes and exportables, and lowers the relative domestic price of less tradeable commodities or commodities that behave as import complements. On the production side, this relative price change induces an expansion in the production of exports and import substitutes, and a contraction in the production of less tradeable commodities and import complements. On the demand side, an expenditure switching effect takes place, followed by a reduction of domestic demand for exportables and import substitutes, and an increase in domestic demand for non-tradeables and import complements. Overall, the effects of the devaluation on the sectoral net prices structure will depend on a number of "sectoral tradeability" characteristics, namely: the share of exports in total production; the share of imports in total domestic demand; the ratio of imports to intermediate inputs; and trade substitution elasticities. A devaluation raises the relative net price of sectors with high export shares; has a larger impact on sectoral net prices, the higher the share of imports in total demand; lowers the relative net price of sectors with a high ratio of imports to intermediate inputs; and tends to raise (lower) the net price in sectors with high (low) trade substitution elasticities, or import substitutes (complements). Sectoral tradeability indicators for each sector are given in Annex 3. The extent to which these effects mutually reinforce or counterbalance each other can be determined only in a general equilibrium framework. The pattern of demand adjustment following the contraction of the inflow of foreign savings is the other key determinant of the structure of relative net prices under a flexible exchange rate regime. When the model is supposed to be savings-driven (see Chapter 2.3), investment adjusts to match available total savings (i.e., domestic and foreign). Although the expenditure-reducing effect of the devaluation may also induce a contraction - 40 - in real private and public consumption, the burden of adjustment falls primarily on investment demand, resulting in a downward pressure in the relative price of capital goods as opposed to consumer goods. On the other hand, when the investment level is to be maintained, savings ad just to bring the domestic gap in line with the foreign exchange gap. In this case, the burden of adjustment falls on consumption, both public and private, and the expenditure-reducing effect of the devaluation is relatively weaker on investment. This results in downward pressure on the relative price of consumer goods and capital goods. Finally, the impact of the devaluation will depend on quantitative import restrictions. The adoption of an import quota will tend to reduce the impact of the devaluation on the net price in the sectors, in cases where quantitative import restrictions are binding. The devaluation will result in this case in a reduction of the import premium accruing to license holders. Correspondingly, in contrast to a scenario where quotas remain binding, the removal of the quota in conjunction with a move to a flexible exchange rate regime will result in stronger sectoral net price changes, as a larger devaluation will be required to reach balance of payments equilibrium, and as its effects will be directly transmitted to import prices. Just as the prices of outputs in all sectors are relative prices, the exchange rate is treated as a relative price in the CGE model. In this sense, changes in the exchange rate refer to changes above the differential between domestic and world inflation. A change in the nominal exchange rate from period to period just offsetting differential world and domestic inflation rates would correspond to a constant real (or price level deflated) exchange rate. Only if the change in the exchange rate exceeds (falls skort of) this differential will there be a depreciation (appreciation) in the real exchange rate. -4 1- 3. Analysis of the Experiments In this section, we analyze the impact on the economy of a tightening of the foreign exchange constraint in a situation where the exchange rate is allowed to adjust. First, we present the results of a base experiment carried out under a central set of estimates for trade substitution and export supply elasticities. We then turn to a series of sensitivity experiments realized under different assumptions on these elasticities. Next, we discuss an experiment carried out under an alternative assumption concerning the adjustment of the investment savings gap. Finally, we present the results of an experiment where balance of payments equilibrium is imposed as a constraint on the economy. These experiments have been carried out under the assumption that produicer prices in the agricultural sector are kept constant in real terms in the short run. This assumption resulted in only marginal variations (generally less than one percent across experiments) in the corresponding value added prices following the adjustments in the exchange rate. Also, import quotas have been assumed to remain in affect throughout the experiments. In the interpretation of this section's results, it must however be kept in mind that the effects of the removal of quotas may be significant, as shown in Chapter 5. The objective of all but the last experiments of this section is a reduction by a half of balance of payment deficit. Such a reduction of CFAF 70 billion out of an overall deficit of CFAF 145 billion in the reference year corresponds to an ex-ante decrease of foreign savings by 3.5% of GDP. This undoubtedly represents a drastic stabilization effort which, nonetheless, remains in line with the targets set under the original IMF program (see chapter 1.2). In the last experiment, the balance of payment objective is a -42- zero deficit. The objective of this experiment is not to simulate the adjustment of the economy to an operational balance of payments target in the short-run, but rather to estimate the theoretical equilibrium exchange rate as defined above. (i) Base Experiment The base experiment has been carried out with the central set of estimates for the trade substitution and export supply elasticities. It shows that the balance of payment target can be achieved with a real devaluation estimated at 9.5%, accompanied by a decrease in GDP at market prices of 2.0%. The most important elements of the adjustment of the structure of the economy called for in the experiment are: a substantial improvement in public finance; a sharp decrease of household income and consumption; and dramatic shifts in the structure of relative prices and external trade. Detailed results are presented in table 3.1 below. The decrease in GDP and the dramatic change in its structure can be seen as essentially the result of the decline in investment (-6.6%) following the reduction of the inflow of foreign savings in the economy induced by the reduction in the balance of payments deficit. Construction value-added declines by a sharp 6.0% and services value-added drops by 1.9%, while agriculture remains practically constant and industry declines by 0.9%. Similar changes occur in the structure of output. Illustrative of the impact of investment demand, for instance, is the fact that the output of the engineering industry sector declines (-2.6%), because the increase in its exports (+1.0%) and the decrease in competing imports (-6.9%) which are induced by the devaluation cannot offset the decline in demand for equipment resulting from the lower investment level. By contrast, output remains practically constant in food processing and textiles (+0.1% and 0.5%, respectively). -43- Table 3.1 Summary of Base Experiment (percentage deviation from reference year) Real Devaluation +9.5 Aggregate Figures (volumes) GDP of Agriculture -0.1 Industrial Exports (in $) +5.0 Industry -0.9 Total Exports (in $) +1.3 Construction -6.0 Total Imports (in $) -6.9 Services -1.9 GDP at Factor Prices -1.6 Composite Goods Prices +2.1 GDP at Market Prices -2.0 Net Prices -0.9 Capital Goods Prices +3.3 Investment -6.6 Import prices +8.4 Government Consumption -2.6 Private Consumption -4.8 Import/Domestic Supply -7e2 Exports/Domestic Output +3.4 Industrial Output -0.8 Total Output -1.8 Sectoral data for some selected sectors Composite Value-added prices prices 1/ Output Exports(in $) Imports(in $) Subsistence crops -2.3 -5.0 -0.1 ---- -10.4 Mining petroleum +8.5 +9.2 0.0 +1.4 -3.3 Food processing +1.0 +1.5 +0.1 +6.4 -9.4 Textiles -2.9 +0.3 +0.5 +5.4 -0.3 Chemicals +5.7 -0.3 -1.2 +2.6 -8.0 Const. materials +3.9 +1.7 +0.4 +4.7 -10.1 Engineering +7.5 -1.2 -2.6 +1.0 -6.9 Utilities +0.1 -1.8 -2.1 ---- ---- Construction +1.4 -3.2 -6.0 ---- ---- Services +0.8 -0.8 -1.9 +0.8 -8.0 1/ Value-added per unit of output. -44- The public finance effect is quite significant, as government revenue increases by 4.8%, resulting in a substantial increase in public savings (+25.8%). This is the result of the combination of two important factors. The first factor is the increase in the Stabilization Fund surplus by 32.4%, bringing its share in government revenue from 16.1% to 20.3%. This results directly from the increased differential between export and domestic prices of agricultural products following the devaluation, as export prices in domestic currency terms increase more than proportionately by comparison with domestic producer prices. This effect demonstrates the importance of the Stabilization Fund mechanism, not only because of its obvious role as a source of revenue for the public budget, but also because of its impact on the overall level of investment through the public savings effect. In particular, if the additional revenue generated by the exchange rate adjustment had been completely passed on to producers, the pattern of demand would have been quite different, and the evolution of the economy would have been substantially affected through the impact on investment and capital formation. The second factor is the 2.5% increase in tariff receipts originating from the fact that the decline in the volume of imports is more than offset by the increase in their domestic price due to the devaluation. The result is an overall increase in government revenue despite a slight fall in direct tax receipts (-4.5%). Since current public expenditures (i.e. consumption and transfers) are assumed to remain constant in nominal terms in the short-run, this translates into a substantial increase in govermnent savings. However, this increase is smaller than that of public income since, on the expenditure side, the domestic currency value of -45- transfers abroad rises as a consequence of the devaluation. - This increase in government savings raises the share of public savings in total savings from 14.4% to 18.8% and nearly offsets the decrease in foreign and domestic savings (-10.0% and -4.5% respectively) which occurs as a result of the reduction in the balance of payments deficit and of the decrease in household income. Overall, nominal savings decline only by 3.6%, and the drop in real investment results mostly from the increase in capital goods prices. As a direct consequence of the decrease in GDP, nominal household income drops by 5.1%. Since households remittances abroad are also fixed in foreign currency terms, -/ disposable income is further deteriorated and domestic savings and consumption by households are reduced by 4.5% and 4.4% respectively in nominal terms. Relative sectoral value-added pries are substantially modified as a result of the exchange rate adjustment, following the marked shifts in demand and supply patterns, and also following the structure of the sectoral tradeability indicators (see 3.2 above). Investment goods (engineering goods and construction) as well as less tradeable sectors (such as utilities or services) see their relative prices drop sharply, whereas the prices of more tradeable sectors increase significantly. Finally, the pattern of foreign trade is significantly modified, as the economy becomes more export oriented and less import dependent. Industrial exports increase by 5.0%, inreasing the ratio of export to domestic output by 3.4% with respect to its reference year value, while the 1/ To the extent that those transfers represent interest payments, this represents a reasonable assumption. 2/ Realistic in the short-run, this assumption will have to be modified in a longer-run analysis. -4 6- ratio of imports to domestic supply decreases by 7.2%, as a direct result of the changes in relative prices induced by the adjustment. (ii) Sensitivity Analysis The role of trade elasticities in the adjustment has been emphasized in section 2 of this chapter. As these elasticities determine the extent of export and import movements following changes in relative prices, it is important to assess the sensitivity of the preceding results to the value of these elasticities. Whereas the base experiment has been carried out under a medium set of values for trade and export elasticities (see the numerical values in chapter 2.2), experiments 1 and 2 presented in this subsection have been carried out assuming, respectively, a high and a low export response, keeping trade elasticities at their medium values. Experiments 3 and 4 have been carried out with high and low trade elasticities, keeping export elasticities constant. The results are presented in the first five columns of Table 3.2 below. The experiments show the expected effects of different assumptions on the elasticities as to the size of exchange rate adjustments, increase in exports and reduction in imports. However, all experiments result in a similar contraction of GDP. Moreover, they confirm the importance of the public finance effect, which prevents a sharp drop in real investment following the increase in public savings. The size of the exchange rate adjustments is consistent with the assumptions on elasticities, increasing as export and trade elasticities decrease. The required devaluation varies between 7.6% and 11.2%, indicating a reasonable estimate for the exchange rate adjustment in the 8 to 10% range. The shifts in the trade structure vary accordingly. The ratio of imports Table 3-2 Summary of Sensitivity Analysis Experiments (Percentage deviation from reference year) Base High Export Low Export High Trade Low Trade Real Equilibrium Experiment Response Response Elasticity Elasticity Investment Exchange Experiment 1 Experiment 2 Experiment 3 Experiment 4 Maintained Rate Real Devaluation +9.5 +7.8 +10.5 +7.6 +11.2 +9.5 +31.1 Real GDP Agriculture -0.1 -0.1 -0.1 -0.1 -0.1 -0.1 -0.4 Industry -0.9 -0.4 -1.2 -1.0 -1.0 -0.4 -2.5 Construction -6.0 -7.0 -5.4 -7.4 -4.8 -0.3 -19.8 Services -1.9 -1.8 -1.9 -1.7 -2.2 -2.6 -6.0 Real GDP at Market Prices -2.0 -1.9 -2.0 -2.1 -2.0 -1.5 -6.6 Investment -6.6 -7.3 -5.8 -7.7 -5.8 0.0 -20.5 Government Consumption -2.6 -2,1 -2.9 -2.i -3.0 -24.4 -7.3 Private Consumption -4.8 -4.4 -5.1 -4.5 -5.3 -4.0 -14.1 Total Output -1.8 -1.8 -1.8 -2.0 -1.7 -1.0 -5.7 Industrial Exports (in $) +5.0 +8,9 +2.1 +3.9 +6.0 +5.7 +14.5 Total Exports (in $) +1.3 +2.0 +0.8 +1.0 +1.5 +1.3 +3.9 Total Imports (in $) -6.9 -6.1 -7.2 -7,0 -6.8 -6.6 -19.4 Composite Goods Prices +2.1 +1.7 +2.4 +1.7 +2.5 +2.2 +6.8 Value-Added Prices -0.9 -0.7 -1,0 -0.6 -1.1 -1.0 -1.6 Capital Goods Prices +3.3 +2.5 +3.8 +2,5 +4.1 +3.5 +10.5 Import Prices +8.4 +6.9 +9.3 +6.7 +10.0 +8.5 +28.3 Imports/Domestic Supply -7.2 -6.2 -7.6 -7.1 -7.3 -7.8 -11.6 Exports/Domestic Output +3.4 +4.1 +2.8 +3.2 +3.6 +2.5 +10.3 -48- to domestic supply varies between -7.2% and -6.1%, and the ratio of export to domestic output ranges from +0.8% to +2.0%. The change in investment varies significantly across experiments, ranging from -7.7% to -5.8%. This is a direct consequence of the public finance effet previously described: the larger the devaluation, the larger the surplus of the Stabilization Fund (between +25.8% and +38.3% the larger the increase in public savings (between +16.7% and +33.3% and therefore the larger real investment. Finally, it is interesting to note that the level of real GDP obtained in the different experiments is of similar magnitude. This is because the various effects of exchange rate adjustments on GDP tend to compensate one another. For instance, the experiment with the lowest exchange rate adjustment is also the one which is done under the assumption of a high export response (explaining why a lower exchange rate adjustment is needed), involving a relatively higher value-added in the industrial sector, which compensates for the relatively lower value added in construction anld services following the decrease in investment demand. Overall, this experiment results in a GDP level comparable to the one obtained under the experiment with the highest investment rate. (iii) Exchange Rate Adjustment with Constant Real Investment While we assumed in the previous analysis that the level of investment adjusts to available savings, in the following equipment government adjusts its savings rate in order to sustain a given level of overall real investment (See Table 3-2; column 6). This policy is preferable to the other ones presented in this chapter as it results in a higher GDP level and an increase in households' real income with an exchange rate adjustment of the same order of magnitude. However, this result is achieved only through a -4 9- severe reduiction in current public expenditure, which would undoubtedly be difficult to implement in the short run. A maintained real investment level and a fall in current public expenditure are the two elements which explain the central results of this experiment, namely an increase in construction value-added which drives overall GDP, a drop in utilities and services sectors activities and prices, and an increase in household real income. The much larger demand for investment goods sectors, as compared with the base experiment, explains why the structure of GDP is distorted in favor of construction, which declines only by 0.3% by comparison with a drop of 6.0% in the base experiment. It also explains the slightly better performance of the industrial sector which, as a result of the substantial increase in engineering industries and construction materials value added, increases slightly above its reference year value. As a consequence, overall GDP is higher than in the base experiment, yielding a higher nominal household income. The reduction in government expenditure is a direct consequence of the assumption made in the model concerning the savings-investment adjustment under an investment-driven regime, i.e. that public savings adjust in order to match the desired level of investment (see chapter 2.3). Thus, in this experiment, public savings increase by 74.0% above their reference year value, whereas the corresponding figure was 25.8% under the base experiment. Since the increase in public revenue due to the devaluation (Stabilization Fund and tariff effects) is not sufficient to meet the additional savings required, government expenditure make up for the difference, resulting in a drop of 22.8%. As government expenditure represents about 16% of the demand for the utilities sector and 9% of the demand for the services sector, this explains the drop in output, prices and value-added of these two sectors. -50- Overall, these two types of effects (boost of investment goods sectors and drop of utilities and services) compensate each other in terms of import and price levels. Compared with the base experiment, the increase in imports required by the larger investment is almost offset by the lower services imports. Consequently, the exchange rate adjustment is of the same order of magnitude (+9.5%) in both cases. A comparable price level is also obtained under both experiments, as the increase in investment goods prices is compensated by a decrease in prices for utilities and services, while the exchange rate is devalued by a similar amount. This, in turn, explains why the increase in nominal household income can be translated into an increase in real income, leaving households better off in this experiment. (iv) Equilibrium Exchange Rate This last experiment simulates the response of the economy to a hypothetical situation where balance of payments equilibrium would need to be reestablished in the short run. The objective of the experiment is not to represent an adjustment to an "operational" balance of payment target, but co obtain an estimate of the real equilibrium exchange rate at a given point in time under a central set of estimates concerning trade substitution and export supply elasticities. The results of the experiment are presented in Table 3.2, column 7. The required equilibrium of the balance of payments is achieved through a 31.1% of devaluation, which is accompanied by a 6.6% drop in real GDP at market prices and by a 20.5% drop in real investment. The adjustment of the trade balance is achieved essentially through a substantial decrease in imports (-19.4%), while total exports grow only by 3.9%. The growth in industrial exports (+14.5%) is significantly higher compared to the result obtained under the base experiment. As expected, the higher devaluation -51- results in a sharp increase in import prices and in capital goods prices, while the decline in value added net prices is more pronounced than under the base experiment. The estimate of the equilibrium exchange rate should not be interpreted as the rate of devaluation which would be needed in order to achieve balance of payment equilibrium within a given year. First, as shown in the previous analysis, this estimate is sensitive to the values assigned to trade substitution elasticities and export supply elasticities. Second, the estimate refers only to the real devaluation needed to yield the required adjustment. Any estimation of the rate of nominal devaluation needed to achieve the balance of payment target would require to take into account the impact of the inflation differential with the rest of the world, which would in turn require to incorporate monetary policy, credit mechanisms, and the process of financial intermediation in the model. However, as an estimate of the marginal value of foreign exchange for the Ivory Coast under the existing system of incentives, it provides a second best shadow price of foreign exchange in that reference year. -52- CHAPTER 4: TARIFFS AND EXPORT SUBSIDIES REFORMS IN THE INDUSTRIAL SECTOR In this chapter, we analyze the short term general equilibrium effects of changes in the tariff and export incentive structure in the industrial sector. In section 1, we discuss the relevance of these policy reforms in the context of the structural adjustment program in the Ivory Coast. We analyze in section 2 the economic mechanisms involved. Finally, in section 3, we present the experiments. 1. Policy Issues Up to the mid-seventies, the development of the industrial sector took place in the context of a relatively simple system of incentives. This system was based on moderate tariffs that applied fairly uniform protection between the different branches of activity, without reliance on quantitative import restrictions. This situation was modified by the 1973 tariff reform,which resulted in a more escalated tariff scale according to the degree of processing, and in substantial distortions in the degree of effective protection given to particular industrial branches. Average effective tariff protection coefficients (ETP) have been estimated on the basis of the average nominal tariff protection coefficients derived both under the pre-1973 and the post- 1973 tariff schedules.1 Between 1971 and 1978, the average effective tariff protection coefficient for the industrial sector as a whole increased from 1.39 to 1.76. The subsectoral ETP varied, in 1971, between 1.19 and 2.06 whereas the interal widened between 0.92 to 2.18 in 197 , with, for instance, increases of the order of 30% in canning and food processing and in grain products. 1/ See M. Noel [1982] -5 3- Starting in 1975, the proliferation of quantitative import restrictions induced increasing distortions in the structure of effective protection across subsectors. Empirical studies conducted at the firm level showed a high degree of correlation between the pattern of effective protection and domestic resource cost estimates, revealing considerable distortions in the allocation of resources toward activities that earn (or save) foreign exchange at a high cost in terms of domestic resources, and away from activities which could have developed at a comparative advantage for the country. Increasing protection for sales on the domestic market resulted in a growing bias against exports. Estimates based on 1978 firm-level data showed, for instance, a bias against exports of 54% in spinning, weaving and dyeing, 122% in other textile products, and 101% in canned and processed food. In this context, the Government is undertaking a comprehensive reform of the system of industrial incentives, based on three main elements: (i) a complete revision of the customs tariff, with the objective to equalize effective tariff protection coefficient across industrial subsectors and, within each subsector, between the various stages of processing; (ii) the implementation of an across-the-board subsidy on export value added for manufactured goods, designed to compensate for the degree of effective tariff protection on domestic sales; and (iii) the removal of quantitative import restrictions and their replacement by temporary import surcharges that will decline over time. The analysis of these reforms in a general equilibrium framework is of particular relevance for policy makers, who have to take into account their impact on internal and external equilibria. Ultimately, this analysis should be carried out in an intertemporal context using a dynamic general equilibrium -54- model. However, at this stage, the estimation of the main macroeconomic adjustments triggered by the reforms in a static general equilibrium framework is of considerable interest as a first step toward improving on analyses carried out in a partial equilibrium context. 2. Economic Mechanisms The analysis carried out in the following experiments consists in simulating the response of the economy to a tightening of the foreign exchange constraint under a fixed exchange rate regime. The adjustment is achieved through a change in the average level and in the structure of tariffs and through an across-the-board subsidy on export value added in the industrial sector, together with the removal of binding quotas. In line with the policy reform, the change in tariffs is constrained to yield equal effective tariff protection across industrial sectors. At the same time, the across-the-board subsidy on export value added is constrained to equal the effective tariff protection rate, in order to neutralize the bias against exports inherent in the system of tariff protection. As a result of the reduction in the inflow of savings from the rest of the world, the economy must achieve a corresponding reduction in its trade balance deficit through a decrease in imports anid an increase in exports. The decrease in imports results (i) from an expenditure-reducing effect, due to the impact on domestic demand of the reduction of foreign savings, and (ii) from an expenditure- and output-switching effect which results, under a fixed exchange rate regime, from the change in the level of tariffs. This switching effect could be achieved through an across-the-board increase in nominal tariffs for industrial products. In s ch an experiment, the structure of effective tariff protection across sectors would be modified only slightly after the adjustment, and the switching effect would result primarily in a -55- modification in the structure of output and demand between tradeables and non- tradeables, and to a lesser extent between tradeables themselves. On the other hand, the expenditure reducing effect can have a substantial impact on the structure of output and demand both between tradeables and non-tradeables and between tradeables themselves, depending on the type of adjustment assumed. In a savings driven model, the level of investment is reduced in line with available savings, and the burden of adjustment falls on investment goods. In an investment driven model, domestic savings must increase in line with the level of investment at the expense of consumption, and the burden of adjustment falls on consumption goods. In the following experiments, the switching effect is achieved through an overall increase in the average nominal tariff level in the industrial sector and through substantial changes in the structure of nominal tariffs to satisfy the constraint of equalization cf effective tariff protection across subsectors. As a result, the switching effect by itself will entail larger changes in the structure of output than in the previous case, in addition to the change induced by the expenditure reducing effect. The change in the structure of tradeables depends, as under ' 5 flexible exchange rate adjustment, on the tradeability characteristics of individual sectors. In the case of an import substitute, an increase in the tariff rate will decrease the demand for imports of that good and increase the demand for its domestic substitute. In the extreme case of a complementary good, (not present in the model), the demand for this good would not decrease and the expenditure devoted to this good would be higher, at the expense of other goods. Changes in the level of tariffs will induce changes in tariff revenue for the government, the direction of the change depending on the 5 6- relative volume and price effects. In turn, this change will induce a proportionate adjustment in another iten of the govermnent budget, with the net impact on the economy depending on the type of adjustment retained (consumption, transfer, savings). Contrary to a change in the level of tariffs, the introduction of an across-the-board subsidy on export value added does not modify directly the structure of domestic prices. The effects of the subsidy work through the possibilities of arbitrage between domestic and export markets for producers of manufactured goods. The value added subsidies make exports more profitable and will therefore generate a shift of production towards export markets, at the expense of domestic sales. This will be accompanied by a change in the level of production since the net price for exporters increases. As production increases, households' income and demand increase. The combined effects of this change in production and demand will affect the domestic price structure. The introduction of export subsidies also will induce a change in government expenditures. As in the case of tariffs, this change induces a proportionate adjustment in another item of the govermnent budget. 3. Analysis of the Experiments In this section, we discuss the effects on the structure of the economy of a tightening of the foreign exchange constraint under a fixed exchange rate of regime, with the adjustment made through an across-the-board tariff-cum-subsidy scheme in the industrial sector. First, we discuss the results of the base experiment, in which the model is run under the assumption that investment adjusts to available savings in the economy, with a central set of estimates for the trade substitution and export supply elasticities. We then turn to a discussioni of the sensitivity analysis experiments carried out on these trade parameters. Finally, we present an alternative version of -5 7- the base experiment where real investment is maintained at its original level in the reference year. As in the previous chapter, the experiments carried out in this section pertain to a reduction in the overall balance of payments deficit by one-half. (i) Base Experiment Under the central set of estimates for trade substitution and export supply elasticities, the balance of payments objective is achieved through an increase in the average tariff level of 77.9% (corresponding to an average nominal tariff rate of 33.8% compared to 19.1% before the adjustment), and through the introduction of an across-the-board subsidy of 57.8% on export value added which exactly offsets the effective protection rate obtained in the industrial sector, At the same time, the equalization of the Corden measure of effective tariff protection rates across industrial subsectors results in substantial changes in the nominal tariff structure. Tariff rates increase by more than 80% in sectors producing mainly intermediate goods, such as chemicals (+117.2%), construction materials (88.2%), and engineering industries (+158.8%), compared with relatively smaller increases in consumption goods sectors, such as food processing (+7.4%) and textiles (+31.6%). The adjustment to the balance of payments target is acccmpanied by an overall decrease of 2.1% of GDP, a substantial improvement in the public finance situation, a decrease in household inccme and consumption, and dramatic shifts in the structure of relative prices and external trade. Table 4-1 presents the main results of the experiment. -5 8- Table 4-1 Summary of Base Experiment (Percentage Deviation from Reference Year 1) Average Tariff Level 33.8% Average Tariff Increase +77.9% Average Effective Protection Level: 57.8% Average Export Subsidy Level: 57.8% Aggregate Figures (Volumes) GDP of Agriculture +0.1 Industrial Exports (in $) +11.4 Industry +1.1 TotalExports (in $) +1.6 Construction -7.5 Total Imports (in $) -6.6 Services -3.6 GDP at Factor Prices -1.8 Composite Goods Prices +2.5 GDP at Market Prices -2.1 Net Prices -2.4 Capital Goods Prices +8.2 Investment -8.4 Import Prices +10.2 Government Consumption -3.4 Private Consumption -4.1 Imports/Domestic Supply -5.6 Export/Domestic Output +4.0 Industrial Output +1.4 Total Output -2.1 Sectoral Data for Selected Sectors Composite Value Added Output Exports Imports Prices Prices _ ($) ($) Subsistence Crops -4. -5.2 -0.1 - -3.8 Mining & Petroleum -0.5 -6.7 +0.0 +0.9 -2.0 Food Processing -2.6 +2.4 +2.0 +10.9 -5.4 Textiles -14.5 +3.8 +1.9 +16.0 +25.7 Chemicals +13.8 +2.0 +1.4 +5.9 -13.2 Construction Materials +4.8 +8.9 +7.3 +19.1 -15.3 Engineering +22.2 -2.3 -2.8 -0.7 -12.7 Utilities -0.5 -3.3 -2.8 - - Construction +2.6 -5.3 -7.5 - - Services -1.1 -3.5 -3.6 -3.3 -4,4 1/ Unless otherwise noted. 7/ Value-added per unit output. -5 9- On the supply side, the decrease in GDP results from the sharp decline in construction sector value added (-7.5%) and in services value added (-3.6%), which contrast with the quasi-stagnation of agriculture (+0.1%) and of the industrial sector (+1.1%). This change in the structure of GDP results from the sharp decline of investment (-8.4%) that follows from the reduction in the inflow of savings from the rest of the world. The combination of the reduction in investment demand and of the sharp changes in the structure of trade incentives results in substantial modifications in the structure of industrial output. Output and exports expand significantly in food processing, textiles, and construction materials. In textiles, the expansion of exports and of output is accompanied by a marked increase in imports following the removal of quantitative import restrictions. This suggests the need for a drastic change in the composition of output inside this sector, toward production lines that can compete on export markets with the subsidy and away from those that were previously protected by the quotas. By contrast, output declines in chemicals and in engineering products, following the decline in domestic investment demand, despite the relatively higher increase in effective tariff protection in these two sectors. The public finance effect of the tariff-cum-subsidy schemes much stronger than the effect obtained through the adjustment of the exchange rate, and the resulting structure of public finances is also markedly different, Government revenue increases by 19.7% as against 4.7% under the experiment with exchange rate adjustment, as tariffs receipts increase by 67.8% from CFAF 158 billion to CFAF 266 billion. On the expend ture side, total utlays for the export subsidy amount to CFAF 27 billion. As the other components of public expenditure remain unchanged, this tznslates into a marked increase in government savings, which amount to CFAF 161 billion after the tariff-cum- -6 0- subsidy adjustment, compared to CFAF 115 billion under the exchange rate adjustment. This sharp increase in government savings practically offsets the decline in foreign savings, resulting in stable level of overall nominal savings and investment after the adjustment. Interestingly, a similar result was also obtained when the adjustment was made through a change in the exchange rate, although the relative weight of the public savings and of the foreign savings adjustments was different in the latter case. Under the exchange rage adjustment, the relatively smaller increase in public savings was compensated by a relatively smaller decline in foreign savings, as the reduction of foreign savings in dollar terms was partly offset by the devaluation. Under both types of adjustment, the decline in real investment therefore results mostly from the increase in capital goods prices. Real investment declines more sharply under the second type of adjustment, however, (-8.4% as compared to -6.6% under the exchange rate adjustment), as the introduction of the tariff-cum-subsidy scheme results in a larger increase in capital goods prices (+8.2% as compar,ed to +3.3% under the previous case). This is because the nominal tariff on capital goods rises more than proportionately with respect to the overall tariff increase in the industrial sector as a result of the equalization of effective tariff protection across subsectors. The structure of sectoral net prices is substantially modified as a result of the tariff-cum-subsidy adjustment. These modifications arise from the demand and supply effects generated by the adjustment, as well as from the change in the nominal tariff structure imposed by the equalization of effective tariff protection rates across sectors. By comparison with the results obtained under the exchange rate adjustment, changes in sectoral net -61- prices ace generally more pronounced, with larger increases obtained for textiles (+3.8% vs. +0.3%) and construction materials (+8.9% vs. +1.7%) and larger declines obtained in the case of non-tradeable sectors. This is because the non-tradeable sectors are negatively affected both by a relatively sharper drop in domestic demand and by a relatively higher increase in the price of their intermediate inputs, following the change in the structure of tariffs, by comparison with the situation resulting from the exchange rate adjustment. Finally, the trade orientation of the economy is substantially affected as a result of the tariff-cum-subsidy adjustment. Industrial exports grow by 11.4%, by comparison with a 5.0% increase under the exchange rate adjustment. (ii) Sensitivity Analysis This section presents a series of alternative experiments carried out under different sets of trade and export supply elasticities. Experiments 1 and 2 have been realized assuming a high and a low export response, respectively. Experiments 3 and 4 have been done with high and low trade elasticities, keeping export elasticities constant. The results are presented in Table 4.2 below. As in the case of the experiments carried out with exchange rate adjustment, the present sensitivity analysis shows the expected effects of the assumptions on trade and export supply elasticities on the level of effective protection and of expo't-' subsidy across sectors, on the increase in exports, and on the decrease in imports. On the other hand, the contraction of GDP following the ad ! istment to lower inflows of foreigni savings is remarkably similar across experiments. Table 4-2 Summary of Sensitivity Analysis Experiments (Percentage deviation from reference year) Base Experiment Experiment I Experiment 2 Experiment 3 Exeriment 4 Experiment 5 High Export Low Export High Trade Low Trade Real Investment Response Response Elasticity Elasticity Maintained Average Tariff Level 33.8 30.1 35.7 32.5 33.9 34.6 Average Tariff Increase +77.9 +58.6 +88.0 +70.8 +78.2 +82.3 Average Effective Protection Level 57.8 50.0 61.8 54.9 57.8 58.9 Average Export Subsidy Level 57.8 50.0 61.8 54.9 57.8 58.9 g 0' Real GDP Agriculture +0.1 +0.2 +0.1 +0.1 +0.1 +0.1 f Industry +1.1 +2.7 +0.1 +1.7 +1.1 .+1.6 Construction -7.5 -8.5 -6.3 -7.7 -5.8 +0.3 Services -3.6 -2.8 -309 -2.9 -3.5 -4.8 Real GDP at Market Prices -2.1 -1.5 -2.4 -1.7 -1.9 -1.7 Investment -8.4 -9..2 -7.8 -8.5 -7.9 0.0 Government Consumption -3.4 -2.6 -3.8 -3.2 -3.6 -33.7 Private Consumption -4.1 -2.8 -4.7 -3.4 -3.8 -3.1 Total Output -2.1 -1.6 -2.3 -1.8 -1.8 -1.1 Industrial Exports +11.4 +20.7 +5.7 +11.7 +11.9 +12.3 Total Exports (in $) +1.6 +3.4 +0.5 +1.8 +1.8 +1.6 Total Imports (in $) -6.6 -4.9 -7.5 -6.3 -6.1 -6.6 Composite Good Prices +2.5 +1.7 +3.0 +2.2 +2.6 +2.7 Value Added Prices -2.4 -1.5 -3.0 -2.3 -2.4 -2.8 Capital Goods Prices +8.2 6.2 +9.4 +7.6 +8.4 +8.8 Import Prices +10.2 +7.1 +11.7 +9.1 +10.2 +10.6 Imports/Domestic Supply -5.6 -3.5 -6.8 -5.3 -5.6 -6.9 Exports/Domestic Output +4.0 +5.5 +3.0 +3.8 +4.0 +2.9 -63- The balance of payments objective is achieved by an increase in the average tariff level ranging from +56.6% to +88.0%, wiich results in an average effective protection level and in a corresponding export subsidy varying between 50.0% and 61.8%. The structure of foreign trade shifts in accordance with the variations in the tariff-cum-subsidy scheme, with the ratio of imports to domestic output varying between -3.5% and -6.8%, and the ratio of exports to domestic output ranging from +3.0% to +5.5%. The contraction of GDP following the adjustment to lower inflows of foreign savings varies only slightly across experiments (from -1.5% to -2.4%), and the corresponding reduction in investment, government consumption and private consumption is also remarkably similar, leaving the structure of expenditure relatively unchanged across experiments. (iii) Tariff-cum-Subsidy Scheme with Constant Real Investment The base experiment was carried out under the hypothesis that the level of investment adjustments to available savings. In the following experiment, we investigate the impact on the economy of a policy in which the adjustment to the new balance of payments target is realized through changes in tariffs and the i-ntroduction of an export subsidy while mainitaining real investment constant, By comparison with the base equipment, the adjustment to the new balance of payment target is realized through similar changes in tariffs and a similar import subsidy level. The average effective protection level, and the corresponding export subsidy level, is estimated at 58.9% compared to 57.8% under the base experiment. By contrast, the contraction of GDP is slightly less pronounced under this experiment (-1,7%) compared to the base run (-2.1%). This is the direct result of the policy option retained in this experiment, as maintaining real investment permits to avoid the drop in the -6 4- construction sector obtained under the base experiment (+0.3% vs. -7.5%). In turn, the contraction of the services sector is more pronounced under the present experiment (-4.8% vs. -3.6%), as a direct result of the very sharp decline in government consumption (-33.7%) required to generate the public savings needed to sustain the level of public investment. The increase in overall government revenue under the present experiment is similar to the one obtained in the base run (+21.2% vs. + 19.7%), and the composition of government revenue remains practically unchanged. By contrast, the structure of government expenditure is fundamentally changed. Government investment, which represented 27.3% of overall government expenditure under the base run, now increases its share to 37.9%. This implies a sharp increase in government savings, from CFAF 161 billion to CFAF 226 billion, which is obtained at the expense of a very sharp cut in goverrnment consumption, who.se share in overall government expenditure declines from 37.1% to 20.9% by comparison with 31.0% under the base run. The trade orientation of the econcmy is also quite similar to the one obtained under the base run, with the same expansion of exports and the same contraction of imports obtained in both cases, while exports of industrial products expand slightly more under the present experiment (+12.3% vs. 11.4%)e -6 5- CHAPTER 5: QUANTITATIVE IMPORT RESTRICTIONS In this chapter, we describe in section 1 the evolution of the system of quantitative import restrictions in the Ivory Coast over the last decade. In section 2, we briefly discuss the economic mechanisms involved. Finally, we preserit in section 3 two experiments designed to estimate the impact of the removal of quantitative import restrictions and of their replacement by tariffs in a situation allowing for adjustments in the exchange rate. 1. Background I! Since the beginning of the sixties, Ivorian authorities have had increasing recourse to various non-tariff protection measures, including import licensing, import prohibition, and prior import agreements. Import licensing and import prohibition were introduced on an ad-hoc basis in the course of the sixties. In 1973, there were 86 industrial product.s! under import licensing, and 8 others under import prohibition. The first comprehensive list (established by Decree in 1975.) contained 310 industrial products, of which 23 were subject to import prohibition. The principal sectors affected were food products, adhesives, textiles and clothing, footwear, and certain products of the mechanical and electrical industries. Products under import prohibition included coffee beans (except arabia), roasted coffee, jute bags and electrical batteries. By January 1982, the number of industrial products under import licensing had grown to 426. Using 1981 data as a reference, imports of industrial products under licensing 1/ See Noel [1982]. 2/ Industrial products have been defined as the 6-digit customs code items under the 1973 Customs Nomenclature belonging to branches 6 through 21 under the National Accounts Nomenclature. -66- represented CFAF 233.6 billion, or 38.4% of total imports in that same year. Products under import licensing are in principle subject to a quota by product and a subquota by importer, and the licenses issued until such time as the quota is filled. In practice, product quotas are sometimes fixed at a level higher than actual import requirements. Also, once an individual importer uses up his subquota, an agreement can be reached in some cases with the Ministry of Commerce to revise the subquota upward and provide new licenses beyond the initial subquota. Parallel to the import licensing system, the Ministry of Commerce issues a list of products whose importation is subject to prior import agreement. For the products belonging to that list, "import intentions" are submitted to special import control commissions whose decisions have executive force, but may be opposed by the Ministry of Commerce. In January 1982, the number of industrial products subject to prior import agreement amounted to 140. Taking 1981 data as a reference, imports of industrial products under prior agreement amounted to CFAF 53.2 billion, or 8.8% of total imports in that same year. The impact of quantitative import restrictions on nominal protection to local manufacturing operations has been studied through the estimation of nominal Protection coefficients (NPC) calculated on the basis of firm-level data for coffee and cocoa pro;essing, textiles and clothing, and chemicals and plastics subsectors in 1980 1/ Among the subsectors under study, the impact of the restrictions was by far the highest in textiles and clothing, while an average NPC on domestic sales of 1.72, or 60 percentage points higher than the corresponding nominal tariff protection coefficient. In the food processing 1/ See Noel [1982]. -67- subsector, a high distortion between the NPC e,timate and nominal tariff protection was found in the case of coffee processing. By contrast, processed cocoa products remained unprotected. However, these findings were not sufficient to infer an estimate of the wedge due to quantitative import restrictions in this subsector. Finally, distortions remained quite low in the chemicals and plastics sub3ector. 2. Economic Mechanisms The imposition of quantitative import restrictions generates two types of effects. First, it raises protection for local manufacturing operations beyond the protection granted by tariffs. Second, it may result in substantial modifications in the distribution of income. The first effect generally reflects the objective pursued by governments faced with pressures for rapid action to safeguard domestic manufacturing operations from foreign competition. However, goveranents do not always recognize the costs involved in protecting local industry through quantitative import restrictions. First, quotas are generally imposed on an ad-hoc basis, resulting in substantial distortions in incentives and therefore in the allocation of resources in the econony. Second, the administration of the quotas system is itself costly. Third, parallel markets for import licenses may develop, generating costs for private finns obliged to divert real resources in order to obtain the licenses. Fourth, although the imposition of quotas is generally presented as temporary, it generally becomes very difficult to phase-out such a system, in particular due to the resulting rents of situation. The imposition of a quota system generally results in the private appropriation of the revenue that would have been generated for the government by its tariff equivalent. This is the incane distribution effect of -6 8 - quantitative import restrictions. In the case of the Ivory Coast, available evidence shows that this effect was quite substantial in the case of textile products. The wedge between domestic prices and world prices plus tariffs resulting from the imposition of quotas in the textiles sector generated a rent was estimated at CFAF 24 billion in 1980, or about 5% of government revenue in that same year. 3. Analysis of the Experiments In this section, we analyze the impact of the removal of quotas in a situation where the exchange rate is allowed to adjust in response to a tightening of the flow of foreign savings. Two alternative experiments have been carried out in this framework. The first experiment simulates the impact on the economy of a reduction in the balance of payments deficit by one-half when quotas in the textiles sector are replaced by import surcharges so as to yield the same level of imports as in the situation with quotas. The second experiment simulates this impact when quotas are removed in the textiles sector without compensating import surcharges. Under the first experiment, the real devaluation required to achieve the balance of payments target is estimated at 10.9%, slightly above the level obtained under the base experiment. The tariff surcharge needed to obtain the same protective effect as the quota in the textiles sector is estimated at 34.6%, yielding an increase in the overall tariff rate from 35.6% to 70.2% in that sector. This results in a 6.0% increase in overall tariff revenues for the government, from CFAF 158.6 billion to CFAF 167.9 billion. This effect combines with the increase in stabilization fund revenues induced by the devaluation to yield a 7.0% increase in government revenue. As a result, the contraction in overall investment is significantly less pronounced than under the base case (-4.6% vs. -6.6%), and construction sector output declines only -69- Table 5. 1 Summary of the Experiments (Percentage deviation from reference year) Replacement of Base Quotas by Removal of Experiment Import Surcharges Quotas Real Devaluation +9.5 +10.9 +10.7 Real GDP Agriculture -0.1 -0.1 -0.1 Industry -0.9 -0.3 0.0 Construction -6.0 -3e9 -3.5 Services -1.9 -1.3 -0.8 Real GDP at Market Prices -2.0 -1.4 -0.9 Investment -6.6 -4.6 -4.2 Government Consumption -2.6 -2.7 -2.6 Private Consumption -4.8 -4.8 -4.0 Total Output -1.8 -1.1 -0.8 Industrial Exports (in $) +5.0 +6.4 +6.8 Total Exports (in $) +1.3 +1.8 +2.0 Total Imports (in $) -6.9 -6.3 -5.8 Composite Goods Prices +2.1 +2.2 +2.0 Value-Added Prices -0.9 -0.6 -1.0 Capital Goods Prices +3.3 +3.9 +3.8 Import Prices +8.4 +8.6 +8.2 Imports/Domestic Supply -7.2 -7.2 -6.1 Exports/Domestic Output +3.4 +3.2 +3.1 -70- by 3.9% in this experiment compared to a decline of 6.0% under the base experiment. This leads to a relatively lower decline in GDP (-1.4% against -2.0% under the base run). Under the second experiment, the real devaluation required to achieve the balance of payments target is estimated at 10.7%, which is practically identical to the result obtained in the previous case. However, the resulting contraction of GDPs significantly less pronounced by comparison with the base experiment (-0.9% vs. -2.0%), as real investment is reduced by 4.2% as compared to 6.6% under the base case. This results in a slightly lower reduction in the volume of private consumption (-4.0% vs. -4.8%). At the same time, industrial exports expand more rapidly by comparison with the base experiment (+6.8% vs. + 5.0%), while the decline in imports is less pronounced (-5.8% vs. -6.9%). This results directly from the removal of quotas in the textile sector, which triggers a sharp increase in the ratio of imports to domestic output in the sector (+26.7%), or about 24 percentage points higher than under the base experiment. At the same time, exports of textiles expand more rapidly than under the base case (+8.0% vs. + 5.4%). The two experiments reveal the impact of the quotas on the distribution of income in the economy. Under the first experiment, tariff receipts by the government increase by 5.9%, from CFAF 159 billion to CFAF 168 billion, by comparison with an increase of only 2.5% under the base experiment. At the same time, urban high incomes are reduced by 8.5%, from CFAF 304 billion to CFAF 278 billion, compared to CFAF 288 billion under the base experiment. A similar pattern of adjustment is observed under the second experiment. In particular, the increase in tariff receipts is of similar magnitude, as the absence of import surcharges is offset by the increase in -71- imports following the removal of quotas in the textiles sector. The removal of quotas in conjunction with the exchange rate adjustment therefore appears to be beneficial for the economy as a whole, although it would imply dramatic shifts in the structure of production in the sector where quotas are removed, away from production lines which are not competitive with imports towards production for exports. An important result of this experiment is the significantly lower contraction of GDP and of investment following the removal of quotas. This is explained by the higher real devaluation required to achieve the balance of payments target. This, in turn, is a direct consequence of the removal of quantitative restrictions, as this generates in the first round a surge in the imports of the previously rationed goods and therefore a larger balance of payments deficit which calls for a larger adjustment of the exchange rate. This results in an increase in stabilization fund revenues (+36.6% against +32.4% in the base experiment) and in tariff revenues (+6.0% against +2.5%). Government savings increase by 38.3%, so that overall investment remains practically constant in nominal terms. The drop in real investment (-4.2%) therefore results almost entirely from the increase in capital goods prices (+3.8%), in sharp contrast with the base experiment where overall investment contracted by an estimated 6% in real terms after the exchange rate adjustment. -72- Conclusion After presenting an overview of the Ivorian economy, we have examined the recent economic developments that have led to the present financial difficulties and to two consecutive years of recession, and we have discussed the main constraints bearing on the future growth of the Ivorian economy. Among those constraints, the distortions in the system of incentives have progressively hampered the growth potential of the country and have reduced its capacity to adjust to rapidly changing external circumstances. The acceleration in the rate of domestic inflation after 1975, and the sharp deterioration in the terms of trade between 1978 and 1982 led to a substantial overvaluation of the CFA Franc vis-a-vis the currencies of Ivory Coast's main trading partners. Although major agricultural products withstood the overvaluation, thanks to the margins provided by their considerable comparative advantage, this was not the case in the industrial sector, where these margins are much narrower. In particular, the potential of the industrial sector has been affected by the strong discrimination of the system of industrial incentives against exports, and by widespread distortions in the system of tariff and non-tariff protection between the different branches of activity and, within them, between the various stages of product processing. This situation is particularly preoccupying, as the most obvious opportunities for exports within the protected CEAO markets have already been exploited, and as the Ivory Coast has to turn to non-regional markets to restore the growth of its industrial base and to generate the foreign exchange earnings required to stabilize its current account situation. In this context, the policy simulations conducted in the framework of the Ivory Coast CGE model take on a particular significance, as they provide a strong basis for assessing the macroeconomic and sectoral impact of -73- major reforms envisioned by the Government in the framework of its structural adjustment program. Although the arrangements presently in force within the Franc Zone prohibit the implementation of an independent exchange rate policy in the Ivory Coast, the experiments carried out under conditions where the exchange rate is allowed to adjust provide key insights into the magnitude of the adjustments required in the economy following a tightening of its foreign exchange constraint. In the short run, the devaluation does not induce too large a decline in overall GDP, as the impact of the reduction in the net inflow of foreign savings is compensated to a large extent by the sharp increase in Stabilization Fund revenues and, to a lesser extent, by the increase in tariff revenues. By contrast, the devaluation has a strong impact on the structure of the economy. Driven by a spur of exports, output expands in sectors like food processing, textiles and construction materials, xhile it contracts sharply in sectors linked with investment (engineering industries, construction) and in less tradeable sectors (utilities, services). Overall, the devaluation induces a sharp change in the trade orientation of the economy, with substantial increase in exports, especially of industrial goods, and reductions in imports that permit to attain the required adjustment of the overall balance. The second set of experiments provide useful guidelines for the design of a tariff-cum-export subsidy scheme as a second best measure to achieve the required adjustment of the overall balance. As in the case of the flexible exchange rate experiments, the CGE model permits to trace the effects of the proposed measures throughout the economy. The experiments show that the structure of net prices is substantially modified as a result of the change in the nominal tariff structure imposed by the equalization of -74- effective tariff protection rates across sectors. Another interesting finding is that the implementation of the tariff-cum-subsidy scheme results in a positive fiscal impact in the short-run. The third set of experiments show that the contractionary impact of a policy of adjustment with an exchange rate change can be reduced to a certain extent by implementing this policy in conjunction with the removal of quotas. By contrast, the replacement of quotas by import surcharges does not substantially affect the overall growth performance of the econamy. It does, however, affect the adjustment required in the corresponding sector. In particular, a policy of complete removal of quotas without corresponding import surcharges involves dramatic compositional changes in the textile sector, with sharp increases in both exports and imports. Ultimately, the analysis of the policies envisaged in the above experiments should be carried out from a medium to long-term perspective in the framework of a dynamic general equilibrium model. Indeed, the timing of the supply and demand responses in the economy, as well as the impact of the various policies on investment, are rich in dynamic implications. However, the investigation of the impact of these policies in a static general equilibrium framework presents new and important insights for the policy makers faced with the need to design far-reaching reforms in a complex system of interdependent economic relationships. -75- ANNEX 1 Equations of the Model 1. Price Equations (for a given sector i). I' Import price - if the quota for the sector is not binding: PM = PW (1 + tm) ER (1) - if the quota for the sector is biLding: PM = g1 (M) (2) where g is the import demand functUon (see below) and M the quota amount, and: t r = -M - t d tm PW.ER where tr is the "trade rent rate". Export price - if the export price for producers is not set by the governiment through the Stabilization Fund: PE = NE (1 + te) ER (4) 1/ Subscripts will be omitted throughout the Annex wherever no confusion is possible. -76- where PWE is the exogenous world price and the subsidy on f.o.b. price (if any). - if the export price for producers is set by the government through the Stabilization Fund: PE = PE (5) where PE is the fixed price, and: twe PE - 1 - te (6) PWE.ER where twe is the margin accruing to (paid by) the Stabilization Fund. Composite price p = PD + PM. M/D (7) h (M/IjT, 1) where PD is the domestic price and h (.,.) is the CES trade aggregation function for aggregating domestic and imported goods M and D (see chapter 2.2). Value added prices define value-added prices on domestic production and exports, as, respectively: PND = PDi (1 - tdi) E- aii P (8) PNEi = PEi (1 - tdi) - aj Pi -77- where tdi the indirect tax rate on sector i. - the value added price is then PN = (1 - ED/XD).PND + (ED/XD).PNE.(l + te') (9) where te' is the subsidy rate on exported value-added. Normalization rule LQ PD. = 1 (10) ii where Qi are predetermined weights for the price index. 2. Production (for a given sector i) - the production level is given by X = f (K,L) (11) where f is a Cobb-Douglas production function. K is the fixed sectoral capital stocks and L is sectoral employment. - sectoral employment stems from the series of first order optimality conditions, which are solved to give the labor demand by sector i for each labor category k (Lki), and from the labor aggregation function: PN x Wk (12) jk i -78- L = h (L.i c...eLmi) (13) where wk is the wage rate of labor category k and h the CES labor aggregation function. - intermediate inputs are in fixed proportion of output Vj = aji Xi (14) where V.i is the demand for intermediate consumption of good j by sector i and aji the correspondent coefficient. 3. International Trade (for a given sector i) Export supply E = X. PE) (15) where p is a logistic function 1/ Import demand M = (T- P).D (16) where 6 and a are, respectively, the share parameter and the elasticity of substitution of the trade aggregation function, and D the total domestic demand for the good i. 1/ The general form of p is p(x) - a + c. 1 + exp [b(x - x)] -79- 4. Income Household income - overall labor income is given by: RE i. i Wk Lki (17) and is distributed among household categories according to employment structure and to the mapping between labor and household categories - overall, non-labor income is R i (PN..X. - wkLki) + Sf F.ER + TR (19) where F is net capital inflow, sF the share of F accruing to households and TR government transfers. Nonlabor factor income is distributed among household categories according to the exogenous structure of capital ownership. Capital inlL,'!,ow is allocated following given fixed shares. Government income RG itmiAiER Mi + EitdiPDiX ikthk«c + ( sF) F.ER + ETtwe1 [PWE.(l + tei)ER-PEiI.EDi t - EVE ER E - te PNE E - TR (20) ie i i i 'i i -80- where Rk is the income tax rate for the household category k, and Rk the total income of that category. The eight elements account for, respectively, tariff, indirect tax and income tax revenue, direct capital inflows, Stabilization Fund surplus, L.o.b. and value-added export subsidies and, finally transfers to households. 5. Final Demand Corsumption - hQusehold consumption demand of good i is, for household category k: Cik = aik (1 k k P (21) where aik is a constant expenditure share and sk the savings rate of household category k. - government consumption is derived in a similar way: cGig =jg (1 - Rg) R (22) Investment and inventories - investment by sector of destination is: I =Pb IPb + 4Pv IPv (23) i i i -81- where .Pb and 0Pv are respectively the exogenous sectoral shares for 1 i public (central government plus public enterprises) and private investment, and IPb and IPv the amounts of public and private investment investment volume by sector of origin is then: I. Z. Z, Z.jP (24) 1 1 1] P. where Zi; are the capital composition coefficients - inventories are merely a constant share of output: I I Z Z i X (25) 6. Equilibrum Conditions Product markets The equilibrium condition for a given sector i is: i id i kik ' Cig + 'j Vij + Zi + Z + E (26) where di is the "domestic use" ratio: d1 1/h(Mi/Di,1) (27) where h is the CES trade aggregation function. - the equilibrating variable can be either the price or the export volume whenever domestic prices are fixed (see chapter 2.3). -82- Investment savings adequation The equilibrium condition writes: Pb -=Pv (8 I + I = ESk k + sgRg (28) The equilibrating variable is either the public investment level or government savings according to the chosen closure rule (see chapter 2.3), reflecting the fact that private investment and savings rates have been assumed constant. Labor markets - equilibrium on a given labor market k requires: Lk - k i (29) k where Lk is the employed amount of labor of category k. In a full employment situatiorn Lk = Lk Since nominal wages are supposed constant, Lk becomes endogenous, making unemployment (Lk - L ) the equilibrating variable. Balance of payments - the equilibrium equation on the foreign exchange market is: .PWiMi - E PWEi Ei - F = 0 (30) where, according to the exchange rate iegime, either capital inflows F are exogenous and the exchange rate adjusts in order to realize the equilibrium or the exchange rate is exogenous and F is endogenous. -83- 7. Equalization of Effective Rates of Protection We use in this model a partial equilibrium concept of effective rate of protection (see Chap. 3.3). In the absence of non-tariff barriers and price control, the effective rate of protection in a given sector i is defined as c P. ~a..p e = - c (31) i P. c p i 7j l+tm J J l+tm where ti = nominal tariff rate c a.. = technical coefficient expressing direct plus indirect use of traded intermediate input j (Corden method)!/ Given a certain set of values for the sectoral effective protections, it is always possible to compute the implied set of nominal tariffs by inverting the linear system defined by equation (31).2/ It is therefore always possible to compute the set of tariffs which would equalize effective protection at a given level. The level at i.Jhich the effective rates of protection are equalized derives from balance of trade considerations. Since, other things being equal, imports are decreasing as effective protection increases, there is a direct relationship between the level of protection and the balance of trade 1/ The Corden method assumes that the elasticity of supply of non-traded goods is non-infinite, and includes indirect protection granted to non- traded goods as a result of protecting a traded sector in the measure of protection of this traded sector. The passage from the observed input- output matrix to the adjusted input-output matrix used in (31) results from these two assumptions. 2/ Note that the coefficients of the equation are fixed for a given input- output matrix and that the solution is unique. -84- deficit. Therefore, in this version of the model, the (uniform) level of effective protection is adjusted to realize equilibrium on the foreign exchange market, defined by (30). In turn, this level of protection is implemented by a set of tariffs computed as described above. Value added export subsidies are introduced and made equal to effective protection. This additional condition does not change the nature of the problem. Since exports are increasing in function of the rate of export subsidy, the combination of the tariff and of the subsidy scheme tends to reduce the trade deficit further. SOCIAL ACCOUNTING MATRIX ACTIVITI COMMODIT LABOR CAPITAL SUM (3- FARMERS RURAL UN URBAN LO URBAN MI URBAN HI EXPENDITURES ES IES 4) SKILLED W INCOME D. INCOM GH INCOM I 2 3 4 5 6 7 8 9 10 RECEIPTS ACTIVITIES I ACTIVITIES 0.00 2564.32 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 COMMODITIES 2 COMMODITIES 1450.21 0.00 0.00 0.00 0.00 353.21 37.00 179.98 531.38 249.77 FACTORS 3 LABOR 524.80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4 CAPITAL 993.61 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5 SUM (3-4) 1518.41 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 HOUSEHOLDS 6 FARMERS 0.00 0.00 0.02 463.61 463.63 0.00 0.00 0.00 0.00 0.00 7RRLUSILD0.00 0.00 43.59 0.00 43.59 0.00 0.00 0.00 0.00 0.00 n 8 URBAN LOW INCOME 0.00 0.00 168.41 0.00 168.41 0.00 0.00 0.00 0.00 0.00 9 URBAN MID. INCOME 0.00 0.00 222.52 322.29 544.81 0.00 0.00 0.00 0.00 0.00 10 URBAN HIGH INCOME 0.00 0.00 90.26 207.71 297.97 0.00 0.00 0.00 0.00 0.00 11 TRADE RENTS 0.00 23.75 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 12 STAB FUND 79.38 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 13 TREASURY 249.39 158.55 0.00 0.00 0.00 32.26 3.03 14.76 48.54 23.91 14 DEBT MANAGEMENT BOAR 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 15 PUB. ENT. CURRENT Fl 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 16 SUM (6-15) 328.77 182.30 524.80 993.61 1518.41 32.26 3.03 14.76 48.54 23.91 17 HOUSEHOLDS CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 18 CENT. GOVT. CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 19 PUB. ENT. CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 20 NON-OIL PRIV. ENT. C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 21 OIL PRI. ENT. CAPITA 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22 SAVINGS 0.00 0.00 0.00 0.00 0.00 40.32. 0.00 0.00 60.66 41.83 23 REST WORLD CURRENT 0.00 870.28 0.00 0.00 0.00 63.84 6.00 29.20 96.05 47.31 24 REST WORLD CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 25 TOTAL 3297.39 3616.90 524.80 993.61 1518.41 489.65 46.03 223.93 736.63 362.82 SOCIAL ACCOUNTING MATRIX(CONTINUED) --------------------- -------------------------------------------------------------------------------------------------------------- TRADE RE STAB FUN TREASURY DEBT MAN PUB. ENT SUM (6-1 HOUSEHOL CENT. GO PUB. ENT NON-OIL EXPENDITURES NTS D AGEMENT . CURREN 5) DS CAPIT VT. CAPI . CAPITA PRIV. EN lt 12 13 14 15 16 17 18 19 20 RECEIPTS ACTIVITIES I ACTIVITIES 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 COMMODITIES 2 COMMODITIES 0.00 0.00 182.67 0.00 0.00 1534.01 30.00 257.40 177.40 .155.78 FACTORS 3 LABOR 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 4 CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 5 SUM (3-4) 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 HOUSEHOLDS 6 FARMERS 0.00 0.00 17.07 0.00 0.00 17.07 0.00 000 000 0.00 7 RURAL UNSKILLED 0.00 0.00 1.60 0.00 0.00 1.60 0.00 0.00 0.00 0.00 8 URBAN LOW INCOME 0.00 0.00 51.43 0.00 0.00 51.43 0.00 0.00 0.00 0.00 9 URBAN MID. INCOME 0.00 0.00 178.35 0.00 0.00 178.35 0.00 0.00 0.00 0.00 10 URBAN HIGH INCOME 23.75 0.00 34.46 0.00 0.00 58.21 0.00 0.00 0.00 0.00 11 TRADE RENTS 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 12 STAB FUND 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 13 TREASURY 0.00 79.38 8.00 0.00 0.00 209.88 0.00 0.00 0.00 0.00 14 DEBT MANAGEMENT BOAR 0.00 0.00 48.80 0.00 0.00 48.80 0.00 0.00 0.00 0.00 15 PUB. ENT. CURRENT Fl 0.00 0.00 38.10 0.00 0.00 38.10 0.00 0.00 0.00 0.00 16 SUM (6-15) 23.75 79.38 377.80 0.00 0.00 603.43 0.00 0.00 O.O0 0.00 17 HOUSEHOLDS CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 18 CENT. GOVT. CAPITAL 0.00 0.00 70.35 0.00 0.00 70.35 0.00 0.00 0.00 0.00 19 PUB. ENT. CAPITAL 0.00 0.00 0.00 0.00 20.70 20.70 0.00 0.00 0.00 0.00 20 NON-OIL PRIV. ENT. C 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 21 OIL PRI. ENT. CAPITA 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 22 SAVINGS 0.00 0.00 0.00 0.00 0.00 142.82 0.00 0.00 0.00 0.00 23 REST WORLD CURRENT 0.00 0.00 32.00 48.80 17.40 340.60 0.00 0.00 0.00 0.00 24 REST WORLD CAPITAL 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 25 TOTAL 23.75 79.38 662.82 48.80 38.10 2711.92 30.00 257.40 177.40 155.78 --------------------~---------------------------------------------------------------------------------------------------------------- SOCIAL ACCOUNTING MATRIX(CONTINUED) OIL PRI. SAVINGS REST WOR REST WOR TOTAL EXPENDITURES ENT. CA LD CURRE LD CAPIT 21 22 23 24 25 RECEIPTS ACTIVITIES I ACTIVITIES 0.00 0.00 733.07 0.00 3297.39 COMMODITIES 2 COMMODITIES 12.10 0.00 0.00 0.00 3616.90 FACTORS 3 LABOR 0.00 0.00 0.00 0.00 524.80 4 CAPITAL 0.00 0.00 0.00 0.00 993.61 5 SUM (3-4) 0.00 0.00 0.00 0.00 1518.41 HOUSEHOLDS b FARMERS 0.00 0.00 8.96 0.00 489.65 7 RURAL UNSKILLED 0.00 0.00 0.84 0.00 46.03 X 8 URBAN LOW INCOME 0.00 0.00 4.10 0.00 223.93 9 URBAN MID. INCOME 0.00 0.00 13.47 0.00 736.63 10 URBAN HIGH INCOME 0.00 0.00 6.64 0.00 362.82 II TRADE RENTS 0.00 0.00 0.00 0.00 23.75 12 STAB FUND 0.0¢ 0.00 0.00 0.00 79.38 13 TREASURY 0.00 0.00 45.00 0.00 662.82 14 DEBT MANAGEMENT BOAR 0.00 0.00 0.00 0.00 48.80 15 PUB. ENT. CURRENT Fl 0.00 0.00 0.00 0.00 38.10 16 SUM (6-15) 0.00 0.00 79.00 0.30 2711.92 17 HOUSEHOLDS CAPITAL 0.00 30.00 0.00 0.00 30.00 18 CENT. GOVT. CAPITAL 0.00 29.95 0.00 157.10 257.40 19 PUB, ENT. CAPITAL 0.00 109.10 0.00 47.60 177.40 20 NON-OIL PRIV. ENT. C 0.00 115.28 0.00 40.50 155.78 21 OIL PRI. ENT. CAPITA 0.00 3.00 0.00 9.10 12.10 22 SAVINGS 0.00 0.00 0.00 144.50 287.32 23 REST WORLD CURRENT 0.00 0.00 0.00 0.00 1210.88 24 REST WORLD CAPITAL 0.00 0.00 398.80 0.00 398.80 25 TOTAL 12.10 287.32 1210.88 398.80 13674.30 PERCENTAGE COMPOSITION OF BASE DATA - 1980 ------------------------------------------ GROSS DOMESTIC VALUE SECTOR OUTPUT EXPORTS IMPORTS DEMAND ADDED LABOR I SUBSISTENCE CROP 10.37 0.12 6.06 15.53 15.84 28.28 2 COFFEE 2.23 8.82 0.00 0.57 3.60 17.24 3 COCOA 3.71 14.72 0.00 0.20 6.02 28.73 4 COTTON 0.36 0.00 0.00 0.00 0.58 2.75 5 OTHER AG 3.23 5.18 '0.08 1.59 3.54 14.04 6 FORESTRY 3.39 14.58 0.00 0.00 3.55 0.21 7 MINING PETROL 4.75 9.56 16.06 0.95 2.16 0.29 8 FOOD FPROCESSING 7.38 10.14 8.00 9.37 3.90 0.78 9 TEXTIL ES 4.12 3.98 6.89 5.42 3.48 0.60 10 CHEMICALS 2.06 1.42 8.42 1.50 1.28 0.18 11 CONSTRUCTION MATERIA 2.40 3.72 2.71 0.13 2.14 0.49 12 ENGINEERING INDUSTRY 4.86 2.75 34.37 4.79 2.99 0.53 13 UTILITIES 1.68 0.12 0.00 1.46 2.22 0.24 14 CONSTRUCTION 15.90 0.11 0.00 26.93 10.92 1.63 15 SERVICES 33.57 24.78 17.41 31.56 37.78 4.02 ALL SECTORS 100.00 100.00 100.00 100.00 10O.00 100.00 --------------------------------------------------------------------------------------- TRADE AND PRODUCTION PARAMETERS - 1980 SECTOR SKAP SEX RMD AMPT TM RHOC RHOD I SUBSISTENCE CROP 0.31 0.24 19.50 22.51 15.31 0.80 0.80 2 COFFEE 3.08 82.92 0.00 43.64 0.00 0.90 0.80 3 COCOA 3.06 83.12 0.00 59.66 0.00 0.90 0.80 4 COTTON 3.09 0.00 0.00 60.78 0.00 0.90 0.80 5 OTHER AG 7.31 33.59 1.30 22.88 0.00 0.90 0.80 6 FORESTRY 23.30 90.04 0.00 34.29 0.00 0.90 0.20 7 MINING PETROL 70.07 42.13 194.65 60.99 21.90 0.90 0.20 8 FOOD PROCESSING 36.76" 28.73 50.67 21,74 28.20 0.60 0.80 9 TEXTILES 20.55 20.22 69.88 32.92 35.56 1.10 0.80 10 CHEMICALS 26.41. 14.47 159.35 46.73 22.19 1.10 0.80 11 CONSTRUCTION MATERIA 39.71 * 32.46 55.81 26.76 23.16 1.10 0.80 12 ENGINEERING INDUSTRY 25.85 11.82 267.10 51.51 21.41 0.60 0.80 13 UTILITIES 59.29 1.55 0.00 43.43 0.00 0.90 0.80 14 CONSTRUCTION "2.42 0.14 0.00 35.81 0.00 0.90 0.80 15 SERVICES .. 10 15.45 20.45 30.28 0.00 0.60 0.80 -----------------------------------------~--------.--------- ------,-----------.----------------- EXPLANATION OF COLUMNS -------~--------------- SKAP = CAPITAL SHARE IN VALUE ADDED (%) SEX = EXPORTS/GROSS OUTPUT (%) RMD * IMPORT/DOMESTIC SUPPLY (%) AMPT * IMPORTED/TOTAL INTERMEDIATES (Y) TM TARIFF RATE (%) RHOC * TRADE SUBSTITUTION ELASTICITY RHOD = CAPITAL-LABOR SUBSTITUTION ELASTICITY 1980 INPUT OUTPUT COEFFICIENTS ------------------------------ SUBSISTE COFFEE COCOA COTTON OTHER AG FORESTRY MINING P FOOD PRO TE,XTILES CHEMICAL COLUMN 1 2 3 4 5 6 7 8 9 10 ROW i SUBSISTENCE CROP 0.1220 0.0000 0.0000 0.0000 0.0041 0.0000 0.0000 0.2451 0.0082 0.0000 2 COFFEE 0.0000 0.0000 0.0000 0.0000 0.0002 0.0000 0.0000 0.0127 0.0000 0.0000 3 COCOA 0.0000 0.0000 0.0000 0.0000 0.0002 0.0000 0.0000 0.0713 0.0000 0.0000 4 COTTON 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0862 0.0000 5 OTHER AG 0.0011 0.0000 0.0000 0.0000 0.1815 0.0000 0.0000 0.0893 0.0000 0.0515 6 FORESTRY 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0012 7 MINING PETROL 0.0048 0.0069 0.0352 0.0061 0.0318 0.0519 0.6503 0.0209 0.0186 0.0706 8 FOOD PROCESSING 0.0099 0.0000 0.0000 0.0000 0.0012 0.0000 0.0000 0.0699 0.0028 0.0078 9 TEXTILES 0.0025 0.0004 0.0002 0.0005 0.0004 0.0000 0.0000 0.0051 0.1954 0.0044 10 CHEMICALS 0.0031 0.0480 0.0220 0.0949 0.0436 0.0207 0.0103 0.0138 0.0733 0.2977 11 CONSTRUCTION MATERIA 0.0000 0.0000 0.0000 0.0000 0.0023 0.0004 0.0000 0.0017 0.0005 0.0040 12 ENGINEERING INDUSTRY 0.0071 0.0152 0.0347 0.0042 0.0306 0.0701 0.0123 0.0657 0.0258 0.0652 13 UTILITIES 0.0001 0.0053 0.0022 0.0000 0.0035 0.0012 0.0012 0.0136 0.0211 0.0132 14 CONSTRUCTION 0.0002 0.0158 0.0000 0.0000 0.0014 0.0004 0.0012 0.0026 0.0011 0.0019 15 SERVICES 0.0104 0.0199 O.O37 0.0031 0.0970 0.2805 0.0751 0.0983 0.1029 0.1401 16 SUM 0.1612 0.1115 0.1080 0.1088 0.3977 0.4252 0.7505 0.7100 0.5359 0.6577 '0 CONSTRUC ENGINEER UTILITIE CONSTRUC SERVICES COLUMN 11 12 13 14 15 ROW I SUBSISTENCE CROP 0.0000 0.0000 0.0000 0.0002 0.0000 2 COFFEE 0.0000 0.0000 0.0000 0.0000 0.0000 3 COCOA 0.0000 0.0000 0.0000 0.0000 0.0000 4 COTTON 0.0000 0.0000 0.0000 0.0000 0.0000 5 OTHER AG 0.0000 0.0000 0.0000 0.0004 0.0000 6 FORESTRY 0.1397 0.0000 0.0000 0.0000 0.0000 7 MINING PETROL 0.0465 0.0191 0.0711 0.0469 0.0504 8 FOOD PROCESSING 0.0000 0.0000 0.0000 0.0000 0.0074 9 TEXTILES 0.0013 0.0023 0.0000 0.0001 0.0032 10 CHEMICALS 0.0114 0.0317 0.0163 0.0385 0.0085 11 CONSTRUCTION MATERIA 0.1655 0.0360 0.0012 0.1140 0.0003 12 ENGINEERING INDUSTRY 0.0330 0.3580 0.0627 0.1362 0.0449 13 UTILITIES 0.0223 0.0095 0.0114 0.0047 0.0143 14 CONSTRUCTION 0.0024 0.0033 0.0162 0.1158 0.0146 15 SERVICES 0.0883 0.2019 0.0958 0.1657 0.2381 16 SUM 0.5102 0.6618 0.2746 0.6226 0.3816 -90- CONSUMPTION SHARES BY HOUSEHOLD GROUPS (PERCENT) SECTOR 1 2 3 4 5 1 SUBSISTENCE CROP 32.28 35.51 8.89 17.54 22.93 2 COFFEE 0.00 0.00 0.00 0.00 0.00 3 COCOA 0.00 0.00 0.00 0.00 0.00 4 COTTON 0.00 0.00 0.00 0.00 0.00 5 OTHER AG 2.07 2.17 1.23 1.64 1.74 6 FORESTRY 0.00 0.00 0.00 0.00 0.00 7 MINING PETROL 0.83 0.43 3.25 2.70 1.36 8 FOOD PROCESSING 26.93 30.54 5.10 12.23 17.27 9 TEXTILES 10.44 9.52 21.39 7.78 8.17 10 CHEMICALS 1.41 1.23 6.16 4.48 5.90 11 CONSTRUCTION MATERIA 0.25 0.23 0.35 0.35 0.28 12 ENGINEERING INDUSTRY 4.35 2.82 5.35 9.03 3.60 13 UTILITIES 0.03 0.06 0.93 1.69 1.85 14 CONSTRUCTION 0.05 0.07 0.16 0.07 0.03 15 SERVICES 21.37 17.43 47.19 42.49 36.87 ALL SECTORS 100.00 100.00 10O.00 100.00 100.00 ------------------------ HOUSEHOLD GROUPS (COLUMNS) 1 FARMERS 2 RURAL UNSKILLED 3 URBAN LOW INCOME 4 URBAN MID. INCOME 5 URBAN HlGH INCOME -91- ANNEX 3 Sectoral Tradeability Indicators Ratio of Share of Imports to Ratio of Exports in Domestic imports to Trade substi- Domestic Production intermediate tution Classif{ Sector Production minus Exports inputs elasticities cation- (in %) (in% ) 1. Food crops 0.24 16.91 21.96 .6 - M1s 2. Coffee 82.92 0.00 44.03 - EX ils 3. Cocoa 83.11 0.00 60.21 - EX 1 4. Cotton 0.00 0.00 61.26 - - hiS 5. Other agric. 33.59 1.30 20.67 .6 EX NS 6. Forestry 90.04 0.0U 34.62 - EX MS 7. Mining & petroleum 42.14 199.98 61.9U .8 LX US 8. Food process. 28.74 54.42 22.12 .6 LX MS 9. Textiles 20.19 67.03 29.31 .8 EX MS 10. Chemicals 14.47 238.87 47.61 .8 EX SX 11. Construction materials 32.45 54.12 26.73 .8 EX MS 12. Engineering industries 11.82 269.76 51.75 .4 EX kc 13. Utilities 1.55 0.00 43.93 - - NT 14. Construction 0.14 U.00 35.98 - - NT 15. Services 15.45 20.45 30.52 .4 EX MIC Source: Michel and Noel L1983J. I/ EX = exportable MS = import substitute MC = import complement NT - non-tradeable -92- References Adelman, I., S. Robinson [1978]: "Income Distribution Policies in Developing Countries." Stanford: Stanford University Press. Amaranand, P., W. Grais [1983]: Macroeconomic and Distribution Implfcations of Sectoral Policy Interventions: An Application to Thailand. World Bank Staff Working Paper No. 627. Armington, P. [1969]: "A Theory of Demand for Products Distinguished by Place of Production." IMF Staff Papers, Vol. 16, pp 159-178. Bhattacharya, A., W. Grais, B. Pleskovic [1983]: "Macroeconomic Implications of Price Reforms in Egypt." World Bank, mimeo. Balassa, B. and Associates [1971]: The Structure of Protection in Developirig Countries. Baltimore: The Johns Hopkins University Press. Bell, C.; T.N. Srinivasan [1983]: "On the Uses and Abuses of Economy-Wide Models in Development Policy Analysis." World Bank, Development Research Department Discussion Paper No. 55. Caisse Centrale de Cooperation Economique and World Bank [1982]: "Rapport sur les Entreprises Publiques en Cote d'Ivoire." World Bank, mimeo. Corden, M. [1971]: The Theory of Effective Protection. London: Oxford University Press. Dervis, K., J. de Melo, S. Robinson [1982]: "A General Equilibrium Analysis of the Causes of a Foreign Exchange Crisis: The Case of Turkey." Weltwirtschaftliches Archiv, Vol. 118(2), pp. 259-280. Dewatripont, M., G. Michel [1983]: "On Closure Rules and Microeconomic Foundations in Applied General Equilibrium Models." World Bank, mimeo. Fargues, Ph. [1981]: "Les Migrations en Cote d'Ivoire." Cahiers du CIRES". Dec. 1981-March 1982 (No. 31-32). Johansen, L. [1960]: A Multi-Sectoral Study of Economic Growth. Amsterdam: North-Holland. Lewis, J., S. Urata [1983]: Turkey: Recent Economic Performances and Medium- Term Prospects, 1978-1990. World Bank Staff Working Paper No. 602. Little, I., T. Scitorsky and M. Scott [1970]: Industry and Trade in Some Developing Countries. London: Oxford University Press. Lysy, F. [1982]: "The Character of General Equilibrium Model' under Alternative Closure Rules." World Bank, mimeo. ?4ichel, G., M. Noel [1983]: "A Social Accounting Matrix for the Ivory Coast: A Technical Note." World Bank, mimeo. -93- Ministere de l'Agriculture [1981]: Recensement National de i'Agriculture. Adidjan. Ministere des Finances [1982]: Budgets Economiques. Abidjan. Noel, M. [1982]: "The Evolution of the System of Industrial Incentives in the Ivory Coast from 1970 to Present." World Baik, mimeo. Ray, A. [1973]: "Non-Traded Inputs and Effective Protection: A General Equilibrium Analysis." Journal of International Economics, Vol. 3, pp. 245-258. Taylor, L. et al [19801: Models of Growth and Distribution for Brazil. London: Oxford University Press;. den Tuinder, B.A. [1978]: Ivory Coast: The Challenge of Success. Baltimore: The Johns Hopkins University Press. World Bank [1979]: "Ivory Coast: Country Economic Memorandum." Washington, D.C. Report No. 2655-IVC. (an internal document Twith restricted circulation). World Bank [1983]: World Development Repaort 1983. New York: Oxford University Press. World Bank [1982]: Yugoslavia: Adjustment Policies and Development Perspectives. Washington, D.C. World Bank Publications of Related Interest Adjustment to Extemal Shocks Adjustment Policies and IVEW in Developing Economies Problems in Developed Countries Bela Balassa Martin Wolf Bureaucracies and the Political Staff Working Paper No. 472. 1981. 31 Staff Working Paper No. 349. 1979. 236 Economy of Protection: pages (including appendix). pages (including references). Refle-ctions of a Continental Stock No. WP 0472, $3. Stock No. WP 0349. $10. European Patrick Messerlin Analyzes three factors that influence Britain's Pattern of the "bureaus" (bureaucrats) in their Specialization in Manufactured Goods with Developing Countries and Trade Protection Vincent Cable and Ivonia Rebelo Staff Working Paper No. 425. 1980. 61 pages (including 3 appendixes). Stock No. WP 0425. $3. decisions affecting protectionism in European Community No. OX 520211, $22.50 hardcover. (A France. Protection against specially priced edition is available in India Staff Working Paper No. 568. 1983. 64 Manufactured Imports from from Oxford University Press branches.) pages. Developing Countries: A Case Industrial Country Policy and ISBN 0-8213-0198-5. Stock No. WP 0568. Study in the Political Economy Adjustment to Imports from $3. of Protection Developing Countries Capital-Importing Oil E. Verreydt and J. Waelbroeck J. M. Finger Exporters; Adjustment Issues Staff Working Paper No. 432. 1980. 25 Staff Working Paper No. 470. 1981. 22 and Policy Choices pages. pages (including references), Alan H. Gebi Stock No. WP 0432. $3. Stock No. WP 0470. $3. Staff Working Paper No. 475. 1981. 38 pages (including 9 tables). Export Promotion Policiess in Stock No. WP 0475. $3. Barend A. de Vries Italian Commercial Policies i The Developing Countries and Staff Working Paper No. 313. 1979. 80 Eno R. Grill pages. International Shipping Stock No. WP 0313. $3. Staff Working Paper No. 428. 1980. 47 Harald Hansen pages. Staff Working Paper No. 502. 1981. 151 NEW Stock No. WP 0428. $3. pages (including 12 annexes, 38 tables, bibliography). Exports of Capital Goods and Stock No. WP 0502. $5. Related Services from the NEW Republic of Korea ' NEW . Larry E. Westphal, Yung W. Rhee, Korea's Competitive Edge: Linsu Kim, Alice Amsden Managing Entry into World Economics and the Politics of Examines Korea's spectacular export Markets Protection: Some Case Studies growth-from $50 million of goods in Yung Whee Rhee, Bruce Ross- 1962 to $25 billion in 1982. Five kinds LasnadGryPuel of Industies of project-related exports are character- Larson, and Garry Pursell Vincent Cable ized (overseas construction, plant ex- How did Korea manage to expand its Looks at factors which effect an indus- ports, direct investments, consulting' exports from less than $100 million a try's attitude toward protection by services, licensing and technical agree- year in the early 1960s to more than analyzing four of Great Britain's indus- ments). Discusses the role of these ex- $20 billion a year in the early 1980s? tries: footwear, knitwear, cutlery, and ports in Korea's strategy for develop- To find out about the underpinnings consumer electronics. Case studies ex- ment. Shows how these strategies of Korea's competitive edge, the au- amine import competition from devel- conform to the country's dynamic thors asked more than 100 major Ko- oping countries and adjustment op- comparative advantage by enlarging its rean exporters what had been impor- tions exercised within each industry. industrial base. tant for them in institutional support, Study includes some explanations for Staff Working Paper No. 629. 1983. 80 in technological development, and in protectionist behavior among indus- pages. markethng overseas. The findings triesiahndadisculssion of th politics of ISBN 0-82Z3-0310-4.Stock No. WP 0629. between exporting and the effective- decisionmaking in regard to tr-ade pol- eso onr' cnmcisiu icy. OE. ness of a countryb s economic institu- Staff Working Paper No. 569. 1983. 80 On Exports and Economic effective institutions, a country mtay pages. G ro dt not be able to implement effective pol- ISBN 0-8213-0199-3.Stock No. WP 0569. Gerson Feder icies for export promotion. Conversely, $3. Staff Working Paper No. 508. 1982. 24 successful exporting appears to give pages (including appendix, references). economic institutions more vitality and Effects of Non Tanff Barrier Stock No. WP 0508. $3. effectiveness. The findings also show to Trade on Prices, how Korea's selectivity-in the acqui- Employment, and Imports: The India's Exports sition of technology and in the market- Case of the Swedish Textile Martin Wolf ing of products overseas-has been an and Cloing Industry Despite improved performance, the important part of Korea's success in Carl Hamilton growth of India's exports continues to the world marketplace. Staff Working Paper No 429 1980 63 lag behind need, potential, and the The Johns Hopkins University Press. June g P . 4 achievements of several of its competi- 1984. About 204 pages. pages (including appendix, bibliography). tors. This study examines India's over- ISBN 0-8018-3266-7. $19.95. Stock No. WP 0429. $3. all export performance in the 1960s Energy, International Trade, and 1970sr f with ncenptv The major prob- On the Political Eccnomy of and1 Econaomic Growth lems and policies are discussed, as Protection in Germany Alan S. Manne and Sehun Kim well as current strategic options. H. H. Glismann and F. D. Weiss Staff Working Paper No. 474. 1981. 30 Oxford University Press. 1982. 224 pages Staff Working Paper No. 427, 1980. 30 pages (including 2 appendixes, references). (including index), pages (including bibliography). Stock No. WP 0474. $3. LC 82-6309: ISBN 0-19-520211-2, Stock Stock No. WP 0427. $3. through which trade police is deter- developing countries grows, the faster NEW mined. Focuses on the evolution of th-se countries will become competi- The Political Economy of France's international commerce in the tive in an even wider range of goods. 1970s and assesses probable trends for Innovation, though a key factor, can- Protection in Italy: Some the future. not determine comparative advantage, Empnrical Evidence Staff Working Paper No. 570. 1983. 64 because innovations spread rapidly Enzo Grilli and Mauro La Noce pages. through the world economy. Thispecif sies the basedon a model that ISBN 0-8213-0200-0. Stock No. WP 0570. Staff Working Paper No. 571. 1983. 84 specifes the demand side of the mnar- 3.pages. ket. Examines 35 industrial subsectors S3. p g . in terms of EEC tariff protection and -NEW $3I Italy's domestic subsidy assistance. _____$3 Finds that tariffs protect Italian exports Real Wages and Exchange The Structure of Protection in in the EEC markets but have less effect Rates in the Philippines, 1956- Developing Countries in keeping out non-EEC imports. Staff Working Paper No. 567. 1983. 48 78: An Application of thie Bela Balassa and others pages. Stolper-Samuelson-Rybczynski The Johns Hopkins University Press, 1971, ISBN 0-8213-0197-7. Stock No. WP 0567 Model of Trade 394 pages (including 5 appendixes, index). $3. Deepak Lal LC 77-147366. ISBN 0-8018-1257-7, Stock Output and Employment Explains the movements of real wages No. JH 1257. $25 hardcover. Changes in a "Trade Sensitive" in the Philippines in terms of a simple Testing for Direction of Sector. Adjustment in the U.S. regression model. Examines the coun- Eixports: India's Exports of Footwear Industry try's postwar economic performnance Manufactures in the 1970s John H. Mutti and Malcolm D. and draws some tentative conclusions Ashok Khanna Bale for economic policy. Staff Working Paper No. 430. 1980. 21 Staff Working Paper No. 604. 1983. 60 Tests the hypothesis that the exports p (Icd. fotntes, ) pages. of a developing country with an ad, Stock No. WP 0430. $3. ISBN 0-8213-0213-2.Storc No. WP 0604. vanced dstinactuinsg the capital inten- Patters of Barriers to Trade in sity of exports will be greater to the '; veden: A Study in the Shadow Prices for Trade more labor abundant destinations, and X heory of Protection Strategy and Investment greater to the more capital abundant Lars Lundberg Planning in Egypt destinations. India's exports of manu- Staff Working Paper No. 494. 1981. 35 John Page, Jr. factures for 1973 and 1978 are used for pages (including 3 aN4 endixes). Staff Working Paper No. 521. 1982. 212 the analysis. Stock No. WP 0494. $3. pages. Staff Working Paper No. 538, 1983. 41 The Political Economy of ISBN 0-8213-0009-1. Stock No. WP 0521. pages. Protection in Belgium $10. ISBN 0-8213-0132-2.Stock No. WP 0538. P.K.M. Tharakan $3, Staff Working Paper No. 431. 1980. 22 Structural Change i Trade in pages (including statistical appendix, refer- Manufactured Goods beibween The Tokyo Round: Results and ences). . Industrial and Developirig Implications for Developing Stock No. WP 0431. $3, Countries Countries The Political Market for Bela Balassa Ria Kemper Protection in Industrial Staff Working Paper No. 396. 1980. 46 Staff Working Paper No. 372. 1980. 38 Countries: Empirical Evidence pages. pages (including annex). Kym Anderson and Robert E. Stock No. WP 0396. $3. Stock No. WP 0372. $3. Baldwin. Trd-dutetPlce n Staff Working Paper No. 492. 1981. 28 NEW Trade Adjustment Policies and pages (including references). Th Income Distribution in Three Stock No. WP 0492. $3. The Stmcture of Intemational Archetype Developing On Protectionism in the Competitiveness in the Federal Economies Netherlands Republic of Germany: An Jaime de Melo and Sherman K.A. Koekkoek, J. Kol, and L.B.M. Appraisal Robinson Mennes Frank D. Weiss Staff Working Paper No. 442. 1980. 91 Staff Wcr?king Paper No. 493. 1981. 70 Probes the comparative trade advan- pages (including appendixes, references). pages (including 3 annexes, references). tage of the Federal Republic of Ger- Stock No, WP 0442. $3. Stock No, WP 0493. $3, many in the 1980s with surprising con- clusions: developing countries are Trade among Developing NEW competitive with the Federal Republic Countries: Theory. Policy Public Assistance to Industries of Germany in a far wider range of Issues, and Principal Trends products than had been previously SafWrigPprN.49 91 1 and Trade Policy in France thought. Suggests probable trends, es- Staff Working Paper No. 479. 1981. 116 Bernard Babe pecially toward developing countries. pages (including 2 appendixes, references). Descibes the instdtutional structure Concludes that the faster income in Stock No. WP 0479. $5. Trade and Employment Trends in International Trade Worker Adjustment to Policies for Industrial in ManumfAured Goods and Liberalized Trade: Costs and Development Structural Change in the Assistance Policies Keith Marsden Industrial Countries Graham Glenday, Glenn P. In the last decade, the developing Bela Balassa with the assistance of Jenkins, and John C. Evans countries have proved that they can Kenneth Meyers Staff Working Paper No. 426. 1980. 87 compete internationally in exporting Exarirnes recent trends in trade in pages (including 2 appendixes, bibliog- manufactured goods, as well as pn- manufactured goods between the in- raphy). mary products and services. This pa- dustrial and the develop:-.rg countries. Stock No. WP 0426. S3. per examines three sets of issues: (a) Analyzes (a) the implications of these World Trade and Output of whether good export perforinance is trends for structural change in the in- Manufactures: Structual attributable to special characteristics of dustrial countries and (b) changes over TrnsadDvlpg the most successful countries or time in the current dollar value and Trends and Developing whether their success can be readily the commodity composition of trade in Countries' Exports replicated in other countries; (b) manufactured goods. Recommends Donald B. Keesing whether the penetration of the mar- policy changes aimed at promoting in- Staff Working Paper No. 316. 1979. kets of indutstrial countries has ternational trade and structural 74 pages (including statistical annex). reached, or will soon reach, a limit; change. and (c) whether trade in manufactures Staff Working Paper No. 611. 1983. $3. among the developing countries can pages. expand further. Concludes with a dis- pages. cussion of the contribution of small ISBN 0-8213-0251-5. Stock No. WP 0611. enterprises to the creation of employ- $3. ment and the alleviation of poverty. Why the Emperor's New 1982. 70 pages (including. annex). Clothes Are Not Made in ISBN 0-8213-0017-2. Stock No. BK 0017. Colombia: A Case Study in $5. Latin American and East Asian Trade in Non-Factor Services: Manufactured Exports Past Trends and Current Issues David Morawetz Andre Sapir and Ernst Lutz Focuses on the exports of a particular Staff Working Paper No. 410. 1980. 140 commoity (clothing) from a particulaEr pages (including 4 annexes). Latin American country (Colombia) in Stock No. WP 0410. $5. an attempt to understand why Latin Trade in Services: Economic America has been so much less suc- Determinants and cessful at exporting manufactured Development-Related Issues goods to date than East Asia. It is the Andre Sapir and Ernst Lutz first study to go into great detail in ex- Staff Working Paper No. 480. 1981. 38 amining the plice, and especially the pages (including appendix, refrences). nonprice, determinants of export suc- Stock No. WP 0480 53. cess. Oxford University Press, 1981. 208 pages Trade Policy for Developing (including appendixes, bibliography). Countries . LC 81-9547. ISBN 0-19-520283-X, Stock Donald B. Keesing No. OX 520283, $22 hlardcover. World Bank Staff Wo?-,'ing Paper No. 353, August 1979, vii + 264 pages (including references). Stock No. WP-0353. $10. Trade Policy Issues for the Developing Countries in the 1980s Isaiah Frank World Bank Staff Working Paper No. 478. August 1981. 52 pages. Stock No. WP-0478. $3. The World Bank Publications Order Form SEND TO: YOlJR LOCAL DISTRIBUTOR OR TO WORLD BANK PUBLICATIONS (See the other side of this form.) P.O. BOX 37525 WASHINGTON, D.C. 20013 U.S.A. Date Name Ship to: (Enter if different from purchaser) Title Name -_- Firm 'r_ ,tle Address_ Firm City State_ _ Postal Code Address Country Telephone c ) City State_ Postal Code_ Purchaser Reference No. Country 'TeIephone ( Check your method of payment. Enclosed is my L Check L International Money Order O Unesco Coupons O Internaticnal Postal Coupon. Make payable to World Bank Publications for U. S. dollars unless you are ordering from your local distributor. Charge my L VISA L MasterCard L American Express L Choice. (Credit cards accepted only for orders addressed to World Bank Publications.) Credit Card Account Number Expiration Date Signature L Invoice me and please reference my Purchase Order No, Please ship me the items listed below. Customer Inlernal Stock Number Author/ Tite Routing Code Quantity Unit Price Total Amount I All prices subject to change. Prices may vary by country, Allow 6-8 weeks for delivery. Subtotal Cost $___ 'Ibtal copies .Air mail surcharge if desired ($2.00 each) $____ Postage and handling for more than two complimentary items ($2.00 each) $___ Tobtal $____ IBtD-006 Thank you for your order. DISTRIBUTORS OF WORLD BANK PUBLICATONS Distributors of World INDONESIA Attn: Mr. J.M. Hernandez Pt. Indira Linmited Castello 37 Bank Publications Attn: Mr. Bambang Wahyudi Madrid JI, Dr. Sam Ratulangi No. 37 SRI LANKA AND THE MALDIVES Jakarta Pusat Lake House Bookshop ARGENTINA IRELAND Attn: Mr. Victor Walatara Carlos Hirsch, SRL, TDC Publishers 41 Wad Ramanayake Mawatha Attn: Ms. Monica Bustos Attn: Mr. James Booth Colombo 2 Florida 165 4° piso 12 North Frederick Street SWEDEN Galeria Guemes Dublin 1 ABCE Fritzes Kungl, Hovbokhandel Buno Aie 130 JAPAN Attn: Mr. Eide Segerback AUSTRALIA, PAPUA NEW GUINEA, Eastem Book Service Re%eninpsgatan 1, Box 16356 FIJI, SOLOMON ISLANDS, Attn: Mr. Terumasa Hirano . Stockholm WESTERN SAMOA, AND 37-3, Hongo 3-Chome, Bunkyo-ku 113 SWITZERLAND VANUATUTokyo Libraie Paot The Australian Financial Review Tokyo L rair e Pa yot Information Service (AFRIS) KENYA Attn: Mr. David Jamieson Africa Book Services (E.A.) Ltd. 6, rue Grenus 235-243 Jones Street Attn: Mr. M.B. Dar 1211 Geneva Broadway P.O. Box 45245 TANZANIA Sydney, fSW 20001 Nairobi Oxford University Press BELGIUM KOREA, REPUBLIC OF Attn: Mr. Anthony Theobold Publications des Nations Unies Pan Korea Book Corporation Maktabaam Attn: Mr. Jean de Lannoy Attn: Mr. Yoon-Sun Kim. Dar es Salaam av. du Roi 202 P.O. Box 101, Kwanghwamun THAILAND 1060 Brussels Seoul Central Department Store, Head Office MALAYSIA Attn: Mrs. atana LANADA University of Malaya Cooperative 306 Silom Road Le A)iffuseur SBookshop Ltd. Bangkok CAP. 8, isoucherville J4B 5E6 Attn: Mr. Mohammed Fahim Htj Thailand Management Association Que 5,bec hrvleJ4 E Yacob Attn: Mrs. Sunan Quebec P.O. Box 1127, Jalan Pantai Baru 308 Silom Road COSTA RICA Kuala Lumpur Bangkok Libreria TreNos MEXICO UNITED KINGDOM AND Attn: Mr. Hugo Chamberlain INFOTEC NORTHERN IRELAND Calle 11-13, Av. Fernandez Guell Attn: Mr. Jorge Cepeda Microinfo Ltd. San Jose San Lorenzo 153-1f, Col. del Valle, Attn: Mr. Roy Selwyn DENMARK Deleg. Benito Juarez Newman Lane, P.O. Box 3 Sanfundslitteratur 03100 Mexico, D.F. Alton, Hampshire GU34 2PG Attn: Mr. Wilfried Roloff NETHERLANDS England Rosenderns Alle 11 MBE BV UNITED STATES DK-1970 Copenhagen V. Attn: Mr. Gerhard van Bussell The World Bank Book Store EGYPT, Arab Republic of Noorderwal 38, 600 19th Street, N.W. Al Ahram 7241 BL Lochem W, 'hington, D.C. 20433 Attn: Mr. Sayed El-Gabri NORWAY (Postal address: P.O. Box 37525 Al Galaa Street Johan Grundt Tanum A.S. Washington, D.C. 20013, U.S.A.) Cairo Attn: Ms. Randi Mikkelborg Baker and Taylor Company FINLAND P.O. Box 1177 Sentrum 501 South Gladiola Avenue Akateeminen Kirjakauppa Oslo 1 Momence, Illinois, 60954 Attn: Mr. Kari Litmanen PANAMA 380 Edison Way Keskuskatu 1, SF-00100 Ediciones Libreria Cultural Panamena Reno, Nevada, 89564 Helsinki 10 Attn: Mr. Luis Fernandez Fraguela R. 50 Kirby Avenue FRANCE Av. 7, Espana 16 Somerville, New Jersey, 08876 World Bank Publications Panama Zone 1 Commerce, Georgia 30599 66, avenue d'Iena PHILIPPINES 75116 Paris National Book Store Bernan Associates GERMANY, Federal Republic of Attn: Mrs. Socorro C. Ramos 9730-E George Palmer H2fhway UNO-Verlag 701 Rizal Avenue Lna,Mrln,27 Attn: Mr. Joachim Krause Manila Blackwell North America, Inc. Simrockstrasse 23 SAUDI ARABIA 1001 Fries Mill Road D-5300 Bonn 1 Jarir Book Store Blackwood, New Jersey 08012 HONG KONG, MACAU Attn: Mr. Akram Al-Agil Sidney Kramer Books Asia 2000 Ltd. P.O. Box 3196 1722 H Street, N.W. Attn: Ms. Gretchen Wearing Smith Riyadh Washington, D.C. 20006 6 Fl., 146 Prince Edward Road SINGAPORE, TAIWAN, BURMA United Nations Bookshop Kowloon Information Publications Private, Ltd. United Nations Plaza INDIA Attn: Ms. Janet David New York, N.Y. 10017 UBS Publishers' Distributors Ltd. 02-06 1st Floor, Pei-Fu Industrial VENEZUELA Attn: Mr. D.P. Veer Building Libreria del Este 5 Ansari Road, Post Box 7015 24 New Industrial Road Attn. Mr. Juan Pericas New Delhi 110002 Singapore Avda Francisco de Miranda, no. 52 (Branch offices in Bombay, Bangalore, SPAIN Edificio Galipan, Aptdo. 60.337 Kanpur, Calcutta, and Madras) Mundi-Prensa Libros, S.A. Caracas 1060-A