Policy, Planning, ana Research WORKING PAPERS International Commodity Markets International Economics Department The World Bank December 1989 WPS 348 Recent Trends and Prospects for Agricultural Commodity Exports in Sub-Saharan Africa Takamasa Akiyama and Donald F. Larson Sub-Saharan Africa's export dependence on coffee and cocoa has increased, but without policy changes to boost these and other agricultural exports, the region can expect only slow growth in its agricultural income terms of trade. The Policy, Planning, and Research Cornplex distributes PPR Working Papers to dissaninate the findings of work in progress and to encourage the exchange of ideas among Bank staff and all others interested in development issues. These papers carry the names of the authors, reflect only their views, and should be used and cited accordingly. The findings, interpretautions, and conclusions ame the authors'own. They should not be attributed to theWorld Bank, its Board of Drectors, its management, or any of its member countines. Plcy, Planning, and Research | International Commodity Markels Sub-Saharan African countries have seen sharp For their major agricultural export products declines in their shares of agricultural export -coffee, cocoa, cotton, groundnut oil, palm oil, markets. But their export dependence on the sugar and tobacco - prospects for world market most important crops - coffee and cocoa - has growth and for increases in Sub-S-`"aran increased. Comparisons in the region and with Africa's market shares are generall) only fair. countries outside the region show the impor- So, without further policy changes to improve tance of appropriate exchange rates and pro- perfonnance, the regions' agricultural income ducer pricing policies, as well as support for terns of trade for these commodities is expected technological advancement, for good perform- to recover only slowly frbm the sharp decline ance in these sectors. Some countries have after 1985. successfully introduced changes in these policy areas. Changes are also ne-,ded in domestic marketing systems to enable flexible responses to market demands in such areas as quality. This paper is a product of the Intemational Economics Department, International Commodity Markets Division. Copies are available free from the World Bank, 1818 H Street NW, Washington DC 20433. Please contact Dawn Gustafson, room S7-044, extension 33714 (58 pages with figures and tables). The PPR Working Paper Series disseminates the findings of work under way in the Bank's Policy, Plaruing, and Research Complex. An objective of the series is to get these findings out quickly, even if presentations are less than fully polished. The findings, interpretations, and conclusions in these papers do not necessarily represent official policy of the Bank. Produced at the PPR Dissemination Center CONTmETS PAGE I. Introduction ....e..e...e..e.....o......o...........o............I II. Recent Developments in Agricultural Commodity Exports.......3 III. Structure of Agricultural Commodity Exports of SSA Countries ooooooe.........................e...............8 IV. Factors Affecting Commodity Exports from SSA...............15 a. Real Exchange Rate Developments ........................17 b. Producer Price Developments........................................20 c. Technological Developments .. e.......................... 24 d. Declining Marginal Revenues ....e...... . .*.. .. .... 27 V. Prospects for Sub-Saharan Agricultural Commodity VI. Conclusions ......... 54*oooooo.. REFERENCES ........... The authors would like to thank Ron Duncan, Elton Thigpen and Brent Borrell of the International Commodity Markets Division for their important contributions to this paper as well as John Underwood of the Debt and International Finance Division for his assistance on t'ie issue of debt in Sub- Saharan Africa. RECKET TRENDS AND PROSPECTS FOR AGRICULTURAL COOMODITY EXPORTS IN SUB-SAHARAN AFRICA I. Introduction 1. Economic development in Sub-Sahardn Africa (SSA) is often considered problematic; recent economic growth has been poor and for many countries economic prospects are dismal. Recent developments in primary commodity prices and export revenues have contributed to Sub-Saharan Africa's predicament. As well, the pessimistic outlook for the primary commodities on which they depend heavily for revenue is one of the main factors responsible for the gloomy economic outlook for tbese countries. 2. While the present situation of sub-Saharan Africa has a long history, much of the current debt servicing problem begins with the commodity boom of the mid-seventies. With their creditworthiness enhanced by inflated commodity prices, many African countries used the boom to take advantage of the expanded availability of credit. As commodity prices fell quickly from boom levels, and then to even lower levels in the 1980s, many of these African countries have only been able to manage their indebtedness through rescheduling or by grants--and yet the burden remains. Total debt for the debt-distressed countries of Africa is around 100% of their combined GNP, and 500% of combined annual exports. Commercial credit has virtually disappeared, and 90% of the debt is held directly or guaranteed by offical creditors. Yet, with commodity prices at very low levels for much of the past decade, the countries of Sub- Saharan Africa collectively remain as dependent upon a few primary commodity exports as during the boom of the mid-seventies. 3. The present article is an update to the paper on the subject by Singh in 1983 and focuses on the recent trends in and prospects for SSA's agricultural commodity exports. Mineral and metal exports have not been discuss2d because, apart from a few countries such as Zaire and Zambia, the majority of SSA countries depend heavily on agricultural commodity exports as sources of their export revenues. The article not only examines agricultural commodity exports in aggregate but also at an individual level. This approach, we believe, reveals the past behavior and prospects in a much clearer manner. II. Recent Developments in Agricultural Comuodity Ezporta 4. During the period 1970-85 world agricultural food exports (SITC 0+1+22+4) and agricultural raw material exports (SITC 2-22-27-28) have increased at annual rates of 2.5% and .07%, respectively, in real terms, i.e., deflated by the unit value of manufactured goods exports (the World Bank's MUV index). These are significantly lower rates than those for fuels (10.4% p.a.) and manufactured goods (5.4% p.a.). However, growth rates of food and agricultural raw material exports vary widely from one region to anotN4r. As shown in Table 1, growth rates of agricultural commodity exports of African 1/ countries were considerably lower than those of developing countries in America and South and South-East Asia. In fact, Africa's food and agricultural raw material exports have had negative growth rates of -2.5% p.a. and -4% p.a. in real terms, respectively, during this period. Growth rates of agricultural commodity exports from South and South East Asian developing countries, measured in nominal terms, were more than twice as high as those of African countries. 5. As a result of this poor export performance, Africa's shares of world exports of all food items declined by one-half during the period 1970 to 1985 (see Table 2). Almost all the loss in their share was taken by developing countries in South and South East Asia. Corresponding figures for exports of 1/ UN statistics are for Africa as a whole, so in this section they are given for Africa and not for SSA. -4- Table 1: GROWTH RATES OF AGRICULTURAL COMMODITY EXPORTS OF DEVELOPING COUNTRIES BY REGION; PERIOD 1970-85 Developing Countries South and World Africa America Southeast Asia …_____________-…(Z p.a.)…---------------- All Food Items In nominal terms 10.3 5.3 10.2 13.1 In real terms a/ 2.5 -2.5 2.4 5.3 Agricultural Raw Materials In nominal terms 8.5 3.8 7.1 8.4 In real terms a/ 0.7 -4.0 -0.7 0.6 a/ Deflated by World Bank's index of the unit value of manufactured grods exports (MUV). Source: UNCTAD; UN. -5- Table 2: WORLD SHARES OF AGRICULTURAL COMMODITY EXPORTS BY DEVELOPING COUNTRY REGIONS All Food Items Developing Countries South and Year Total Africa America Southeast Asia -------------------------(X)-------------------------- 970 32.4 7.6 15.6 6.6 1975 29.2 5.1 14.6 7.7 1985 31.5 3.8 15.6 9.6 Agricultural Raw Materials Developing Countries South and Year Total Africa America Southeast Asia …__ _ _ _ _ __ _ _ _ _ _(z)…__ _ _ _ _ __ _ _ _ _ _ 1970 30.9 7.2 5.6 15.1 1975 25.2 5.4 4.6 12.8 1985 25.5 3.7 4.5 14.9 Source: UNCTAD; UN. -6- agricultural raw materials show thie same trend for Africa, i.e., halving during the period 1970-85. 6. Despite Africa's declining market share, the dependence of many Sub- Saharan African nations on agricultural exports remains startling. Twenty-two countries rely on agricultural exports for more than 50% of their total merchandise trade revenue. Agriculture provides more than 75% of export revenue for 14 nations and six countries (Burundi, Ethiopia, Malawi, Sudan, Somalia, Uganda) earn more than 90% of their export revenue from agriculture. 7. There is an important difference in the destination of agricultural exporti from African a:.d Asian developing countries. In 1985 only about 20% of Africa's agricultural commodity exports went to other developing countries while about 40% of Asian agricultural commodity exports went to other developing countries. This difference is important, because developing countries' imports of agricultural commodities increased by about six-fold during this period while those of industrial countries increased by only one- half as much. Which implies that Asian developing countries have been exporting commodities demanded in the fastest growing markets and/or marketed their commodities well into these markets. 8. As discussed in detail below, the major reason for Africa's declining world market shares in primary commodity exports is not due to an unfavorable composition of commodity exports, but rather due to a failure of Africa's production to competitively meet growing world demand. Given the relatively low rates of production growth, demand from an expanding population--which grew nearly 56% from 1970 to 1985 while the population in developing nations as a whole grew 44X--strained domestic resources as well. - 8 - III. Structure of Agricultural Commodity Exports of SSA Countries 9. A prominent characteristic of agricultural commodity exports of SSA countries iP that a few commodities have large shares of the total of agricultural exports. As shown in Table 3 and Figure 1, exports of cofit cocoa, tea, cotton and tobacco constituted 21% and 71% of total exports .4.d total agricultural exports of SSA countries, respectively, For the period 1984-86. The combinee share of coffee and cocoa alone wPa 54% of total agricultural exports. 10. The statistics in Table 3 also show that SSA's agricultural exports were considerably more diversified in the early 1960s. While coffee and cocoa were the two most important agricultural commodities exported by the SSA countries during that period, their combined share was only 32%. Coffee, cocoa and tea are the only Sub-Saharan agricultural commodities for which exports have increased during the last 16 years despite growing markets in a number of other commodities (Figures 2-4). It should be noted, however, that the export growth of tea was almost entirely due to the growth in Kenya's exports and that of cocoa was almost entirely due to increased exports from C6te d'Ivoire. If these two cases are - scluded, it is only coffee which recorded export growth among all the agricultural commodity exports by SSA. Table 3: SUB-SAiARAN AFRICA: SHARE OF TOTAL AGRICULTURAL EXPCRT REVENUE FOR SELECTED COM0D0ITIES Groundnut Cocoa and Eight Years Coffee Cocoa Tea Cotton Tobacco Sugar Oil Palm Cil Coffee Commodity Total ________ - - - - (S- 1961-63 15.8 15.9 1.9 10.6 5.1 4.2 10.6 2.7 31.7 66.8 1971-73 22.0 18.6 2.5 11.8 3.2 3.7 5.4 1.0 40.6 68.2 1984-86 30.2 24.2 4.3 7.8 4.4 4.5 1.3 0.5 54.4 77.2 Source: FAO. Figure 1: Export Revenue for Sub-Sahara Africa 21~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 20- 19 Forest Products 18 17 Cotton 16 -Cocoa 15 14- 13 12 11 10 ~~~~~~~~~~~~~~~~~~~~~~~~Coffee 'S 9 7 6 5 4 3 N ~~~~~~~~~~~~~~~~~~~~~~~Other 2 I 1961 1966 1971 1976 1981 1986 Figure 2: GROUNDNUT OIL EXPORTS 700 Others 600 a ~~~~~~~~SSA 0 ~~500 0 0~~~0 400 1 1 0 1~~~0 0~~~~ 100 I~90 98 18 Figure 3: Palm oil exports 8000 7000 __ _ __ _ _ __ _ __ _ _ __ _ __ _ _ __ _ _ _ _ __Others 6000 _______ SSA o 5000 _ a~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~- o 3000 2000 __r_____ 1000 - - - 196519698 1970 1980 1989 Figure 4: Tobacco exports 1400 Others 1200- Lq I ~~~~~~~~SSA S 1000 0 200 I I I 10~~~~ 1965 1975 1985 1970 1980 1989 - 14 - 11. Agricultural exports for the SSA region remain at the same share of total merchandise trade (about 29%) as a decade ago. 1/ In some countries the share of agricultural exports has grown. Burundi, Benin, Burkina Faso, Cote d'Ivoire, Comoros Islands, Ethiopia, Ghana, Guinea-Bissau, Kenya, Lesotho, Madagascar, Mali, Mozambique, Malawi, Rwanda, Sudan, Somalia, Sao Tome and Principle, Swaziland, Chad, Tanzania and Uganda all rely primarily on agricultural exports to generate export revenues (see Appendix Tables 1 and 2 for further details). 12. This general dependency is further exacerbated by several countries' dependency on only a handful of commodities. Export revenues generated from coffee and cocoa by Cote d'Ivoire, which amounted to more than 50% of total merchandise trade in 1974-76, accounted for more than 60% of revenues in 1984- 86--with nearly 40% coming from cocoa exports alone. Burundi is dependent on coffee alone for 85Z of its export revenues (up from 81% a decade ago); Kenya has become more dependent on coffee (up from 22% to 31%) and tea (up from 11% to 23%); coffee's share of export revenues has risen from 31% to 41% in Madagascar; Malawi remains dependent upon tobacco and tea for 65% of its export revenues (up slightly from a decade ago); nearly 37% of Mali's export revenues are derived from cotton sales (up from a decade ago); 41Z of Tanzania's export revenues are derived from coffee (up from 21% a decade ago); Uganda exports coffee and little else (87%--up from 77% a decade ago); Zaire has increased its dependency on coffee exports, which now make up nearly 26% of export revenues. 1/ See Appendix Tables 1 and 2 for details. - 15 - IV. Factors Affecting Agricultural Commodity Ezports from SSA 13. There are doubtless a large number of factors which explain the poor agricultvral commodity export performance of SSA. These include macro- economic oolicies, commodity-specific policies, developments in world agricultural commodity markets, declining marginal export revenues, technological developments, marketing efficiencies, land availability, and population growth. Here, the relevance of four factors is examined--real exchange rate movements, producer pricing policies, technological developments and declining marginal export revenues. 14. To limit the analysis, cocoa, coffee and cotton are the only agricultural exports treated in detail--however, these are important export commodities for SSA. Table 4 gives an idea of what has been happening to the shares of SSA countries in the world markets for these exports over the past 20 years. The shares of some major competitors are also shown for comparison. The dramatic examples in the table are the declines in the shares of cocoa exports held by Ghana and Nigeria and the increases in the shares of C8te d'Ivoire and Malaysia. The figures for Brazilian coffee exports are somewhat misleading as Brazil's coffee production has varied sharply due to frosts and drought. It should also be noted that export performance by cotton producers may not be a useful indicator of production performance; unlike coffee, and to a lesser extent cocoa, the development of cotton manufacturing in many developing countries has led to a reduction of exports of cotton fiber. Table 4: SH4ARES OF WORLD EXPORT MARKETS FOR COCOA, COFFEE AND COTTON, SELECTED COUNTRIES, 1969-71 TO 1987 (%) 1969-71 1979-81 1987 /a Countries Cocoa Coffee Cotton Cocoa Coffee Cotton Cocoa Coffee Cotton Cameroon 7.9 1.9 - 7.3 2.7 - 6.3 2.7 C6te d'lvoire 12.2 6.3 0.3 22.9 6.5 0.8 30.1 5.4 1.6 Egypt - - 7.8 - - 3.8 - - 1.8 Ethiopia - 2.5 - - 2.2 - - 1.9 - Ghana 31.0 - - 16.6 - - 11.6 - Kenya - 1.6 - - 2.0 - - 2.3 Mali - - 0.5 - - 0.9 - - 1.3 Nigeria 17.6 - - 9.7 - - 4.1 -- Sudan - - 5.7 - - 2.3 - - 2.6 Uganda - 5.8 - - 3.9 - - 3.3 - Zimbabwe - - 0.9 - - 1.2 - - 1.2 Brazil 11.7 33.7 8.0 17.3 25.1 1.5 16.0 24.8 2.3 Colombia - 12.1 - - 16.2 - - 16.5 Indonesia 0.2 2.6 - 0.4 6.1 - 2.0 6.7 Malaysia 0.2 - - 2.1 - - 8.4 - /a Estimated. Sources: International Coffee Organization, International Cocoa Organization, and Intcrnational Cotton Advisory Co_ittee. - 17 - a. Real Exchange Rate Developments 15. The real exchange rate is one of the most important factors affecting the competitiveness of a country's export sector. There is considerable evidence being accumulated to support the statement that over-valued exchange rates have been an important reason, if not the most important reason, why SSA's agricultural exports have fared so poorly in the past (see, for example, World Bank and the UNDP (1989)). Although important, it should be noted that real exchange rates and real producer prices affect the competitiveness of perennial crops only in the long-run as short-run elasticities are small. 16. Table 5 shows the real exchange rate movements 1/ of most SSA countries since 1970 as well as rates for some major agricultural commodity exporting countries in other regions. The aggregate real exchange rate index calculated for SSA was estimated by using agricultural commodity export values as weights. Movements since 1970 in the real exchange rates of the SSA in aggregate and for some SSA and non-SSA countries are shown in Figure 5. 17. Table 5 and Figure 5 show that, although some SSA countries have adjusted their real exchange rates to more competitive levels recently (e.g., Ghana, Nigeria, Somalia, Tanzania, Zaire and Zaimbia), real exchange rates for 1/ Real exchange rates of a country are usually calculated by multipiying the nominal exchange rates of the country concerned by the ratio of some inflation index of the country and the weighted average inflation index of its trading partners. Here they are calculated by multiplying nominal exchange rates by the CPI of the country concerned and did not take trading partners into consideration because the interest is in inter- country comparisons of competitiveness of producing certain commodities. - 18 - Table 5: REAL EXCHANGE RATES, SELECTED COUNTRIES, 1970-87 Country 1970 1975 1980 1985 1986 1987 Sub-Saharan Africa 100 179 376 39 298 306 Burundi 100 170 314 357 384 381 CBte d'Ivoire 100 185 405 250 346 420 Cameroon 100 213 358 278 372 470 Congo, People's Republic 100 189 300 240 319 376 Ethiopia 100 144 299 430 388 379 Gabon 100 212 395 295 407 465 Ghana 100 193 1,095 503 382 310 Gambia, The 100 190 324 269 236 286 Burkina Faso 100 178 300 206 260 292 Kenya 100 88 160 135 142 147 Liberia 100 167 258 304 315 331 Madagascar 100 201 318 252 283 206 Mauritius 100 164 280 215 250 263 Niger 100 186 374 253 317 341 Nigeria 100 220 549 793 425 205 Rwanda 100 208 357 448 512 586 Suaan 100 207 337 292 334 336 Senegal 100 243 342 282 389 429 Sierra Leone 100 139 231 370 212 279 Somalia 100 165 451 459 342 300 Swaziland 100 155 277 195 212 268 Seychelles 100 224 398 435 504 570 Togo 100 204 343 223 301 347 Tanzania 100 182 322 564 399 264 Zaire 100 233 559 193 236 239 Zambia 1O0 157 259 165 101 118 Zimbabwe 100 161 229 184 204 230 Brazil 100 147 179 145 162 186 Colombia 100 136 265 241 210 207 Indonesia 100 213 283 254 233 199 Malaysia 100 183 251 276 267 276 Source: IMF, "International Financial Statistics", various issues. Figure 5:Re al Exchange Rates 450 - 400 - 350 K 0 ~~~~~~~~~~~~~~~~~~~~~~~~~~SSA 0 250 I- 200 -Indonesia IBrazil 150 - ~~~~~~~~~~~~~~~~~~~~~~~Kenya 100 50- l l l l l l l l l l l l l l 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 - 20 - the region as a whole and for many individual countries are cuasiderably overvalued compared with some major non-SSA agricultural commodity exporting countries. In fact, using 1970 as the base year, Kenya and Zambia are the only two countries which have kept their exchange rates more competitive than Brazil, Colombia and Indonesia. This is arguably one of the reasons why Kenya's coffee exports have done well in competition with these latter three countries which are the largest coffee exporters. On the other hand, the fact that Ghana's real exchange rate appreciated considerably over the period 1970- 80 relative to other major cocoa producers such as Brazil, Cate d'Ivoire and Malaysia, was obviously an important reason for its loss of world cocoa market share to those countries. b. Producer Price Developments 18. One of the key micro-economic variables that determine production and hence export quantities of agricultural commodities is the producer prices in real terms. Recent trends of real producer prices for coffee, cocoa and cotton in major SSA countries and for major producers outside the region are given below. The marked differences in price trends between countries are generally consistent with the differences in trends in production and exports. 19. Coffee: Table 6 shows real producer price indices for coffee for major SSA coffee-producing countries and for the three most important producers outside the region. Except for Kenya and Zaire, real coffee producer prices for all the SSA coffee-exporting countries shown in Table 6 have been considerably lower in the mid-1980s compared with their level in the - 21 - Table 6: REAL PRODUCER PRICE INDICES FOR COFFEE, SELECTED COUNTRIES /a (1970-71 = 100) Country 1974-75 1980-81 1985-86 Burundi 83.3 99.3 72.4 Cameroon 72.9 93.4 80.4 C8te d'Ivoire 104.4 82.9 84.9 Ethiopia 77.8 57.4 48.8 Kenya 96.6 109.3 132.2 Madagascar 85.6 70.0 57.1 Rwanda 76.7 100.7 78.6 Tanzania 72.3 55.3 36.2 Zaire 82.7 89.1 153.5 Brazil 124.5 134.9 312.7 Colombia 97.4 96.8 112.2 Indonesia 76.1 131.0 163.4 Source: International Coffee Organization. ia Deflated by the CPI of each country. - 22 - early 1970s. By comparison, there was an increase in real prices in the world's three largest coffee exporters over this period. Not surprisingly, coffee production and export shares have declined in countries such as Cate d'Ivoire and Ethiopia where real producer prices have declined sharply in the last 15 years (sees Table 4 and 6). 20. Cocoa As shown in Table 7, since the early 1970s real producer prices of cocoa have been kept fairly stable in Cate d'Ivoire and Cameroon but declined sharply in Ghana and Nigeria--although real producer prices in the latter two countries have increased significantly in the last two years. However, in two major non-SSA cocoa-producing countries, Brazil and Malaysia, real producer prices have been considerably higher in the 1980s than in the early 1970s. As can be seen from Table 4, there has been quite a contrast in export performance between C6te d'Ivoire on the one hand and Ghana and Nigeria on the other; and between Brazil and Malaysia and Ghana and Nigeria. 21. Cotton. The real producer prices of cotton in SSA countries (given in Table 8) have been fairly stable over the period shown, except for Senegal and Sudan. It is to be ncted that among the SSA countries in Table 8 Senegal and Sudan are the two countries whose cotton production has stagnated over the last ten years. Recent experiences in SSA countries have shown that apart from favorable changes in real producer prices, timely availability of credit and inputs, the convenience of marketing services, the assurance of payment for marketed output, and appropriate land-use rotation are also important to the viability of the cotton industry. Countries such as Burkina Faso and Cameroon succeeded in increasing cotton production substantially in the 1980s, - 23 - Table 7: REAL PRODUCER PRICE INDICES FOR COCOA, SELECTED £C'UNTRIES /a (1970-71 a 100) Country 1974-75 1980-81 1983-86 Ghana 106.8 66.0 82.4 C6te d'Ivoire 144.0 104.4 107.2 Cameroon 86.1 112.6 99.9 Nigeria 115.6 20.8 36.9 Brazil 209.8 223.1 270.7 Malaysia 132.1 153.7 133.1 /a Deflated by the CPI of each country. Source: The World Bank. Table 8: REAL PRODUCER PRICE INDICES FOR COTTON, SELECTED COUNTRIES /a (1970-71 = 100) Country 1974-75 1980-81 1985-86 Burkina Paso 95.0 78.9 99.5 Cameroon 92.9 98.6 113.1 Senegal 72.3 77.7 68.3 Sudan 90.4 78.8 54.5 Tanzania 102.9 83.8 94.1 Zimbabwe 143.4 113.8 99.9 Egypt 107.4 114.2 103.7 China 99.8 130.5 120.8 /a Pakistan 120.5 96.1 77.8 Turkey 155.2 94.6 95.1 /a Deflated by the CPI of individual countries. 7i Average of 1984-85. Source: FAO; World Bank. - 24 - due to favorable producer prices and other factors such as those mentioned above. c. Technological Developments 22. A country can increase its production of a commodity even when real producer prices are declining, if advances in the production technology result in a significant cut in production costs. It is very difficult, however, to estimate production costs for a country or region. Various estimates of production costs of agricultural commodities show that an important variable that affects production costs per unit weight is yield per unit of harvested area. They show that, generally, production costs decline with yield increases. Hence in this section, the analysis is focused on the developments in yields of major agricultural export commodities in SSA. 23. Coffee. Over the last 15 years research efforts in many coffee- producing countries have focused on the development of high-yielding and disease-resistant varieties and optimal input packages. 24. Table 9 shows changes in coffee yields for some major coffee producing countries, including countries outside the SSA region. Except for Kenya and Rwanda, yields in SSA countries are lower than for other countries in the table. Yields are high in Kenya and Rwanda because there is very limited arable land per capita; so intensive cultivation methods are employed. These countries also have very well organized coffee research facilities. Madagascar and C6te d'Ivoire have the lowest yields among the countries in - 25 - Table 9: COFFEE YIELDS, SELECTED COUNTRIES, 1970-71 TO 1985-86 (kg/ha) Countries 1970-71 1975-76 1980-81 1985-86 Cameroon 351 356 464 523 C6te d'Ivoire 327 334 263 230 Kenya 726 1,037 726 781 Madagascar 372 352 339 288 Rwanda 862 1,344 1,044 1,082 Uganda 494 528 461 441 Zaire 409 351 390 461 Brazil 561 576 593 593 Colombia 564 570 863 722 Costa Rica 1,201 1,037 1,560 1,610 Mexico 559 604 608 756 Indonesia 469 568 630 564 Source: USDA, World Coffee Situation, August, 1988. - 26 - Table 9; moreover, they have been declining over time. It is believed that the main reason for the decline in C6te d'Ivoire's yield has been the decline in coffee producer prices compared with prices paicA to cocoa producers. Because of the movement in relative prices against coffee, little new planting has taken place and the coffee trees are aging fast. Madagascar's poor yield development has also been due to declining real producer prices. By comparison, the development and planting of new high-yielding varieties is reflected in the high and increasing yields in Colombia and Costa Rica. 25. Cocoa. A major technological development in cocoa over the last 15 years has also been the development of hybrid, high-yielding varieties. Significant areas of hybrid varieties have been planted in Malaysia and Brazil. Although complete data are scarce, development and application of high-yielding varieties are reportedly quite limited in SSA. The high level of yields in these major cocoa producing countries, by comparison with other countries shown in Table 10, reflects the impact of the hybrids. In Malaysia, which has the highest yield among the countries in Table 10, new clonal varieties have recently been developed wh4ch can yield more than 2,000 kg/ha. Table 10: COCOA YIELDS IN MAJOR COCOA PRODUCING COUNTRIES IN 1986-87 (kg/ha) Cameroon 373 Cote d'Ivoire 555 Ghana 250 Nigeria 335 Brazil 595 Indonesia 730 Malaysia 745 Source: The World Bank. - 27 - 26. Cotton. Table 11 gives cotton yields in major producing countries over the period since 1970. The data show that cotton yields in SSA countries are generally lower than in non-SSA cotton-producing countries, except for Brazil. Table 11 also shows that cotton yields in some West African countries, i.e., C6te d'Ivoire, Cameroon, Burkina Faso and Togo have increased sharply in the 1980s. These countries have benefited from attractive pricing policies ant well-functioning institutional arrangements that have encouraged the use of a package of modern inputs. The primary difference between yields in the West African countries and in the large cotton-producing countries outside the region is accounted for by the fact that most of the cotton in West Africa is being grown under rainfed conditions, whereas a large proportion of the cotton produced in the non-SSA countries shown in the table, except for Brazil, is grown with irrigation. d. Declining Marginal Revenues 27. As noted in Section II, SSA's two most successful agricultural commodity export activities have been cocoa in C6te d'Ivoire and tea in Kenya. Ivorian cocoa exports increased at an average rate of 7.4Z p.a. during the period 1970-86 and the corresponding figure for Kenya's tea exports was 7.5Z p.a. These fast growth rates were, in the main, the result of appropriate pricing policies and the ready availability of suitable land. 28. It is doubtful that in the future these products will be able to contribute is much to the export revenue growth of these countries as they did in the past. A major reason is their declining marginal revenues. - 28 - Table 11: COTTON YIELDS IN MAJOR PRODUCING COUNTRIES, 1970-71 TO 1985-86 (kg/ha) Country 1970-71 1975-76 1980-81 1985-86 SSA Countries Sudan 484 335 343 431 Mali 314 424 450 487 C6te d'Ivoire 357 437 447 562 Chad 122 190 192 257 Cameroon 150 247 493 517 Burkina Fasso 123 241 318 529 Senegal 366 335 282 357 Tanzania 250 139 154 125 Togo 289 296 338 543 Zimbabwe 402 580 454 352 Non-SSA Countries Egypt 770 716 1,010 927 Syria 614 810 877 913 Pakistan 337 246 342 525 Turkey 760 768 745 835 China 444 450 560 815 Brazil 231 247 311 333 Mexico 778 863 925 953 Source: USDA, World Cotton Situation, May 1988. - 29 - 29. A decline in marginal revenue occurs when a country with a large share of world exports of a commodity facing low price elasticity of import demand increases its exports. The additional export quantity depresses the world price of the commodity. Because the country is already exporting a large quantity of the commodity, the reduction of export revenues caused by this price decline is large. The price elasticities of world import demand for cocoa and tea are very Low--estimated to be about 0.24 in the short run and 0.3 in the long run for cocoa, and slightly lower for tea. Given the low price elasticity of world import demand for these products and the large market shares which C8te d'Ivoire has in cocoa (over 30% of world exports in 1987) and Kenya has in the tea (about 15% of world black tea exports in 1987), the percentage increase of export revenues would be considerably less than any percentage increase of export quantities. For example, if COte d'Ivoire increases its cocoa exports by 1%, its export revenue from cocoa is estimated to increase by less than 0.6% (see Akiyama, 1988). Conversely, because of its small market share, a 1% increase in tea exports from Halawi would have little impact on the world tea price. Finding itself in such a situation does not necessarily imply that a country should not increase further its production and export of the commodity in question; but it does imply that the economic and financial analysis of any projects involving production increases of these commodities needs to take account of the marginal revenue impact of the project. - 30 - V. Prospects for Sub-Saharan Agricultural Commodity Exports 30. As discussed in the previous section, a few commodities constitute a large share of the total agricultural commodity exports of the SSA . Thus, the prospects for the agricultural commodity exports of these countries can, in the main, be evaluated by examining the prospects of these few commodities. Table 12 contains export volume and value projections for SSA and the world for the commodities which have a significant impact on SSA terms of trade. The total values are revealing: over the projection period world epxort value for the eight important SSA export commodities (in real terms) is projected to increase nearly 28Z while exports from SSA are expected to climb less than 16X. Individual commodity prospects are discussed below. 31. (a) Coffee. There is great uncertainty surrounding the world coffee market at present. 1/ The International Coffee Agreement's (ICA) export quota system, which has been in operation since October 1980, was suspended in early July of 1989. At the time of writing, the quota system is not operating and there are no plans to re-introduce it. Because of the suspension, world coffee prices fell by about 40% between early June and mid-October. 1/ Because of increasing world coffee prcduction relative to demand, the ICA quota system was necessary to maintain stable world prices. Disagreements among ICA members over the issues of exports to non-ICA members and of quota distribution among exporting members led to the suspension of the quota system in July 1989. Table 12: ECR VAUIE AND 'JE V FRCWrIGN6 Fat SEILED) DtDTTS COcs COffee Cotton Growumiit Oil Palm SOS Tea Ibbacoo Total Year Volume Valu Volume Value Volume Vale Volume Val Volm Val Volum Value Volume Value Voume Valu Value 1988 1.13 1,295 0.93 2,044 0.65 659 0.15 64 0.20 63 1.66 270 0.27 36 0.18 267 4,697 1989 1.14 1,039 1.05 1,901 0.67 824 0.14 75 0.20 52 1.70 351 0.28 39 0.18 275 4,555 1990 1.17 805 1.05 1,487 0.70 875 0.14 65 0.20 50 1.75 436 0.28 42 0.17 281 4,040 1991 1.19 837 1.03 1,548 0.72 895 0.14 65 0.20 52 1.80 428 0.30 44 0.18 277 4,147 1992 1.21 871 1.01 1,605 0.75 914 0.14 66 0.19 53 ^c 419 0.31 47 0.18 273 4,248 1993 1.24 905 0.99 1,659 0.77 932 0.14 67 0.19 55 1.91 408 0.32 49 0.18 269 4,343 1994 1.26 940 0.96 1,708 0.80 949 0.15 67 0.18 57 1.96 397 0.34 51 0.19 264 4,432 1995 1.28 876 0.94 1,753 0.82 965 0.15 68 0.18 58 2.01 384 0.35 53 0.19 259 4,516 1996 1.29 1,079 0.94 1,799 0.84 974 0.15 69 0.16 52 2.07 409 0.37 56 0.19 260 4,698 1997 1.30 1,183 0.94 1,846 0.86 982 0.16 70 0.15 46 2.13 434 0.38 59 0.19 261 4,881 I998 1.31 1,289 0.94 1,893 0.88 990 0.16 70 0.13 40 2.19 460 0.40 62 0.20 262 5,066 1999 1.32 1,396 0.95 1,939 0.90 997 0.16 71 0.12 35 2.25 487 0.42 65 0.20 263 5,253 2000 1.33 1,505 0.95 1,986 0.92 1,003 0.16 72 0.10 30 2.31 515 0.44 68 0.20 264 5,442 Wbrld 1988 1.97 2,268 4.00 8,794 5.03 5,080 0.59 254 6.53 2,070 27.83 4,537 1.77 230 1.36 2,049 25,281 1989 2.02 1,842 4.49 8,135 5.05 6,212 0.60 327 7.34 1,923 27.53 5,671 1.80 248 1.35 2,089 26,447 1990 2.06 1,421 4.54 6,399 5.10 6,375 0.60 281 7.58 1,886 27.72 6,902 1.82 272 1.33 2,139 25,673 1991 2.10 1,479 4.56 6,834 5.15 6,360 0.61 287 8.04 2,120 28.08 6,667 1.87 280 1.35 2,105 26,131 1992 2.14 1,538 4.57 7,272 5.21 6,343 0.63 293 8.51 2,368 28.45 6,423 1.92 288 1.37 2,070 26,596 1993 2.18 1,598 4.59 7,713 5.26 6,325 0.65 300 8.98 2,629 28.81 6,171 1.97 296 1.38 2,034 27,066 1994 2.22 1,660 4.61 8,157 5.32 6,305 0.67 306 9.45 2,904 29.18 5,911 2.01 305 1.40 1,995 27,542 1995 2.27 1,722 4.63 8,604 5.37 6,283 0.69 311 9.91 3,192 29.54 5,642 2.06 313 1.42 1,955 28,024 1996 2.30 1,916 4.66 8,886 5.45 6,288 0.70 315 10.28 3,257 29.94 5,911 2.11 324 1.44 1,963 28,860 1997 2.33 2,115 4.69 9,171 5.53 6,291 0.71 319 10.65 3,317 30.35 6,185 2.17 334 1.46 1,971 29,702 1998 2.36 2,318 4.72 9,459 5.61 6,291 0.72 322 11.01 3,374 30.75 6,464 2.22 345 1.47 1,978 30,550 l999 2.39 2,526 4.75 9,749 5.69 6,289 0.74 325 11.38 3,427 31.15 6,748 2.28 355 1.49 1,984 31,404 2000 2.42 2,738 4.78 10,043 5.76 6,284 0.75 329 11.74 3,476 31.56 7,037 2.33 366 1.51 1,990 32,264 NMY!: Vohume Is In malon tons, value in million 1985 US$. Souree: International CQaunodity Markets Division. - 32 - 32. Should the suspension of the coffee export quota system continue through the year 2000, the Division's global coffee model predicts real coffee prices will be much lower than the recent level for the next 4 years or so. Although export-quantity increases are expected to partially compensate for price declines, coffee export revenues of SSA are expected to be considerably lower over the next several years than in the recent past (see Figure 6). These low prices may well have adverse effects on many coffee producing countries including those in SSA countries. Unless significant changes are made in the tax systems or in exchange rates, these low world prices would result in lower producer prices in these countries. This, in turn, would reduce farmers' incentives to replace aged trees, to plant new higher-yielding varieties and provide good management to an existing tree stand. As a result many SSA countries would not only suffer from low world prices but also from low production and hence low export quantities for several years. 33. Due to expected low coffee production in many countries because of low prices in the early 1990s, prices should start strengthening by the mid- 1990s. This helps SSA's coffee income terms trade to increase in the second half of the 1990s. 1/ In spite of this recovery, as shown in Figure 6 their income terms of trade is projected to be about 40% and 60% lower than in the 1960s and the 1970s, respectively. 1/ The income terms of trade is defined as export revenue from the commodity deflated by the MUV. It is preferred to present the income terms of trade rather than the net barter terms of trade as the former illustrates the impact of productivity on export supplies (assuming real wages are not deteriorating). Figure6: Income terms of trade Coffee 320 - 300 - 280 - 260 - 2a0 - I A' 220 - 0 0 200- 180-~~~~~~~1 160 140- 100-"~A A ,A too 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 1 Africa + World - 34 - 34. A current trend in the world coffee market, which we believe will persist whether or not there is an ICA, is relatively weak prices and import demand for robusta as compared with arabica varieties. Almost two-thirds of coffee produced in SSA is robusta--mainly from Western African countries. Hence, under any circumstances, SSA's coffce exports are likely to fare worse than other region's. 35. Coffee production in several SSA countries is expected to stagnate or decline, mainly due to aging trees and lack of new plantings in recent years. These countries include Cameroon, C6te d'Ivoire, Ethiopia, and Madagascar. A number of Communante Financiere Africaine (CFA)-zone countries and Rwanda have potential for increasing production but their over-valued currencies may make it difficult for the governments of these countries to implement policies to increase production. Countries such as Burundi, Kenya and Rwanda face acute shortages of arable land and their production expansion depends on whether they can increase yield. 36. Recent positive developments include development of high-yielding varieties in several SSA countries and increasing shares of washed coffee in Ethiopia and Burundi which fetches considerably higher unit export prices than unwashed or semi-washed coffee. There is also a plan to introduce arabicas in Madagascar where up to now only low-quality robustas have been grown. This program, if successful, should increase that country's coffee export revenues substantially. - 35 - 37. (b) Cocoa. The world cocoa market is entering its fifth consecutive year of excess supply. According to recent estimates, by the end of the current crop year (1988/89) world stocks will increase by about 300,000 tons. This increase will bring cocoa closing stocks to over 1.1 million tons, or over half a year's consumption--the highest level in the post-World War II period. For the current crop year, production is exoected to increase by about 9.1% and consumption by 3.6% over 1987/88 levels. (The annual growth rates for 1987/88 were 9.5% and 4.9% for production and consumption, respectively.) The recent high rates of growth in production are due to the past high levels of new plantings in the major producer countries. Production in Cote d'Ivoire, Malaysia, Indonesia, and Ghana has increased significantly over the last four years. Cote d'Ivoire's average rate of growth in production over the past four years was 6.5% p.a., while for Malaysia and Indonesia it was 27% and 22% p.a., respectively. In Chana, recent policy changes benefiting Lhe cocoa sector have led to a production increase from 159,000 tons in 1983/84 to 305,000 tons in 1988/89. World consumption has also been increasing, with consumers taking advantage of lower prices--but not fast enought to keep up with production. Prices in both real and nominal terms have been steadily falling since 1984, reflecting the persistent imbalances between demand and supply. We expect this trend to continue into the early 1990s. However, the lower prices will induce lower levels of new plantings and less use of pesticides and fertilizers, which will eventually reduce the growth of production. On the other hand, consumption will continue - 36 - growing as a response to lower prices. 1/ Eventually, the imbalance between supply and demand will close and prices will increase in real terms. We expect this increase to take place in the late 1990s. 38. SSA's export volume is expected to increase at 1.6Z p.a. As in the case of coffee, SSA's income terms of trade for cocoa is expected to be much lower in the 1990s than for the period 1976-85. Past and projected income terms of trade for cocoa are shown in Figure 7. Export volumes from Ghana and Nigeria should show healthy growth in the 1990s--but this is from a relatively low base. C6te d'Ivoire, which has recorded export volume growth of over 7% p.a. and which has grown to become the world's largest producer, is expected to increase its cocoa exports only at about 2% p.a. in the 1990s. 39. The much lower production growth rate expected in C6te d'Ivoire is due in large part to its overvalued exchange rate. Because of a sharp decline in world cocoa prices and substantial appreciation of the CFA against the US dollar in the last three years, producer prices have been considerably higher than the f.o.b. prices after deductions for domestic marketing and transportation costs. Because the government has not been able to pay in full the posted producer prices, the cocoa subsector has been in chaos recently. The large reduction recently made in producer prices should discourage them from increasing cocoa production. 1/ The chocolate industry expects rates of increase in demand for cocoa of around 3.0-4.5Z p.a. for the next 4-5 years. Figure7: Income terms of tradce Cocoa 450 - 400 - 350 - 300 - 0 0 VI 100 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 0 Africa + World - 38 - 40. Although total export revenues from cocoa in Cote d'Ivoire and Ghana are expected to increase with increased export volume, Cote d'Ivoire, and to a lesser extent, Ghana, face declining marginal revenues because of their large share of exports. Thus, further expansion of their shares of world markets will become increasing1y more difficult to justify from the point of view of the country as a whole. 41. (c) Tea. SSA's income terms of trade for tea is projected to do well in the 1990s (See Figure 8). Real world tea prices are expected to show a steady increase from the very low levels of recent years. SSA's tea export volume is also projected to increase at a high rate of 4.1X p.a. over the next 12 years. However, this high projected growth in SSA's income terms of trade for tea is, as in the past, primarily due to growth in Kenya's exports. 42. As in the case of Cote d'Ivoire with cocoa, Kenya's future export revenues are not likely to increase in proportion to its export quantities because of declining marginal revenues. Furthermore, availability of suitable land will increasingly become a serious constraint to production growth in Kenya and Malawi. 43. (d) Cotton. World cotton prices are projected to increase over the period to 2000 from the very depressed levels of the mid-1980s. The cotton surplus accumulated in 1984/85 (when Chinia's production reached a record level and output in the United States increased sharply as planting restrictions under the payment-in-ki.id support policy were relaxed), has been liquidated. World cotton consumption has increased strongly in recent years and fiber prices have recovered to attractive levels. Figure 8: Income terms of trade Tea 320 - 300 - 280 - 260 - 240 - 220 - 200 - 180- y <<,1,, 60 1961 1964 1967 1970 1973 1976 1979 1982 1985; 1988 1991 1994 1997 2000 O Africa -1 World - 40 - 44. The strengthening of cotton prices will stimulate production in the early 1990s and likely lead to another round of low prices. In the longer term, cotton's declining competitiveness in the fiber market, along with declining population growth and modest income growth prospects in the industrial regions, means that cotton consumption will grow at only a modest pace. The tendency for synthetic fibers to be substituted for cotton in developing countries as incomes grow and technology allows will reinforce the tendency towards slower growth and further declines in cotton's share of the fiber market. 45. SSA's cotton-producing countries were particularly concerned about food availability during the 1970s and early 1980s and, as a result, relatively less attention was paid to the cotton subsector and production stagnated. Cotton production has recovered in many SSA countries, and during each of the three most recent seasons it has exceeded the 1979-81 period average output by more than 50Z. Also, their cotton exports have doubled in recent years. This impressive increase in cotton production was achieved by acreage expansion and increased productivity, particularly in the francophone countries in West Africa. Their increased use of improved varieties, chemical fertilizers, and pesticides in block plantings contributed substantially to the uptrend in cotton yields. 46. In response to heavy losses in cotton, due to extremely low export prices during 1986/87, cotton sector adjustment efforts were mounted in several SSA countries to reduce the costs of production and marketing. In Chad, cotton growing is now concentrated in the more productive areas. This - 41 - concentration of the cotton area has reduced the cost of collecting seedcotton and only 6 gins are now operating compared to 20 earlier when production was more widely spread. With the emphasis on higher productivity, future increases in cotton production in Chad are expected from improved yields rather than from expanded cotton-growing area. Chad has also changed its pricing system so that producers' returns are based on prevailing export prices rather than fixed, guaranteed prices. In Mali, the cotton-growing area increased sharply in 1988/89, but additional expansion is likely to be constrained by concerns for the adverse effects of cotton monoculture on soil fertility and erosion. There is growing concern in several other West African cotton-producing countries about this problem and appropriate rotation systems are being sought. In those countries where the tendency to intercrop cotton and foodcrops is strong and the use of modern inputs is not generally practiced, yields are low and production increases are conscrained. 47. Many of the SSA countries plan to continue increasing their cotton production in the coming year. With the projected increase in world cotton prices and in SSA's export volumes, SSA's income terms of trade for cotton is projected to show healthy growth (see Figure 9). Opportunities for cotton export growth are attractive in those SSA countries following supportive macroeconomic policies and developing means for assured input supply and efficient marketing structures. This, in addition to relatively low cost structures, makes prospects promising for further gains in cotton exporting in Benin, Burkina Faso, Mali, C6te d'Ivoire and Togo. Cotton production in Tanzania has recovered strongly since 1986 under the policies of the Economic Recovery Program. This increase in output was largely due to acreage Figure 9: Income terrms of trade Cotton 220 210 - 200- 190 - 170- 160- 0 150- 0 140 to 130 - 120 - i IR -A 0 -, 50 - 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 01 Africa + World - 43 - expansion but, if conditions remain favorable and institutional inadequacies are corrected, there is the potential for further gains from improved productivity. Cotton's future in Sudan is fraught with uncertainty,but there is an opportunity to supply a larger share of the extra-long staple market. Cotton production in Zimbabwe has fluctuated between 80,000 and 100,000 tons in recent seasons, but new ginning capacity was added in 1989 and further capacity is planned for 1991. Cotton acreage continues to expand in the communal areas, and two new varieties have been released for the 1989 crop. The new varieties produce longer staple fiber with improved strength and fineness which sells at a premium in the export market relative to previous varieties. 48. (e) Tobacco. Tobacco is one of the few agricultural commodities in which SSA's export performance has been better than that of other regions. Because of consumer health concerns, world demand growth has been slowing down over the last 15 years and should slow further in the future. World tobacco prices in real termo are projected to stay stagnant in the 1990s and SSA's export volume is projected to increase at 0.7% p.a. As a result, only low growth of SSA's income terms of trade for tobacco is projected (see Figure 10). The direction of trade is also changing as new investment in cigarette manufacturing capacity is shifting to those developing countries where rising incomes are increasing the share of the population that consumes manufactured cigarettes. The major African tobacco producers, Malawi and Zimbabwe, are expected to continue increasing production by increasing yields--although acreage expansion may be constrained. Figure 10:Income terms of trade Tobacco 180 - 170 - 160 - 150 - 'k 140 - 0 03 130 I~~~~~~~~~~~~~~~~~ 1*0 OOh 9*I~9 110 - 100 7 - . . | . . * . . ' . . | . . * . , ' . . | . . I . . * . 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 O Africa + World - 45 - 49. (f) Vegetable Oils and Oilseeds. Over the longer run, the emergence of the EEC as a significant producer of oilseed products and protein meals and the continued expansion of low-cost vegetable oil and oilseed production in Argentina, Brazil, Indonesia and Malaysia will pressure international prices, despite growing world demand. Harvested soybean area in Brazil grew from 1.3 million hectares in 1970 to 10 million hectares in 1985. Argentine soybean area grew tenfold in the 1980s--from 360,000 hretares in 1980 to 3.6 million hectares in 1987. Palm oil production doubled in Malaysia from 1962 to 1967. It had doubled again by 1970, 1973, 1977 and 1982. By 1995 it will double once more as expansion over the next decade is gueranteed by past planting practices. Over the past ten years, new plantings in Malaysia have averaged over 85,000 hectares per year. New plantings in Indonesia have increased by over 40,000 hectares per year during the same period. These cohorts of genetically-improved trees will move into full production over the next decade. 50. However, the world market for vegetable oils remains viable and dynamic. Investments in low-cost producing areas still offer high rates of return despite the expectation of falling real prices. World demand, spurred by falling prices and expanding supplies is expected to grow significantly-- especially among developing countries. For example, per capita consumption grew from 3.08 kg in Indonesia in 1970 to 8.7 kg in 1980; in China, from 2.65 kg to 7.5 kg; and in India from 5.48 kg to 7.2 kg; and yet all three countries are still substantially below the world's average of 13.1 kg per capita and significantly below the EEC-US average of more than 38 kg per capita. As a result, production, consumption and trade are all expected to grow, as prices - 46 - fall. Low-cost producers and low-cost vegetable oils are likely to gain a larger share of these growing world markets. 51. Palm Oil. The story of palm oil in Africa is closely linked to developments in Nigeria. In 1958 Nigeria was the world's largest producer of palm oil, generating over one-half the world's supply. Production fell abruptly during the late 1960s due to civil strife, then again in the 1970s as trees were neglected during the petroleum boom years. By 1985? Nigerian production was at 441 of 1958 levels, despite a growing domestic and international market. As production levels grew in Malaysia and Indonesia and Nigerian production fell, SSA's share of the world market fell precipitously and, as can be seen in Figure 11, SSA's income terms of trade languished as world trade soared. 52. Production gains in Cameroon, the Congo, Cate d'Ivoire and Zaire have only partially offset declines in Nigeria. Both Camerooyn and Cate d'Ivoire have made significant gains in palm oil production and processing, yet these two countries are severely limited in terms of areas with sufficient rainfall. The Congo River Basin offers a potentially large productive area for the Congo and Zaire, yet transportation remains a problem for both countries. The potential for SSA to recover markets lost in oil palm remains limited in the foreseeable future. Though indigenous to West Africa, varieties of oil palm have been engineered which thrive in the damp forests of South East Asia. While plantations in Malaysia and Indonesia routinely produce over 20 tons of fresh fruit bunches per hectare, yields fall to 16 tons in C6te d'Ivoire, 14 tons in Nigeria and 9 tons in Ghana. Figure 11: Income terms of trade Palm oil 800 - 700 - 600 - 500 - 0 0 '4 300 - 100 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 O Africa + World - 48 - 53. Groundnut Oil. While SSA missed a growing market in palm oil, it has been squeezed by a limited export market for groundnut oil. China, India and the United States are large sources of final demand, yet groundnut oil imports are limited in these markets due to import restrictions or producer subsidies, leaving a very thin market for groundnut oil trade. In 1987, only 380,000 tons was traded. Of this, 274,000 tons were imported by the EEC. However, as EEC production of sunflowerseed oil--which has similar cooking properties to peanut oil--has increased in recent years, the demand for imported groundnut oil has fallen. SSA, which has traditionally exported much of its oil to Europe, has been particularly hard hit. While prospects for groundnut export growth are not bright, some price recovery is expected in the short run. Additionally, groundnut export quantities are expected to stabilize and grow slowly from current low levels. Hence, the income terms of trade is expected to show a slight increase over the long term (see Figure 11). 54. (g) Sugar. Sugar is a commodity which is projected to do better in the 1990s compared to its very poor performance during much of the 1980s. However, prices are likely to remain extremely variable. 55. Ending stocks are expected to decline for the fifth consecutive year in 1988/89. Recent price increases, which reflect the decline in stocks, may encourage increased production in some low-cost exporting countries. Nevertheless, if world sugar consumption continues to increase by around 2.5 million tons a year, with stocks already at a very low level, even small shortfalls in production could have a large impact on price. Figure 12: Income terms of trade Groundnut oil 160 - 150 - _ 140 - 130 - 120 - 110 - 100 - 0 o 90 - 80- tD~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~4 70 %0 6 0 -_ _ _ _ _ _ _ _ _ _ _ 50 - 40- 30 - 20 - 10- 0 - . A . . I 1 I U I * ''-I I ' ' I ' I ' ' I ' ' I ' ' I' ' I ' 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 0 Africa + World - 50 - 56. Because of the high degree of goverment intervention in sugar industries around the world, producers are not highly responsive to changes in world prices. It is unlikely, therefore, that growth in production will keep pace with growth in consumption over the next three to four seasons, despite improving prices. In the past, governments have changed their sugar policies in ways which encouraged investment only after a period of very high world prices. Synchronized changes in policies worldwide, and the insulation which policies afforded producers from low world prices, have eventually led to a large overexpansion of sugar-producing capacity. As a result, stocks have built up and world prices have slumped. Because of producers' insultation from low prices, high levels of sugar production have been maintained and the slumps in world prices have lasted for several years. During these years world prices fell below producer costs in many SSA countries. This cycle of events is likely to be repeated. 57. As the timing of any production shortfall cannot be predicted, the timing of a boom and subsequent slump in world prices is difficult to predict. However, the most lirely scenario is for a short period of very high prices early in the 1990s followed by another prolonged period of low prices. 58. As shown in Figure 13, SSA's income terms of trade for sugar is projected overall to fare considerably better than that for other regions due mainly to a projected 2.8% p.a. increase in exports over the next 12 years. In the decade to 1987, SSA's exports expanded at an annualized rate of 3.2%-- albeit from a low base. With some improvement in prices into the 1990s low-cost producers of SSA (Malawi, Mauritius, Swaziland, and Zimbabwe) are Figure 13 Imcome terms of trade Sugar 900 - 800 - 700 - _ _ _ _ _ _ _ _ _ _ _ _ 600 - . .. 1_ _-- ---- - - . 0 500- 8,, 400 200X,1 g&zi 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 O3 Africa t Wotrld - 52 - expected to achieve further growth. However, resource constraints such as the availability of water for irrigation are likely to reduce the rate of growth compared to the previous decade. 59. Figure 14 shows the past and projected income terms of trade for the aggregate of the eight commodities discussed above. It shows SSA's income terms of trade for these commodities to increase slightly over the forecast period. The figure also shows the SSA's future income terms of trade from these commodities to be much lower than *n the 1970s and even lower than in the 1960s. 60. While the income terms of trade for Sub-Saharan Africa is expected to recover partially by the 1990s, its growth in Africa will lag world growth due to the mix of products exported from Africa. Although tea, tobacco and cotton are expected to perform well in Africa, tea and tobacco face limited growth in final demand. Cotton prospects are more favorable, but only a limited number of countries in Africa are likely to benefit. Sugar prices are expected to recover in the near term and several of the island nations of Africa are expected to gain from the recovery; however, the area in Africa suited for expansion of sugar production is limited and little production growth is expected. In the vegetable oil market, world production is expected to expand rapidly--but not in Africa--with production gains centered in Asia and South America. The future for Sub-Saharan Africa is, by default, closely tied to the prospects for cocoa and coffee which are expected to grow quite slowly. Figure 14: Income terms of trade Aggregate 240 - 230 - _ 220 - 200 - A 0 190- h [Al IS0- . o 170- 0 160- 140- 130 - NM 120- I10 - 90 - 80 70- * 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 0 Africa + World - 54 - VI. Conclusions 61. Sluggish agricultural production and export growth has reduced SSA's agricultural commodity export revenues in the mid-1980s to only about 60Z of what they were 15 years ago in real terms. Food imports by SSA, meanwhile, have increased at a very high rate. As for the structure of agricultural commodity exports, reliance on a few commodities--coffee, cocoa and cotton-- has even increased. Because of slow world import demand growth for these commodities, the world market outlook for these commodities, especially for coffee and cocoa, is not encouraging. 62. Comparisons within the Region and with countries in other regions show that inappropriate exchange rate and producer pricing policies, as well as lack of support for technological advancement, have been major factors in SSA's poor performance in its export sector. Many SSA countries are acutely aware of the poor performance of their agricultural sectors and they have taken policy reforms, sometimes drastic ones, including currency devaluation and producer price increases, to change the unfavorable trends of the past. With these adjustments the agricultural export sectors in these countries face a much more encouraging environment. However, these macro-economic and taxation policies alone may not be sufficient to increase agricultural commodity exports in a significant way, especially in the short-term. Actions in each sub-sector in the areas of production technology, processing, marketing and sub-sector level policies may be required in many countries. In a dynamic world, the agricultural sectors of these countries need to be - 55 - flexible and dynamic enough to cope with changing tastes and import requirements in the established as well as potential markets. It cannot be emphasized enough that a successful trading country or company is the one that is flexible and dynamic enough to produce, market and export products which are in high demand. 63. The examples of tea in Kenya and cocoa in C8te d'Ivoire suggest that proper policies can result in dynamic agriculture sub-sectors in SSA countries. However, in the absence of fundamentally sound and viable policies, the projections of export volumes and revenue from agricultural primary commodities of SSA are dismal. In fact, income terms of trade of agricultural commodity exports in the next five years are expected to be about 40X below the level which prevailed in the first half of the 1980s and 50% below the levels of the 1970s. The growth rate of income terms of trade of agriculture is expected to be a mere 281 per annum for the next decade which would make the income terms of trade in the year 2000 about 17% lower than the level which prevailed in the first half of the 1980s. - 56 - Appexdix Table 1: 91ARE OF IOUL MEHIANDISE IRALE, 1974-1976 AVERA Orondnut Total Gomtry Coffc_ kcoa Tea Fruit Cotton Tobacco Sugar Fore.L Spices Oil Agriculture (Z) Angola 15.8 0.0 0.0 0.0 1.5 0.5 0.0 0.5 0.0 0.0 21.9 Bururdi 81.9 - 2.3 0.0 3.8 0.0 - - - - 98.7 Benin 1.6 8.4 - 0.1 28.5 2.4 0.0 0.0 0.0 4.8 70.1 Botswana 0.0 0.0 0.0 0.0 0.0 0.0 - 0.0 0.1 39.7 Burkdna Faso - - 0.0 0.0 28.2 0.0 0.0 0.0 0.0 13.4 95.6 Central African Rep. 28.6 0.1 0.0 16.2 1.9 0.1 29.1 - 0.0 58.1 Cote dIvoire, Iepublic 26.6 24.8 0.0 0.0 1.6 0.0 0.0 19.6 0.0 0.0 65.7 Caeron 27.0 32.3 0.1 0.0 2.1 0.5 0.1 12.5 0.0 1.1 71.6 Congo, Peoplers Rep. 0.8 1.5 0.0 0.0 0.0 0.1 3.4 16.5 0.0 0.0 6.6 Camoros Islands 0.6 0.4 - - - - - - 0.0 - 63.7 Cape Verde 0.0 0.0 - 0.0 - - 0.0 - - 0.0 11.7 Djibouti - - - - - - - - 0.0 Ethiopia 38.3 - 0.0 0.0 1.4 0.0 1.9 0.0 0.1 0.5 91.8 Gabon 0.0 0.2 - - - - 0.0 9.3 - 0.0 0.2 Ghana 0.4 71.6 - 0.0 - 0.1 0.0 10.8 0.0 - 73.1 Gainea 1.1 2.5 - 0.0 - - - 2.4 - 1.0 11.1 Gambia, te - - - 0.0 0.0 - - - - 84.5 94.6 Qainea-Bissau - - - 0.0 0.0 0.0 1.2 2.9 - 58.4 82.1 Equatorial Qdnea 8.9 67.9 - - - - - 8.2 - - 77.3 Kenya 22.4 0.1 10.6 0.1 0.5 0.0 0.0 0.8 0.0 0.0 59.4 LIberia 1.2 1.0 0.0 - 0.0 0.0 0.0 5.3 - - 16.6 Lesotho - - - 0.0 - - - - - - 73.7 Madagascar 30.6 0.7 0.0 0.0 0.2 0.5 4.7 0.3 0.0 0.1 75.0 Malu 0.0 - 0.1 0.0 27.1 0.1 0.0 - - 9.9 80.1 azamhique 0.0 0.0 3.8 0.0 10.4 0.4 14.3 7.1 0.0 1.1 70.7 Mauritania - - - - - - - - - - 11.6 mauritius 0.0 0.0 1.2 0.0 - 0.0 80.8 - 0.1 0.0 84.4 Malawi 0.3 0.0 17.6 0.1 1.9 45.3 11.5 0.3 0.1 6.0 89.4 Nanmbia - - - - - - - - - - - Niger 0.0 0.0 0.0 0.0 0.5 0.0 0.0 0.0 0.0 6.8 44.1 Nigeria 0.0 3.7 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.1 5.4 Rwanda 66.3 - 7.2 - - - - - - 0.0 88.8 Sudan - - - 0.0 43.7 0.0 - - 0.0 20.5 95.2 Seneal 0.0 0.0 0.0 0.0 2.1 - 0.1 - 0.0 36.5 50.7 Sierra leone 6.9 7.7 - 0.0 - 0.0 0.0 - 0.3 - 26.6 Somalia - - 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 88.5 Sao 1Tom And Principe 0.8 73.7 - - - - - 0.0 - - 84.7 Swaziland - 0.1 - 1.4 1.1 0.1 37.9 17.8 - - 51.8 Seydele- - 0.0 0.0 0.0 - - 0.0 - - - 31.1 Cald - - 0.0 0.0 43.1 0.1 0.0 - - 0.0 85.6 TogP 8.9 15.1 0.0 0.0 1.5 0.0 0.0 0.0 - 0.3 28.5 Tanzania 21.1 0.2 3.0 0.0 14.6 5.2 1.3 0.4 0.0 0.0 79.9 Uganda 77.1 0.0 4.6 0.0 9.5 0.6 0.1 0.1 0.0 0.0 94.7 Zaire 11.4 0.7 0.4 0.0 0.1 0.0 0.0 1.0 0.0 0.0 19.9 Z7ia 0.0 0.0 - 0.0 0.0 0.8 0.0 0.0 0.0 0.1 1.3 ZIirala. 0.8 0.0 0.5 0.1 5.1 13.3 5.8 0.3 0.0 0.6 40.3 Sub-Saharn Africa 6.9 6.3 0.6 0.0 2.3 1.0 2.0 2.9 0.0 1.6 29.6 World 0.6 0.3 0.1 0.2 0.5 0.3 1.1 3.1 0.0 0.1 13.8 - 57 - Apperdix Table 2: MARE OF TOML MWNDIE IR,E 1984-1986 AVERAGE Grounut Total Canttry Coffee Coca Tea Fruit Ootton Tobacco Sugar Forest Spices 011 Agriculture (2) Argola 3.4 0.0 0.0 0.0 0.1 0.2 0.0 0.0 0.0 0.0 3.7 rdi 85.2 - 5.1 0.0 0.4 0.0 - - - - 93.9 Berdn 3.7 18.7 - 0.0 22.8 0.0 0.0 0.0 0.0 0.0 56.6 Botswana 0.0 0.0 0.0 0.0 0.0 0.0 - - 0.0 0.0 8.4 Burkdna Faso - - 0.0 0.0 47.4 0.0 0.0 0.0 0.0 0.4 82.7 Central African Rep. 18.4 0.0 0.0 - 8.4 0.4 0.0 12.8 - 0.0 33.4 Cote dIvoire, Republic 20.3 40.3 0.0 0.0 2.8 0.0 0.3 8.3 0.0 0.0 74.6 Cwerom 18.8 16.3 0.1 0.0 1.6 0.4 0.1 7.0 0.0 0.0 40.9 Como People"s Rep. 0.3 0.3 0.0 0.0 0.0 0.0 0.7 4.6 0.0 0.0 1.4 CGros Islands 0.0 0.0 - - - - - - 0.1 - 87.6 Cape Verde 0.0 0.0 - 0.0 - - 0.0 - - 0.0 16.3 Djibout - - - - - - - - - - 29.7 Ethiopia 67.4 - 0.0 0.2 0.3 0.0 1.0 0.0 0.1 0.0 93.5 Gabmn 0.1 0.3 - - - - 0.0 7.2 - 0.0 0.5 Ghana 0.1 52.1 - 0.0 - 0.0 0.0 3.0 0.0 - 52.5 GuinBa 0.8 1.5 - 0.0 - - - 0.2 - 0.0 4.4 GmmbJa, The - - - 0.0 1.6 - - - - 37.1 41.7 Guinea-Bissau - - - 0.0 4.7 0.0 0.0 2.9 - 16.2 74.4 Equatorial Ginea 5.2 43.7 - - - - - 30.0 - - 49.3 Kemya 31.3 0.0 23.4 0.1 0.1 0.1 0.0 0.2 0.0 0.0 70.7 Liberia 4.5 2.8 0.0 - 0.0 0.0 0.0 8.6 - - 28.0 lasotho - - - 0.0 - - - - - - 67.6 Madagasar 40.7 1.2 0.0 0.0 1.1 0.4 1.2 0.0 0.0 0.0 79.5 Mali 0.0 - 0.0 0.0 36.5 0.0 0.0 - - 0.9 83.7 Moambique 0.0 0.0 5.6 0.0 10.6 0.0 8.3 1.8 0.0 1.8 52.6 MauritanLa - - - - - - - - - - 9.9 Mauritius 0.0 0.0 2.9 0.1 - 0.0 43.9 - 0.1 0.0 48.8 PsIa 2.9 0.0 20.6 0.0 1.4 49.3 8.5 0.1 0.2 1.7 93.3 Nadbia - - - - - - - - - - - Niger 0.0 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 23.6 Nigeria 0.0 3.1 0.0 0.0 0.0 0.0 0.0 0.1 0.0 0.0 3.6 Rwarda 71.3 - 8.1 - - - - - - 0.0 85.8 Sudan - - - 0.1 45.6 0.0 - - 0.0 2.7 95.5 Seiawl 0.0 0.0 0.0 0.2 2.8 - 0.0 - 0.0 11.9 19.5 Sierra leone 12.8 17.0 - 0.0 - 0.7 0.0 - 0.1 - 34.4 SnMII A - - 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 92.7 Sao Txe nld Principe 0.4 51.6 - - - - - 0.0 - - 58.9 Swazilamd - 0.0 - 3.2 2.0 0.0 38.1 36.2 - - 57.6 Saychefles - 0.0 0.0 0.0 - 0.0 0.0 - - - 5.9 Ra - 0 0.0 0.0 15.8 0.0 0.0 - - 0.0 68.5 Togo 9.6 14.6 0.0 0.0 12.6 0.0 0.0 0.0 - 1.3 41.6 Tanzania 41.5 0.6 5.3 0.0 9.7 3.2 0.9 0.5 0.0 0.0 79.3 Ugada 86.9 0.1 0.6 0.0 1.9 0.3 0.0 0.0 0.0 0.0 91.0 Zaire 25.5 1.0 0.5 0.0 0.0 0.0 0.0 2.2 0.0 0.0 32.0 ZaiLA 0.2 0.0 - 0.0 0.6 0.9 0.9 0.0 0.0 0.0 3.5 Zimbabw 2.8 0.0 1.4 0.0 8.1 21.3 3.8 0.7 0.0 0.1 45.6 Sul-Saihar Africa 8.8 7.0 1.2 0.0 2.2 1.3 1.3 2.2 0.0 0.4 28.8 World 0.6 0.3 0.1 0.1 0.3 0.2 0.5 2.6 0.0 0.0 10.9 - 58 - REFERENCES Akiyama, T. (1988), Cocoa and Coffee Pricing Policies in C6te d'Ivoire, World Bank PPR Working Paper, WPS 64. Akiyama, T. and P. Varangis (1989), Impact of the International Coffee Agreement's Export Quota System on the World's Coffee Market, World Bank PPR Working Paper, WPS 148. FAO, Statistics Series No. 82, Production Vol. 41 (1987), Rome. FAO, Statistics Series No. 84, Trade and Commerce, Vol. 41 (1987), Rome. Humphreys, C. and J. Underwood (1989), "The External Debt Difficulties of Low Income Africa" (draft), World Bank. International Monetary Fund, International Financial Statistics, (various issues), Washington, D.C. Singh, S. (1983), Sub-Saharan Agriculture-Synthesis and Trade Prospects, World Bank Staff Working Paper No. 608, July 1983. UNCTAD (1988), Handbook of International Trade and Development Statistics, New York. World Bank (1981), Accelerated Development in Sub-Saharan Africa: An Agenda for Action, Washington, D.C. ___ __ and UNDP (1989), Africa's Adjustment and Growth in the 1980s, Washington, D.C. _________ (1988), Price Prospects for Major Primary Commodities, 1989-2000, Washington, D.C., November. PPR Working Paper Series Contact Illis ~~~~~~~Author DIfor pagper WPS333 Secondary Market Prices under Stijn Claessens Alternative Debt Reduction Sweder van Wijnbergen Strategies: An Option Pricing Approach with an Application to Mexico WPS334 Built-in Tax Elasticity and the Jaber Ehdaie Revenue Impact of Discretionary Tax Measures: An Econometric Estimation Method with Application to Malawi and Mauritius WPS335 Macroeconomic Adjustment and Ramon E. Lopez Labor Market Structure in Four Luis A. Riveros Latin American Countries: An Econometric Study WPS336 Complementary Approaches to Germano Mwabu Financing Health Services in Africa WPS337 Projecting Mortality for All Rodolfo A. Bulatao Countries Eduard Bos WPS338 Supply and Use of Essential Drugs S. D. Foster in Sub-Saharan Africa: Issues and Possible Solutions WPS339 Private Investment and Macro- Luis Serven economic Adjustment: An Andres Solimano Overview WPS340 Prudential Regulation and Banking Vincent P. Polizatto Supervision: Building an Institutional Framework for Banks WPS341 Cost-of-Living Differentials between Martin Ravallion Urban and Rural Areas of Indonesia Dominique van de Walle WPS342 Human Capital and Endogenous Patricio Arrau Growth in a Large Scale Life-Cycle Model WPS343 Policy Determinants of Growth: William R. Easterly Survey of Theory and Evidence Deborah L. Wetzel WPS344 Policy Distortions, Size of William Easterly Government, and Growth PPR Working Paper Series Contact Ima Auuthor WPS345 Essays on Private Transfers and Donald Cox Public Policy in Developing Emmanuel Jimenez Countries: A Survey and Case Study for Peru WPS346 Indiia. The Growing Conflict Hans Jurgen Peters between Trade and Transport: Issues and Options WPS347 Housing Finance in Developing Robert M. Buckley Countries: A Transaction Cost Approach WPS348 Recent Trends and Prospects Takamasa Akiyama December 1989 D. Gustafson for Agricultural Commodity Exports Donald F. Larson 33714 in Sub-Saharan Africa